2
0
mirror of https://github.com/boostorg/python.git synced 2026-01-19 16:32:16 +00:00

Compare commits

...

17 Commits

Author SHA1 Message Date
Dave Abrahams
0eb039a500 NULL shared_ptr conversions, more tests for custom to-python shared_ptr registrations
[SVN r18263]
2003-04-16 14:32:14 +00:00
nobody
ba3b0a7c5d This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.2'.

[SVN r17835]
2003-03-12 12:47:45 +00:00
Dave Abrahams
b990d7580d Changes for MPL v2
[SVN r13504]
2002-04-16 14:29:30 +00:00
nobody
536dbe28a6 This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.2'.

[SVN r13257]
2002-03-24 15:19:57 +00:00
Dave Abrahams
27f99ba783 no message
[SVN r10384]
2001-06-21 22:49:06 +00:00
nobody
4c30bb98f9 This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.2'.

[SVN r9436]
2001-03-04 19:00:43 +00:00
Ralf W. Grosse-Kunstleve
582dc47885 Makefile for Windows mingw32 (a gcc type compiler).
[SVN r9388]
2001-03-03 02:54:03 +00:00
Ralf W. Grosse-Kunstleve
5666c21037 Enhanced pickle safety measures.
[SVN r9358]
2001-02-27 18:56:57 +00:00
Ralf W. Grosse-Kunstleve
16264a3f2d Safer pickle support (explicit auto_pickle parameter)
[SVN r9313]
2001-02-22 20:47:36 +00:00
Ralf W. Grosse-Kunstleve
046fb553c5 map [plain] char '\0' <-> "" (Python string of length 0)
[SVN r9306]
2001-02-21 10:00:27 +00:00
Ralf W. Grosse-Kunstleve
3eb98f54a1 Mapping of std::complex<T> <-> Python complex.
[SVN r9305]
2001-02-21 08:01:54 +00:00
Ralf W. Grosse-Kunstleve
f1d4a38121 Added to/from_python converters for [plain] char.
Reverted to original converters for un/signed char.


[SVN r9046]
2001-02-09 03:31:32 +00:00
Ralf W. Grosse-Kunstleve
6d3b960cb7 -lm on link-line for comprehensive.so eliminates warning.
[SVN r9021]
2001-02-08 01:37:27 +00:00
Ralf W. Grosse-Kunstleve
2aed6f022e rational.hpp (used in comprehensive.cpp) requires -ftemplate-depth-21
[SVN r9020]
2001-02-08 01:35:55 +00:00
nobody
724c91b7b2 This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.2'.

[SVN r9018]
2001-02-07 23:58:45 +00:00
Ralf W. Grosse-Kunstleve
1e272b9c14 BPL support for exporting/importing class wrappers.
[SVN r8788]
2001-01-27 17:35:02 +00:00
nobody
2ceaa9133d This commit was manufactured by cvs2svn to create branch
'unlabeled-1.1.2'.

[SVN r8787]
2001-01-27 17:35:01 +00:00
550 changed files with 2140 additions and 59363 deletions

2
build/.cvsignore Normal file
View File

@@ -0,0 +1,2 @@
Makefile
patch_tru64_to_gcc

Binary file not shown.

View File

@@ -1,84 +1,36 @@
# (C) Copyright David Abrahams 2001. Permission to copy, use, modify, sell and
# distribute this software is granted provided this copyright notice appears
# in all copies. This software is provided "as is" without express or implied
# warranty, and with no claim as to its suitability for any purpose.
#
# Boost.Python library Jamfile
# declare the location of this subproject relative to the root
subproject libs/python/build ;
# bring in the rules for python
SEARCH on <module@>python.jam = $(BOOST_BUILD_PATH) ;
include <module@>python.jam ;
SOURCES = classes.cpp conversions.cpp extension_class.cpp functions.cpp
init_function.cpp module_builder.cpp
objects.cpp types.cpp cross_module.cpp ;
if [ check-python-config ]
PYTHON_ROOT ?= c:/tools/python ;
PYTHON_INCLUDES = $(BOOST_ROOT) $(PYTHON_ROOT)$(SLASH)include ;
lib libboost_python : ../src/$(SOURCES) : <include>$(PYTHON_INCLUDES) <shared-linkable>true ;
rule boost-python
{
local bpl-linkflags ;
dll $(<) : $(>) :
$(3) <include>$(PYTHON_INCLUDES)
<msvc><library-path>$(PYTHON_ROOT)/libs
<gcc><*><library-file>$(PYTHON_ROOT)/libs/python20.lib
: $(4) ;
}
if $(UNIX) && ( $(OS) = AIX )
{
bpl-linkflags = <linkflags>"-e initlibboost_python" ;
}
boost-python boost_python_test : <lib>libboost_python ../test/comprehensive.cpp ;
# Enabling intrinsics (/0i) or maximize speed (/02) seem to cause
# internal compiler errors with this toolset.
local msvc-stlport-workarounds
= <optimization>off "<cxxflags>-Ogty -O1 -Gs" ;
local sources =
numeric.cpp
list.cpp
long.cpp
dict.cpp
tuple.cpp
str.cpp
boost-python abstract : <lib>libboost_python ../example/abstract.cpp ;
boost-python getting_started1 : <lib>libboost_python ../example/getting_started1.cpp ;
boost-python getting_started2 : <lib>libboost_python ../example/getting_started2.cpp ;
boost-python simple_vector : <lib>libboost_python ../example/simple_vector.cpp ;
boost-python do_it_yourself_converters : <lib>libboost_python ../example/do_it_yourself_converters.cpp ;
boost-python pickle1 : <lib>libboost_python ../example/pickle1.cpp ;
boost-python pickle2 : <lib>libboost_python ../example/pickle2.cpp ;
boost-python pickle3 : <lib>libboost_python ../example/pickle3.cpp ;
boost-python noncopyable_export : <lib>libboost_python ../example/noncopyable_export.cpp ;
boost-python noncopyable_import : <lib>libboost_python ../example/noncopyable_import.cpp ;
boost-python ivect : <lib>libboost_python ../example/ivect.cpp ;
boost-python dvect : <lib>libboost_python ../example/dvect.cpp ;
aix_init_module.cpp
converter/from_python.cpp
converter/registry.cpp
converter/type_id.cpp
object/enum.cpp
object/class.cpp
object/function.cpp
object/inheritance.cpp
object/life_support.cpp
object/pickle_support.cpp
errors.cpp
module.cpp
converter/builtin_converters.cpp
converter/arg_to_python_base.cpp
object/iterator.cpp
object_protocol.cpp
object_operators.cpp
;
dll boost_python
: ../src/$(sources)
: $(BOOST_PYTHON_V2_PROPERTIES)
<define>BOOST_PYTHON_SOURCE
$(bpl-linkflags)
<msvc-stlport><release>$(msvc-stlport-workarounds)
;
lib boost_python
: # sources
../src/$(sources)
: # requirements
$(BOOST_PYTHON_V2_PROPERTIES)
<define>BOOST_PYTHON_SOURCE
<define>BOOST_STATIC_LIB
$(bpl-linkflags)
<msvc-stlport><release>$(msvc-stlport-workarounds)
;
stage bin-stage : <dll>boost_python <lib>boost_python
: <tag><debug>"_debug"
<tag><debug-python>"_pydebug"
:
debug release
;
}

View File

@@ -1,85 +0,0 @@
import os ;
# Use a very crude way to sense there python is locatted
local PYTHON_PATH ;
if [ GLOB /usr/local/include/python2.2 : * ]
{
PYTHON_PATH = /usr/local ;
}
else if [ GLOB /usr/include/python2.2 : * ]
{
PYTHON_PATH = /usr ;
}
if [ os.name ] in CYGWIN NT
{
lib_condition = <link>shared: ;
defines = USE_DL_IMPORT ;
# Declare a target for the python interpreter library
lib python : : <name>python2.2.dll ;
PYTHON_LIB = python ;
}
else
{
lib python : : <name>python2.2 ;
PYTHON_LIB = python ;
}
if $(PYTHON_PATH) {
project boost/python
: source-location ../src
: requirements <include>$(PYTHON_PATH)/include/python2.2
$(lib_condition)<library-path>$(PYTHON_PATH)/lib/python2.2/config
<link>shared:<library>$(PYTHON_LIB)
<define>$(defines)
: usage-requirements # requirement that will be propageted to *users* of this library
<include>$(PYTHON_PATH)/include/python2.2
# We have a bug which causes us to conclude that conditionalized
# properties in this section are not free.
# $(lib_condition)<library-path>$(PYTHON_PATH)/lib/python2.2/config
# <shared>true:<find-library>$(PYTHON_LIB)
<library-path>$(PYTHON_PATH)/lib/python2.2/config
<library>$(PYTHON_LIB)
;
lib boost_python
:
numeric.cpp
list.cpp
long.cpp
dict.cpp
tuple.cpp
str.cpp
aix_init_module.cpp
converter/from_python.cpp
converter/registry.cpp
converter/type_id.cpp
object/enum.cpp
object/class.cpp
object/function.cpp
object/inheritance.cpp
object/life_support.cpp
object/pickle_support.cpp
errors.cpp
module.cpp
converter/builtin_converters.cpp
converter/arg_to_python_base.cpp
object/iterator.cpp
object_protocol.cpp
object_operators.cpp
: <link>static:<define>BOOST_PYTHON_STATIC_LIB
<define>BOOST_PYTHON_SOURCE
: <link>shared
;
}

194
build/Makefile.gcc Normal file
View File

@@ -0,0 +1,194 @@
# Usage:
#
# Create a new empty directory anywhere (preferably not in the boost tree).
# Copy this Makefile to that new directory and rename it to "Makefile"
# Set the MYBOOST pathname below.
#
# make softlinks Create softlinks to source code and tests
# make Compile all sources
# make test Run doctest tests
# make clean Remove all object files
# make unlink Remove softlinks
MYBOOST= /net/cci/rwgk/myboost
PYEXE= /usr/local/Python-1.5.2/bin/python
PYINC= -I/usr/local/Python-1.5.2/include/python1.5
#PYEXE= /usr/local/Python-2.0/bin/python
#PYINC= -I/usr/local/Python-2.0/include/python2.0
#STLPORTINC= -I/usr/local/STLport-4.1b3/stlport
#STLPORTOPTS= \
# -D__USE_STD_IOSTREAM \
# -D__STL_NO_SGI_IOSTREAMS \
# -D__STL_USE_NATIVE_STRING \
# -D__STL_NO_NEW_C_HEADERS \
# -D_RWSTD_COMPILE_INSTANTIATE=1
#STLPORTINC= -I/usr/local/STLport-4.1b4/stlport
#STLPORTOPTS= -D__NO_USE_STD_IOSTREAM -D__STL_NO_SGI_IOSTREAMS
#STLPORTINC= -I/net/cci/xp/C++_C_headers
STDOPTS= -ftemplate-depth-21
WARNOPTS=
CPP= g++
CPPOPTS= $(STLPORTINC) $(STLPORTOPTS) -I$(MYBOOST) $(PYINC) \
$(STDOPTS) $(WARNOPTS) -g
LD= g++
LDOPTS= -shared
BPL_SRC = $(MYBOOST)/libs/python/src
BPL_TST = $(MYBOOST)/libs/python/test
BPL_EXA = $(MYBOOST)/libs/python/example
SOFTLINKS = \
$(BPL_SRC)/classes.cpp \
$(BPL_SRC)/conversions.cpp \
$(BPL_SRC)/extension_class.cpp \
$(BPL_SRC)/functions.cpp \
$(BPL_SRC)/import_extension_class.cpp \
$(BPL_SRC)/init_function.cpp \
$(BPL_SRC)/module_builder.cpp \
$(BPL_SRC)/objects.cpp \
$(BPL_SRC)/types.cpp \
$(BPL_TST)/comprehensive.cpp \
$(BPL_TST)/comprehensive.hpp \
$(BPL_TST)/comprehensive.py \
$(BPL_TST)/doctest.py \
$(BPL_EXA)/abstract.cpp \
$(BPL_EXA)/dvect.cpp \
$(BPL_EXA)/dvect.h \
$(BPL_EXA)/getting_started1.cpp \
$(BPL_EXA)/getting_started2.cpp \
$(BPL_EXA)/getting_started3.cpp \
$(BPL_EXA)/getting_started4.cpp \
$(BPL_EXA)/getting_started5.cpp \
$(BPL_EXA)/passing_char.cpp \
$(BPL_EXA)/cmplx.cpp \
$(BPL_EXA)/ivect.cpp \
$(BPL_EXA)/ivect.h \
$(BPL_EXA)/noncopyable_export.cpp \
$(BPL_EXA)/noncopyable_import.cpp \
$(BPL_EXA)/store.h \
$(BPL_EXA)/test_abstract.py \
$(BPL_EXA)/test_getting_started1.py \
$(BPL_EXA)/test_getting_started2.py \
$(BPL_EXA)/test_getting_started3.py \
$(BPL_EXA)/test_getting_started4.py \
$(BPL_EXA)/test_getting_started5.py \
$(BPL_EXA)/tst_passing_char.py \
$(BPL_EXA)/tst_cmplx.py \
$(BPL_EXA)/tst_dvect.py \
$(BPL_EXA)/tst_ivect.py \
$(BPL_EXA)/tst_noncopyable.py
OBJ = classes.o conversions.o extension_class.o functions.o \
import_extension_class.o init_function.o module_builder.o \
objects.o types.o
.SUFFIXES: .o .cpp
all: libbpl.a test.so abstract.so \
getting_started1.so getting_started2.so getting_started3.so \
getting_started4.so getting_started5.so \
passing_char.so cmplx.so \
noncopyable_export.so noncopyable_import.so \
ivect.so dvect.so
softlinks:
@ for pn in $(SOFTLINKS); \
do \
bn=`basename "$$pn"`; \
if [ ! -e "$$bn" ]; then \
echo "ln -s $$pn ."; \
ln -s "$$pn" .; \
else \
echo "info: no softlink created (file exists): $$bn"; \
fi; \
done
unlink:
@ for pn in $(SOFTLINKS); \
do \
bn=`basename "$$pn"`; \
if [ -L "$$bn" ]; then \
echo "rm $$bn"; \
rm "$$bn"; \
elif [ -e "$$bn" ]; then \
echo "info: not a softlink: $$bn"; \
fi; \
done
libbpl.a: $(OBJ)
rm -f libbpl.a
ar r libbpl.a $(OBJ)
test.so: $(OBJ) comprehensive.o
$(LD) $(LDOPTS) $(OBJ) comprehensive.o -o test.so -lm
abstract.so: $(OBJ) abstract.o
$(LD) $(LDOPTS) $(OBJ) abstract.o -o abstract.so
getting_started1.so: $(OBJ) getting_started1.o
$(LD) $(LDOPTS) $(OBJ) getting_started1.o -o getting_started1.so
getting_started2.so: $(OBJ) getting_started2.o
$(LD) $(LDOPTS) $(OBJ) getting_started2.o -o getting_started2.so
getting_started3.so: $(OBJ) getting_started3.o
$(LD) $(LDOPTS) $(OBJ) getting_started3.o -o getting_started3.so
getting_started4.so: $(OBJ) getting_started4.o
$(LD) $(LDOPTS) $(OBJ) getting_started4.o -o getting_started4.so
getting_started5.so: $(OBJ) getting_started5.o
$(LD) $(LDOPTS) $(OBJ) getting_started5.o -o getting_started5.so
passing_char.so: $(OBJ) passing_char.o
$(LD) $(LDOPTS) $(OBJ) passing_char.o -o passing_char.so
cmplx.so: $(OBJ) cmplx.o
$(LD) $(LDOPTS) $(OBJ) cmplx.o -o cmplx.so -lm
noncopyable_export.so: $(OBJ) noncopyable_export.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) noncopyable_export.o -o noncopyable_export.so
noncopyable_import.so: $(OBJ) noncopyable_import.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) noncopyable_import.o -o noncopyable_import.so
ivect.cpp: ivect.h dvect.h
ivect.so: $(OBJ) ivect.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) ivect.o -o ivect.so
dvect.cpp: ivect.h dvect.h
dvect.so: $(OBJ) dvect.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) dvect.o -o dvect.so
.cpp.o:
$(CPP) $(CPPOPTS) -c $*.cpp
test:
$(PYEXE) comprehensive.py
$(PYEXE) test_abstract.py
$(PYEXE) test_getting_started1.py
$(PYEXE) test_getting_started2.py
$(PYEXE) test_getting_started3.py
$(PYEXE) test_getting_started4.py
$(PYEXE) test_getting_started5.py
clean:
rm -f $(OBJ) libbpl.a libbpl.a.input
rm -f comprehensive.o test.so
rm -f abstract.o abstract.so
rm -f getting_started1.o getting_started1.so
rm -f getting_started2.o getting_started2.so
rm -f getting_started3.o getting_started3.so
rm -f getting_started4.o getting_started4.so
rm -f getting_started5.o getting_started5.so
rm -f passing_char.o passing_char.so
rm -f cmplx.o cmplx.so
rm -f noncopyable_export.o noncopyable_export.so
rm -f noncopyable_import.o noncopyable_import.so
rm -f ivect.o ivect.so
rm -f dvect.o dvect.so
rm -f so_locations *.pyc
rm -rf cxx_repository

269
build/Makefile.mingw32 Normal file
View File

@@ -0,0 +1,269 @@
# Usage:
#
# Create a new empty directory anywhere (preferably not in the boost tree).
# Copy this Makefile to that new directory and rename it to "Makefile"
# Set the MYBOOST_* pathnames below.
#
# The idea is that the build directory is on a Unix filesystem that
# is mounted on a PC using SAMBA. Use this makefile under both Unix
# and Windows:
#
# Unix: make softlinks Create softlinks to source code and tests
# Win: make Compile all sources
# Win: make test Run doctest tests
# Unix: make clean Remove all object files
# Unix: make unlink Remove softlinks
# To install mingw32, follow instructions at:
# http://starship.python.net/crew/kernr/mingw32/Notes.html
# In particular, install:
# ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/mingw32/gcc-2.95.2/gcc-2.95.2-msvcrt.exe
# ftp://ftp.xraylith.wisc.edu/pub/khan/gnu-win32/mingw32/gcc-2.95.2/fixes/quote-fix-msvcrt.exe
# http://starship.python.net/crew/kernr/mingw32/Python-1.5.2-mingw32.zip
# Unpack the first two archives in the default locations and update your PATH.
# Unpack the third archive in \usr.
# Note: comprehensive.cpp generates compiler errors and later crashes.
# Could this be fixed with compiler options?
MYBOOST_UNIX= /net/cci/rwgk/myboost
MYBOOST_WIN= "L:\myboost"
PYEXE= "C:\Program files\Python\python.exe"
PYINC= -I"C:\usr\include\python1.5"
PYLIB= "C:\usr\lib\libpython15.a"
STDOPTS= -ftemplate-depth-21
WARNOPTS=
CPP= g++
CPPOPTS= $(STLPORTINC) $(STLPORTOPTS) -I$(MYBOOST_WIN) $(PYINC) \
$(STDOPTS) $(WARNOPTS) -g
LD= g++
LDOPTS= -shared
BPL_SRC = $(MYBOOST_UNIX)/libs/python/src
BPL_TST = $(MYBOOST_UNIX)/libs/python/test
BPL_EXA = $(MYBOOST_UNIX)/libs/python/example
SOFTLINKS = \
$(BPL_SRC)/classes.cpp \
$(BPL_SRC)/conversions.cpp \
$(BPL_SRC)/extension_class.cpp \
$(BPL_SRC)/functions.cpp \
$(BPL_SRC)/import_extension_class.cpp \
$(BPL_SRC)/init_function.cpp \
$(BPL_SRC)/module_builder.cpp \
$(BPL_SRC)/objects.cpp \
$(BPL_SRC)/types.cpp \
$(BPL_TST)/comprehensive.cpp \
$(BPL_TST)/comprehensive.hpp \
$(BPL_TST)/comprehensive.py \
$(BPL_TST)/doctest.py \
$(BPL_EXA)/abstract.cpp \
$(BPL_EXA)/dvect.cpp \
$(BPL_EXA)/dvect.h \
$(BPL_EXA)/getting_started1.cpp \
$(BPL_EXA)/getting_started2.cpp \
$(BPL_EXA)/getting_started3.cpp \
$(BPL_EXA)/getting_started4.cpp \
$(BPL_EXA)/getting_started5.cpp \
$(BPL_EXA)/passing_char.cpp \
$(BPL_EXA)/cmplx.cpp \
$(BPL_EXA)/ivect.cpp \
$(BPL_EXA)/ivect.h \
$(BPL_EXA)/noncopyable_export.cpp \
$(BPL_EXA)/noncopyable_import.cpp \
$(BPL_EXA)/store.h \
$(BPL_EXA)/test_abstract.py \
$(BPL_EXA)/test_getting_started1.py \
$(BPL_EXA)/test_getting_started2.py \
$(BPL_EXA)/test_getting_started3.py \
$(BPL_EXA)/test_getting_started4.py \
$(BPL_EXA)/test_getting_started5.py \
$(BPL_EXA)/tst_passing_char.py \
$(BPL_EXA)/tst_cmplx.py \
$(BPL_EXA)/tst_dvect.py \
$(BPL_EXA)/tst_ivect.py \
$(BPL_EXA)/tst_noncopyable.py
DEFS= \
test \
abstract \
getting_started1 \
getting_started2 \
getting_started3 \
getting_started4 \
getting_started5 \
passing_char \
cmplx \
noncopyable_export \
noncopyable_import \
ivect \
dvect
OBJ = classes.o conversions.o extension_class.o functions.o \
import_extension_class.o init_function.o module_builder.o \
objects.o types.o
.SUFFIXES: .o .cpp
all: libbpl.a test.pyd abstract.pyd \
getting_started1.pyd getting_started2.pyd getting_started3.pyd \
getting_started4.pyd getting_started5.pyd \
passing_char.pyd cmplx.pyd \
noncopyable_export.pyd noncopyable_import.pyd \
ivect.pyd dvect.pyd
softlinks: defs
@ for pn in $(SOFTLINKS); \
do \
bn=`basename "$$pn"`; \
if [ ! -e "$$bn" ]; then \
echo "ln -s $$pn ."; \
ln -s "$$pn" .; \
else \
echo "info: no softlink created (file exists): $$bn"; \
fi; \
done
unlink: rmdefs
@ for pn in $(SOFTLINKS); \
do \
bn=`basename "$$pn"`; \
if [ -L "$$bn" ]; then \
echo "rm $$bn"; \
rm "$$bn"; \
elif [ -e "$$bn" ]; then \
echo "info: not a softlink: $$bn"; \
fi; \
done
defs:
@ for def in $(DEFS); \
do \
echo "EXPORTS\n\tinit$$def" > $$def.def; \
done
rmdefs:
@ for def in $(DEFS); \
do \
rm $$def.def; \
done
libbpl.a: $(OBJ)
del libbpl.a
ar r libbpl.a $(OBJ)
DLLWRAPOPTS= --driver-name g++ -s \
--entry _DllMainCRTStartup@12 --target=i386-mingw32
test.pyd: $(OBJ) comprehensive.o
dllwrap $(DLLWRAPOPTS) \
--dllname test.pyd \
--def test.def \
$(OBJ) comprehensive.o $(PYLIB)
abstract.pyd: $(OBJ) abstract.o
dllwrap $(DLLWRAPOPTS) \
--dllname abstract.pyd \
--def abstract.def \
$(OBJ) abstract.o $(PYLIB)
getting_started1.pyd: $(OBJ) getting_started1.o
dllwrap $(DLLWRAPOPTS) \
--dllname getting_started1.pyd \
--def getting_started1.def \
$(OBJ) getting_started1.o $(PYLIB)
getting_started2.pyd: $(OBJ) getting_started2.o
dllwrap $(DLLWRAPOPTS) \
--dllname getting_started2.pyd \
--def getting_started2.def \
$(OBJ) getting_started2.o $(PYLIB)
getting_started3.pyd: $(OBJ) getting_started3.o
dllwrap $(DLLWRAPOPTS) \
--dllname getting_started3.pyd \
--def getting_started3.def \
$(OBJ) getting_started3.o $(PYLIB)
getting_started4.pyd: $(OBJ) getting_started4.o
dllwrap $(DLLWRAPOPTS) \
--dllname getting_started4.pyd \
--def getting_started4.def \
$(OBJ) getting_started4.o $(PYLIB)
getting_started5.pyd: $(OBJ) getting_started5.o
dllwrap $(DLLWRAPOPTS) \
--dllname getting_started5.pyd \
--def getting_started5.def \
$(OBJ) getting_started5.o $(PYLIB)
passing_char.pyd: $(OBJ) passing_char.o
dllwrap $(DLLWRAPOPTS) \
--dllname passing_char.pyd \
--def passing_char.def \
$(OBJ) passing_char.o $(PYLIB)
cmplx.pyd: $(OBJ) cmplx.o
dllwrap $(DLLWRAPOPTS) \
--dllname cmplx.pyd \
--def cmplx.def \
$(OBJ) cmplx.o $(PYLIB)
noncopyable_export.pyd: $(OBJ) noncopyable_export.o
dllwrap $(DLLWRAPOPTS) \
--dllname noncopyable_export.pyd \
--def noncopyable_export.def \
$(OBJ) noncopyable_export.o $(PYLIB)
noncopyable_import.pyd: $(OBJ) noncopyable_import.o
dllwrap $(DLLWRAPOPTS) \
--dllname noncopyable_import.pyd \
--def noncopyable_import.def \
$(OBJ) noncopyable_import.o $(PYLIB)
ivect.cpp: ivect.h dvect.h
ivect.pyd: $(OBJ) ivect.o
dllwrap $(DLLWRAPOPTS) \
--dllname ivect.pyd \
--def ivect.def \
$(OBJ) ivect.o $(PYLIB)
dvect.cpp: ivect.h dvect.h
dvect.pyd: $(OBJ) dvect.o
dllwrap $(DLLWRAPOPTS) \
--dllname dvect.pyd \
--def dvect.def \
$(OBJ) dvect.o $(PYLIB)
.cpp.o:
$(CPP) $(CPPOPTS) -c $*.cpp
test:
$(PYEXE) comprehensive.py
$(PYEXE) test_abstract.py
$(PYEXE) test_getting_started1.py
$(PYEXE) test_getting_started2.py
$(PYEXE) test_getting_started3.py
$(PYEXE) test_getting_started4.py
$(PYEXE) test_getting_started5.py
clean:
rm -f $(OBJ) libbpl.a libbpl.a.input
rm -f comprehensive.o test.pyd
rm -f abstract.o abstract.pyd
rm -f getting_started1.o getting_started1.pyd
rm -f getting_started2.o getting_started2.pyd
rm -f getting_started3.o getting_started3.pyd
rm -f getting_started4.o getting_started4.pyd
rm -f getting_started5.o getting_started5.pyd
rm -f passing_char.o passing_char.pyd
rm -f cmplx.o cmplx.pyd
rm -f noncopyable_export.o noncopyable_export.pyd
rm -f noncopyable_import.o noncopyable_import.pyd
rm -f ivect.o ivect.pyd
rm -f dvect.o dvect.pyd
rm -f so_locations *.pyc
rm -rf cxx_repository

199
build/Makefile.tru64 Normal file
View File

@@ -0,0 +1,199 @@
# Usage:
#
# Create a new empty directory anywhere (preferably not in the boost tree).
# Copy this Makefile to that new directory and rename it to "Makefile"
# Set the MYBOOST pathname below.
#
# make softlinks Create softlinks to source code and tests
# make Compile all sources
# make test Run doctest tests
# make clean Remove all object files
# make unlink Remove softlinks
MYBOOST= /net/cci/rwgk/myboost
PYEXE= /usr/local/Python-1.5.2/bin/python
PYINC= -I/usr/local/Python-1.5.2/include/python1.5
#PYEXE= /usr/local/Python-2.0/bin/python
#PYINC= -I/usr/local/Python-2.0/include/python2.0
#STLPORTINC= -I/usr/local/STLport-4.1b3/stlport
#STLPORTOPTS= \
# -D__USE_STD_IOSTREAM \
# -D__STL_NO_SGI_IOSTREAMS \
# -D__STL_USE_NATIVE_STRING \
# -D__STL_NO_NEW_C_HEADERS \
# -D_RWSTD_COMPILE_INSTANTIATE=1
#STLPORTINC= -I/usr/local/STLport-4.1b4/stlport
#STLPORTOPTS= -D__NO_USE_STD_IOSTREAM -D__STL_NO_SGI_IOSTREAMS
STLPORTINC= -I/net/cci/xp/C++_C_headers
STDOPTS= -std strict_ansi
WARNOPTS= -msg_disable 186,450,1115
# use -msg_display_number to obtain integer tags for -msg_disable
CPP= cxx
CPPOPTS= $(STLPORTINC) $(STLPORTOPTS) -I$(MYBOOST) $(PYINC) \
$(STDOPTS) $(WARNOPTS) -g
LD= cxx
LDOPTS= -shared -expect_unresolved 'Py*' -expect_unresolved '_Py*'
#HIDDEN= -hidden
BPL_SRC = $(MYBOOST)/libs/python/src
BPL_TST = $(MYBOOST)/libs/python/test
BPL_EXA = $(MYBOOST)/libs/python/example
SOFTLINKS = \
$(BPL_SRC)/classes.cpp \
$(BPL_SRC)/conversions.cpp \
$(BPL_SRC)/extension_class.cpp \
$(BPL_SRC)/functions.cpp \
$(BPL_SRC)/import_extension_class.cpp \
$(BPL_SRC)/init_function.cpp \
$(BPL_SRC)/module_builder.cpp \
$(BPL_SRC)/objects.cpp \
$(BPL_SRC)/types.cpp \
$(BPL_TST)/comprehensive.cpp \
$(BPL_TST)/comprehensive.hpp \
$(BPL_TST)/comprehensive.py \
$(BPL_TST)/doctest.py \
$(BPL_EXA)/abstract.cpp \
$(BPL_EXA)/dvect.cpp \
$(BPL_EXA)/dvect.h \
$(BPL_EXA)/getting_started1.cpp \
$(BPL_EXA)/getting_started2.cpp \
$(BPL_EXA)/getting_started3.cpp \
$(BPL_EXA)/getting_started4.cpp \
$(BPL_EXA)/getting_started5.cpp \
$(BPL_EXA)/passing_char.cpp \
$(BPL_EXA)/cmplx.cpp \
$(BPL_EXA)/ivect.cpp \
$(BPL_EXA)/ivect.h \
$(BPL_EXA)/noncopyable_export.cpp \
$(BPL_EXA)/noncopyable_import.cpp \
$(BPL_EXA)/store.h \
$(BPL_EXA)/test_abstract.py \
$(BPL_EXA)/test_getting_started1.py \
$(BPL_EXA)/test_getting_started2.py \
$(BPL_EXA)/test_getting_started3.py \
$(BPL_EXA)/test_getting_started4.py \
$(BPL_EXA)/test_getting_started5.py \
$(BPL_EXA)/tst_passing_char.py \
$(BPL_EXA)/tst_cmplx.py \
$(BPL_EXA)/tst_dvect.py \
$(BPL_EXA)/tst_ivect.py \
$(BPL_EXA)/tst_noncopyable.py
OBJ = classes.o conversions.o extension_class.o functions.o \
import_extension_class.o init_function.o module_builder.o \
objects.o types.o
.SUFFIXES: .o .cpp
all: libbpl.a test.so abstract.so \
getting_started1.so getting_started2.so getting_started3.so \
getting_started4.so getting_started5.so \
passing_char.so cmplx.so \
noncopyable_export.so noncopyable_import.so \
ivect.so dvect.so
softlinks:
@ for pn in $(SOFTLINKS); \
do \
bn=`basename "$$pn"`; \
if [ ! -e "$$bn" ]; then \
echo "ln -s $$pn ."; \
ln -s "$$pn" .; \
else \
echo "info: no softlink created (file exists): $$bn"; \
fi; \
done
unlink:
@ for pn in $(SOFTLINKS); \
do \
bn=`basename "$$pn"`; \
if [ -L "$$bn" ]; then \
echo "rm $$bn"; \
rm "$$bn"; \
elif [ -e "$$bn" ]; then \
echo "info: not a softlink: $$bn"; \
fi; \
done
libbpl.a: $(OBJ)
rm -f libbpl.a
ls -1 ./cxx_repository/*.o > libbpl.a.input
ar r libbpl.a $(OBJ) -input libbpl.a.input
rm -f libbpl.a.input
test.so: $(OBJ) comprehensive.o
$(LD) $(LDOPTS) $(OBJ) comprehensive.o -o test.so -lm
abstract.so: $(OBJ) abstract.o
$(LD) $(LDOPTS) $(OBJ) abstract.o -o abstract.so
getting_started1.so: $(OBJ) getting_started1.o
$(LD) $(LDOPTS) $(OBJ) getting_started1.o -o getting_started1.so
getting_started2.so: $(OBJ) getting_started2.o
$(LD) $(LDOPTS) $(OBJ) getting_started2.o -o getting_started2.so
getting_started3.so: $(OBJ) getting_started3.o
$(LD) $(LDOPTS) $(OBJ) getting_started3.o -o getting_started3.so
getting_started4.so: $(OBJ) getting_started4.o
$(LD) $(LDOPTS) $(OBJ) getting_started4.o -o getting_started4.so
getting_started5.so: $(OBJ) getting_started5.o
$(LD) $(LDOPTS) $(OBJ) getting_started5.o -o getting_started5.so
passing_char.so: $(OBJ) passing_char.o
$(LD) $(LDOPTS) $(OBJ) passing_char.o -o passing_char.so
cmplx.so: $(OBJ) cmplx.o
$(LD) $(LDOPTS) $(OBJ) cmplx.o -o cmplx.so -lm
noncopyable_export.so: $(OBJ) noncopyable_export.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) noncopyable_export.o -o noncopyable_export.so
noncopyable_import.so: $(OBJ) noncopyable_import.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) noncopyable_import.o -o noncopyable_import.so
ivect.cpp: ivect.h dvect.h
ivect.so: $(OBJ) ivect.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) ivect.o -o ivect.so
dvect.cpp: ivect.h dvect.h
dvect.so: $(OBJ) dvect.o
$(LD) $(LDOPTS) $(OBJ) $(HIDDEN) dvect.o -o dvect.so
.cpp.o:
$(CPP) $(CPPOPTS) -c $*.cpp
test:
$(PYEXE) comprehensive.py
$(PYEXE) test_abstract.py
$(PYEXE) test_getting_started1.py
$(PYEXE) test_getting_started2.py
$(PYEXE) test_getting_started3.py
$(PYEXE) test_getting_started4.py
$(PYEXE) test_getting_started5.py
clean:
rm -f $(OBJ) libbpl.a libbpl.a.input
rm -f comprehensive.o test.so
rm -f abstract.o abstract.so
rm -f getting_started1.o getting_started1.so
rm -f getting_started2.o getting_started2.so
rm -f getting_started3.o getting_started3.so
rm -f getting_started4.o getting_started4.so
rm -f getting_started5.o getting_started5.so
rm -f passing_char.o passing_char.so
rm -f cmplx.o cmplx.so
rm -f noncopyable_export.o noncopyable_export.so
rm -f noncopyable_import.o noncopyable_import.so
rm -f ivect.o ivect.so
rm -f dvect.o dvect.so
rm -f so_locations *.pyc
rm -rf cxx_repository

View File

@@ -1,882 +0,0 @@
# Microsoft Developer Studio Project File - Name="boost_python" - Package Owner=<4>
# Microsoft Developer Studio Generated Build File, Format Version 6.00
# ** DO NOT EDIT **
# TARGTYPE "Win32 (x86) Dynamic-Link Library" 0x0102
CFG=BOOST_PYTHON - WIN32 RELEASE
!MESSAGE This is not a valid makefile. To build this project using NMAKE,
!MESSAGE use the Export Makefile command and run
!MESSAGE
!MESSAGE NMAKE /f "boost_python.mak".
!MESSAGE
!MESSAGE You can specify a configuration when running NMAKE
!MESSAGE by defining the macro CFG on the command line. For example:
!MESSAGE
!MESSAGE NMAKE /f "boost_python.mak" CFG="BOOST_PYTHON - WIN32 RELEASE"
!MESSAGE
!MESSAGE Possible choices for configuration are:
!MESSAGE
!MESSAGE "boost_python - Win32 Release" (based on "Win32 (x86) Dynamic-Link Library")
!MESSAGE "boost_python - Win32 Debug" (based on "Win32 (x86) Dynamic-Link Library")
!MESSAGE
# Begin Project
# PROP AllowPerConfigDependencies 0
# PROP Scc_ProjName ""
# PROP Scc_LocalPath ""
CPP=cl.exe
MTL=midl.exe
RSC=rc.exe
!IF "$(CFG)" == "boost_python - Win32 Release"
# PROP BASE Use_MFC 0
# PROP BASE Use_Debug_Libraries 0
# PROP BASE Output_Dir "Release"
# PROP BASE Intermediate_Dir "Release"
# PROP BASE Target_Dir ""
# PROP Use_MFC 0
# PROP Use_Debug_Libraries 0
# PROP Output_Dir "../bin-stage"
# PROP Intermediate_Dir "release-obj"
# PROP Ignore_Export_Lib 0
# PROP Target_Dir ""
F90=df.exe
# ADD BASE CPP /nologo /MT /W3 /GX /O2 /D "WIN32" /D "NDEBUG" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "BPL_EXPORTS" /YX /FD /Zm800 /Zm800 /Zm800 /c
# ADD CPP /nologo /MD /W3 /GR /GX /O2 /I "../../../../" /D "NDEBUG" /D "WIN32" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "BOOST_PYTHON_DYNAMIC_LIB" /D "BOOST_PYTHON_SOURCE" /FD /Zm800 /Zm800 /Zm800 /Zm800 /c
# SUBTRACT CPP /YX
# ADD BASE MTL /nologo /D "NDEBUG" /mktyplib203 /win32
# ADD MTL /nologo /D "NDEBUG" /mktyplib203 /win32
# ADD BASE RSC /l 0x1409 /d "NDEBUG"
# ADD RSC /l 0x1409 /d "NDEBUG"
BSC32=bscmake.exe
# ADD BASE BSC32 /nologo
# ADD BSC32 /nologo
LINK32=link.exe
# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /dll /machine:I386
# ADD LINK32 kernel32.lib user32.lib /nologo /dll /machine:I386
!ELSEIF "$(CFG)" == "boost_python - Win32 Debug"
# PROP BASE Use_MFC 0
# PROP BASE Use_Debug_Libraries 1
# PROP BASE Output_Dir "Debug"
# PROP BASE Intermediate_Dir "Debug"
# PROP BASE Target_Dir ""
# PROP Use_MFC 0
# PROP Use_Debug_Libraries 1
# PROP Output_Dir "../bin-stage"
# PROP Intermediate_Dir "debug-obj"
# PROP Ignore_Export_Lib 0
# PROP Target_Dir ""
F90=df.exe
# ADD BASE CPP /nologo /MTd /W3 /Gm /GX /ZI /Od /D "WIN32" /D "_DEBUG" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "BPL_EXPORTS" /YX /FD /Zm800 /Zm800 /Zm800 /GZ /c
# ADD CPP /nologo /MDd /W3 /GR /GX /Zi /Od /I "../../../../" /D "_DEBUG" /D "WIN32" /D "_WINDOWS" /D "_MBCS" /D "_USRDLL" /D "BOOST_PYTHON_DYNAMIC_LIB" /D "BOOST_PYTHON_SOURCE" /FD /Zm800 /Zm800 /Zm800 /Zm800 /Zm800 /GZ /c
# ADD BASE MTL /nologo /D "_DEBUG" /mktyplib203 /win32
# ADD MTL /nologo /D "_DEBUG" /mktyplib203 /win32
# ADD BASE RSC /l 0x1409 /d "_DEBUG"
# ADD RSC /l 0x1409 /d "_DEBUG"
BSC32=bscmake.exe
# ADD BASE BSC32 /nologo
# ADD BSC32 /nologo
LINK32=link.exe
# ADD BASE LINK32 kernel32.lib user32.lib gdi32.lib winspool.lib comdlg32.lib advapi32.lib shell32.lib ole32.lib oleaut32.lib uuid.lib odbc32.lib odbccp32.lib /nologo /dll /debug /machine:I386 /pdbtype:sept
# ADD LINK32 kernel32.lib user32.lib /nologo /dll /incremental:no /debug /machine:I386 /out:"../bin-stage/boost_python_debug.dll" /pdbtype:sept
!ENDIF
# Begin Target
# Name "boost_python - Win32 Release"
# Name "boost_python - Win32 Debug"
# Begin Group "Source Files"
# PROP Default_Filter "cpp;c;cxx;rc;def;r;odl;idl;hpj;bat"
# Begin Source File
SOURCE=..\..\src\aix_init_module.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\converter\arg_to_python_base.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\converter\builtin_converters.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\class.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\dict.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\enum.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\errors.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\converter\from_python.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\function.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\inheritance.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\iterator.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\life_support.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\list.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\long.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\module.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\numeric.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object_operators.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object_protocol.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\object\pickle_support.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\converter\registry.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\str.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\tuple.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\converter\type_id.cpp
# End Source File
# End Group
# Begin Group "Header Files"
# PROP Default_Filter "h;hpp;hxx;hm;inl"
# Begin Group "detail"
# PROP Default_Filter ""
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\aix_init_module.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\api_placeholder.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\arg_tuple_size.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\borrowed_ptr.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\call_object.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\caller.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\char_array.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\config.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\construct.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\convertible.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\cv_category.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\decorated_type_id.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\def_helper.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\defaults_def.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\defaults_gen.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\dependent.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\destroy.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\exception_handler.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\force_instantiate.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\if_else.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\indirect_traits.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\make_keyword_range_fn.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\make_tuple.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\map_entry.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\member_function_cast.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\module_base.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\module_init.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\msvc_typeinfo.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\none.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\not_specified.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\operator_id.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\overloads_fwd.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\pointee.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\preprocessor.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\python22_fixed.h
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\raw_pyobject.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\referent_storage.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\result.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\returning.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\scope.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\string_literal.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\target.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\translate_exception.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\type_list.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\type_list_impl.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\type_list_impl_no_pts.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\type_list_utils.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\unwind_type.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\void_ptr.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\void_return.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\detail\wrap_python.hpp
# End Source File
# End Group
# Begin Group "converter"
# PROP Default_Filter ""
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\arg_from_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\arg_to_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\arg_to_python_base.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\builtin_converters.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\constructor_function.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\convertible_function.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\from_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\implicit.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\obj_mgr_arg_from_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\object_manager.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\pointer_type_id.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\pyobject_traits.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\pyobject_type.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\pytype_arg_from_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\pytype_object_mgr_traits.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\registered.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\registered_pointee.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\registrations.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\registry.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\return_from_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\rvalue_from_python_data.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\converter\to_python_function_type.hpp
# End Source File
# End Group
# Begin Group "object"
# PROP Default_Filter ""
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\add_to_namespace.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\class.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\class_converters.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\class_detail.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\class_wrapper.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\construct.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\enum_base.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\find_instance.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\forward.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\function.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\function_handle.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\function_object.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\inheritance.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\instance.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\iterator.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\iterator_core.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\life_support.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\make_holder.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\make_instance.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\pickle_support.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\pointer_holder.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\py_function.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\select_holder.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\value_holder.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object\value_holder_fwd.hpp
# End Source File
# End Group
# Begin Source File
SOURCE=..\..\..\..\boost\python\arg_from_python.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\args.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\args_fwd.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\back_reference.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\base_type_traits.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\bases.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\borrowed.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\call.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\call_method.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\cast.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\class.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\class_fwd.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\copy_const_reference.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\copy_non_const_reference.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\data_members.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\def.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\default_call_policies.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\dict.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\enum.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\errors.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\exception_translator.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\extract.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\handle.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\handle_fwd.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\has_back_reference.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\implicit.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\init.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\instance_holder.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\iterator.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\list.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\long.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\lvalue_from_pytype.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\make_function.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\manage_new_object.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\module.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\module_init.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\numeric.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_attributes.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_call.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_core.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_fwd.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_items.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_operators.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_protocol.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_protocol_core.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\object_slices.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\operators.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\operators2.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\other.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\overloads.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\pointee.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\proxy.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\ptr.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\refcount.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\reference_existing_object.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\return_internal_reference.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\return_value_policy.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\scope.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\self.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\signature.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\slice_nil.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\str.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\tag.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\to_python_converter.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\to_python_indirect.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\to_python_value.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\tuple.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\type_id.hpp
# End Source File
# Begin Source File
SOURCE=..\..\..\..\boost\python\with_custodian_and_ward.hpp
# End Source File
# End Group
# Begin Group "Resource Files"
# PROP Default_Filter "ico;cur;bmp;dlg;rc2;rct;bin;rgs;gif;jpg;jpeg;jpe"
# End Group
# End Target
# End Project

View File

@@ -1,29 +0,0 @@
Microsoft Developer Studio Workspace File, Format Version 6.00
# WARNING: DO NOT EDIT OR DELETE THIS WORKSPACE FILE!
###############################################################################
Project: "boost_python"=".\boost_python.dsp" - Package Owner=<4>
Package=<5>
{{{
}}}
Package=<4>
{{{
}}}
###############################################################################
Global:
Package=<5>
{{{
}}}
Package=<3>
{{{
}}}
###############################################################################

Binary file not shown.

View File

@@ -1,853 +0,0 @@
+++++++++++++++++++++++++++++++++++++++++++
Building Hybrid Systems with Boost.Python
+++++++++++++++++++++++++++++++++++++++++++
:Author: David Abrahams
:Contact: dave@boost-consulting.com
:organization: `Boost Consulting`_
:date: $Date$
:Author: Ralf W. Grosse-Kunstleve
:status: Draft
:copyright: Copyright David Abrahams and Ralf W. Grosse-Kunstleve 2003. All rights reserved
.. contents:: Table of Contents
.. _`Boost Consulting`: http://www.boost-consulting.com
==========
Abstract
==========
Boost.Python is an open source C++ library which provides a concise
IDL-like interface for binding C++ classes and functions to
Python. Leveraging the full power of C++ compile-time introspection
and of recently developed metaprogramming techniques, this is achieved
entirely in pure C++, without introducing a new syntax.
Boost.Python's rich set of features and high-level interface make it
possible to engineer packages from the ground up as hybrid systems,
giving programmers easy and coherent access to both the efficient
compile-time polymorphism of C++ and the extremely convenient run-time
polymorphism of Python.
==============
Introduction
==============
Python and C++ are in many ways as different as two languages could
be: while C++ is usually compiled to machine-code, Python is
interpreted. Python's dynamic type system is often cited as the
foundation of its flexibility, while in C++ static typing is the
cornerstone of its efficiency. C++ has an intricate and difficult
compile-time meta-language, while in Python, practically everything
happens at runtime.
Yet for many programmers, these very differences mean that Python and
C++ complement one another perfectly. Performance bottlenecks in
Python programs can be rewritten in C++ for maximal speed, and
authors of powerful C++ libraries choose Python as a middleware
language for its flexible system integration capabilities.
Furthermore, the surface differences mask some strong similarities:
* 'C'-family control structures (if, while, for...)
* Support for object-orientation, functional programming, and generic
programming (these are both *multi-paradigm* programming languages.)
* Comprehensive operator overloading facilities, recognizing the
importance of syntactic variability for readability and
expressivity.
* High-level concepts such as collections and iterators.
* High-level encapsulation facilities (C++: namespaces, Python: modules)
to support the design of re-usable libraries.
* Exception-handling for effective management of error conditions.
* C++ idioms in common use, such as handle/body classes and
reference-counted smart pointers mirror Python reference semantics.
Given Python's rich 'C' interoperability API, it should in principle
be possible to expose C++ type and function interfaces to Python with
an analogous interface to their C++ counterparts. However, the
facilities provided by Python alone for integration with C++ are
relatively meager. Compared to C++ and Python, 'C' has only very
rudimentary abstraction facilities, and support for exception-handling
is completely missing. 'C' extension module writers are required to
manually manage Python reference counts, which is both annoyingly
tedious and extremely error-prone. Traditional extension modules also
tend to contain a great deal of boilerplate code repetition which
makes them difficult to maintain, especially when wrapping an evolving
API.
These limitations have lead to the development of a variety of wrapping
systems. SWIG_ is probably the most popular package for the
integration of C/C++ and Python. A more recent development is SIP_,
which was specifically designed for interfacing Python with the Qt_
graphical user interface library. Both SWIG and SIP introduce their
own specialized languages for customizing inter-language bindings.
This has certain advantages, but having to deal with three different
languages (Python, C/C++ and the interface language) also introduces
practical and mental difficulties. The CXX_ package demonstrates an
interesting alternative. It shows that at least some parts of
Python's 'C' API can be wrapped and presented through a much more
user-friendly C++ interface. However, unlike SWIG and SIP, CXX does
not include support for wrapping C++ classes as new Python types.
The features and goals of Boost.Python_ overlap significantly with
many of these other systems. That said, Boost.Python attempts to
maximize convenience and flexibility without introducing a separate
wrapping language. Instead, it presents the user with a high-level
C++ interface for wrapping C++ classes and functions, managing much of
the complexity behind-the-scenes with static metaprogramming.
Boost.Python also goes beyond the scope of earlier systems by
providing:
* Support for C++ virtual functions that can be overridden in Python.
* Comprehensive lifetime management facilities for low-level C++
pointers and references.
* Support for organizing extensions as Python packages,
with a central registry for inter-language type conversions.
* A safe and convenient mechanism for tying into Python's powerful
serialization engine (pickle).
* Coherence with the rules for handling C++ lvalues and rvalues that
can only come from a deep understanding of both the Python and C++
type systems.
The key insight that sparked the development of Boost.Python is that
much of the boilerplate code in traditional extension modules could be
eliminated using C++ compile-time introspection. Each argument of a
wrapped C++ function must be extracted from a Python object using a
procedure that depends on the argument type. Similarly the function's
return type determines how the return value will be converted from C++
to Python. Of course argument and return types are part of each
function's type, and this is exactly the source from which
Boost.Python deduces most of the information required.
This approach leads to *user guided wrapping*: as much information is
extracted directly from the source code to be wrapped as is possible
within the framework of pure C++, and some additional information is
supplied explicitly by the user. Mostly the guidance is mechanical
and little real intervention is required. Because the interface
specification is written in the same full-featured language as the
code being exposed, the user has unprecedented power available when
she does need to take control.
.. _Python: http://www.python.org/
.. _SWIG: http://www.swig.org/
.. _SIP: http://www.riverbankcomputing.co.uk/sip/index.php
.. _Qt: http://www.trolltech.com/
.. _CXX: http://cxx.sourceforge.net/
.. _Boost.Python: http://www.boost.org/libs/python/doc
===========================
Boost.Python Design Goals
===========================
The primary goal of Boost.Python is to allow users to expose C++
classes and functions to Python using nothing more than a C++
compiler. In broad strokes, the user experience should be one of
directly manipulating C++ objects from Python.
However, it's also important not to translate all interfaces *too*
literally: the idioms of each language must be respected. For
example, though C++ and Python both have an iterator concept, they are
expressed very differently. Boost.Python has to be able to bridge the
interface gap.
It must be possible to insulate Python users from crashes resulting
from trivial misuses of C++ interfaces, such as accessing
already-deleted objects. By the same token the library should
insulate C++ users from low-level Python 'C' API, replacing
error-prone 'C' interfaces like manual reference-count management and
raw ``PyObject`` pointers with more-robust alternatives.
Support for component-based development is crucial, so that C++ types
exposed in one extension module can be passed to functions exposed in
another without loss of crucial information like C++ inheritance
relationships.
Finally, all wrapping must be *non-intrusive*, without modifying or
even seeing the original C++ source code. Existing C++ libraries have
to be wrappable by third parties who only have access to header files
and binaries.
==========================
Hello Boost.Python World
==========================
And now for a preview of Boost.Python, and how it improves on the raw
facilities offered by Python. Here's a function we might want to
expose::
char const* greet(unsigned x)
{
static char const* const msgs[] = { "hello", "Boost.Python", "world!" };
if (x > 2)
throw std::range_error("greet: index out of range");
return msgs[x];
}
To wrap this function in standard C++ using the Python 'C' API, we'd
need something like this::
extern "C" // all Python interactions use 'C' linkage and calling convention
{
// Wrapper to handle argument/result conversion and checking
PyObject* greet_wrap(PyObject* args, PyObject * keywords)
{
int x;
if (PyArg_ParseTuple(args, "i", &x)) // extract/check arguments
{
char const* result = greet(x); // invoke wrapped function
return PyString_FromString(result); // convert result to Python
}
return 0; // error occurred
}
// Table of wrapped functions to be exposed by the module
static PyMethodDef methods[] = {
{ "greet", greet_wrap, METH_VARARGS, "return one of 3 parts of a greeting" }
, { NULL, NULL, 0, NULL } // sentinel
};
// module initialization function
DL_EXPORT init_hello()
{
(void) Py_InitModule("hello", methods); // add the methods to the module
}
}
Now here's the wrapping code we'd use to expose it with Boost.Python::
#include <boost/python.hpp>
using namespace boost::python;
BOOST_PYTHON_MODULE(hello)
{
def("greet", greet, "return one of 3 parts of a greeting");
}
and here it is in action::
>>> import hello
>>> for x in range(3):
... print hello.greet(x)
...
hello
Boost.Python
world!
Aside from the fact that the 'C' API version is much more verbose,
it's worth noting a few things that it doesn't handle correctly:
* The original function accepts an unsigned integer, and the Python
'C' API only gives us a way of extracting signed integers. The
Boost.Python version will raise a Python exception if we try to pass
a negative number to ``hello.greet``, but the other one will proceed
to do whatever the C++ implementation does when converting an
negative integer to unsigned (usually wrapping to some very large
number), and pass the incorrect translation on to the wrapped
function.
* That brings us to the second problem: if the C++ ``greet()``
function is called with a number greater than 2, it will throw an
exception. Typically, if a C++ exception propagates across the
boundary with code generated by a 'C' compiler, it will cause a
crash. As you can see in the first version, there's no C++
scaffolding there to prevent this from happening. Functions wrapped
by Boost.Python automatically include an exception-handling layer
which protects Python users by translating unhandled C++ exceptions
into a corresponding Python exception.
* A slightly more-subtle limitation is that the argument conversion
used in the Python 'C' API case can only get that integer ``x`` in
*one way*. PyArg_ParseTuple can't convert Python ``long`` objects
(arbitrary-precision integers) which happen to fit in an ``unsigned
int`` but not in a ``signed long``, nor will it ever handle a
wrapped C++ class with a user-defined implicit ``operator unsigned
int()`` conversion. Boost.Python's dynamic type conversion
registry allows users to add arbitrary conversion methods.
==================
Library Overview
==================
This section outlines some of the library's major features. Except as
neccessary to avoid confusion, details of library implementation are
omitted.
------------------
Exposing Classes
------------------
C++ classes and structs are exposed with a similarly-terse interface.
Given::
struct World
{
void set(std::string msg) { this->msg = msg; }
std::string greet() { return msg; }
std::string msg;
};
The following code will expose it in our extension module::
#include <boost/python.hpp>
BOOST_PYTHON_MODULE(hello)
{
class_<World>("World")
.def("greet", &World::greet)
.def("set", &World::set)
;
}
Although this code has a certain pythonic familiarity, people
sometimes find the syntax bit confusing because it doesn't look like
most of the C++ code they're used to. All the same, this is just
standard C++. Because of their flexible syntax and operator
overloading, C++ and Python are great for defining domain-specific
(sub)languages
(DSLs), and that's what we've done in Boost.Python. To break it down::
class_<World>("World")
constructs an unnamed object of type ``class_<World>`` and passes
``"World"`` to its constructor. This creates a new-style Python class
called ``World`` in the extension module, and associates it with the
C++ type ``World`` in the Boost.Python type conversion registry. We
might have also written::
class_<World> w("World");
but that would've been more verbose, since we'd have to name ``w``
again to invoke its ``def()`` member function::
w.def("greet", &World::greet)
There's nothing special about the location of the dot for member
access in the original example: C++ allows any amount of whitespace on
either side of a token, and placing the dot at the beginning of each
line allows us to chain as many successive calls to member functions
as we like with a uniform syntax. The other key fact that allows
chaining is that ``class_<>`` member functions all return a reference
to ``*this``.
So the example is equivalent to::
class_<World> w("World");
w.def("greet", &World::greet);
w.def("set", &World::set);
It's occasionally useful to be able to break down the components of a
Boost.Python class wrapper in this way, but the rest of this article
will stick to the terse syntax.
For completeness, here's the wrapped class in use: ::
>>> import hello
>>> planet = hello.World()
>>> planet.set('howdy')
>>> planet.greet()
'howdy'
Constructors
============
Since our ``World`` class is just a plain ``struct``, it has an
implicit no-argument (nullary) constructor. Boost.Python exposes the
nullary constructor by default, which is why we were able to write: ::
>>> planet = hello.World()
However, well-designed classes in any language may require constructor
arguments in order to establish their invariants. Unlike Python,
where ``__init__`` is just a specially-named method, In C++
constructors cannot be handled like ordinary member functions. In
particular, we can't take their address: ``&World::World`` is an
error. The library provides a different interface for specifying
constructors. Given::
struct World
{
World(std::string msg); // added constructor
...
we can modify our wrapping code as follows::
class_<World>("World", init<std::string>())
...
of course, a C++ class may have additional constructors, and we can
expose those as well by passing more instances of ``init<...>`` to
``def()``::
class_<World>("World", init<std::string>())
.def(init<double, double>())
...
Boost.Python allows wrapped functions, member functions, and
constructors to be overloaded to mirror C++ overloading.
Data Members and Properties
===========================
Any publicly-accessible data members in a C++ class can be easily
exposed as either ``readonly`` or ``readwrite`` attributes::
class_<World>("World", init<std::string>())
.def_readonly("msg", &World::msg)
...
and can be used directly in Python: ::
>>> planet = hello.World('howdy')
>>> planet.msg
'howdy'
This does *not* result in adding attributes to the ``World`` instance
``__dict__``, which can result in substantial memory savings when
wrapping large data structures. In fact, no instance ``__dict__``
will be created at all unless attributes are explicitly added from
Python. Boost.Python owes this capability to the new Python 2.2 type
system, in particular the descriptor interface and ``property`` type.
In C++, publicly-accessible data members are considered a sign of poor
design because they break encapsulation, and style guides usually
dictate the use of "getter" and "setter" functions instead. In
Python, however, ``__getattr__``, ``__setattr__``, and since 2.2,
``property`` mean that attribute access is just one more
well-encapsulated syntactic tool at the programmer's disposal.
Boost.Python bridges this idiomatic gap by making Python ``property``
creation directly available to users. If ``msg`` were private, we
could still expose it as attribute in Python as follows::
class_<World>("World", init<std::string>())
.add_property("msg", &World::greet, &World::set)
...
The example above mirrors the familiar usage of properties in Python
2.2+: ::
>>> class World(object):
... __init__(self, msg):
... self.__msg = msg
... def greet(self):
... return self.__msg
... def set(self, msg):
... self.__msg = msg
... msg = property(greet, set)
Operator Overloading
====================
The ability to write arithmetic operators for user-defined types has
been a major factor in the success of both languages for numerical
computation, and the success of packages like NumPy_ attests to the
power of exposing operators in extension modules. Boost.Python
provides a concise mechanism for wrapping operator overloads. The
example below shows a fragment from a wrapper for the Boost rational
number library::
class_<rational<int> >("rational_int")
.def(init<int, int>()) // constructor, e.g. rational_int(3,4)
.def("numerator", &rational<int>::numerator)
.def("denominator", &rational<int>::denominator)
.def(-self) // __neg__ (unary minus)
.def(self + self) // __add__ (homogeneous)
.def(self * self) // __mul__
.def(self + int()) // __add__ (heterogenous)
.def(int() + self) // __radd__
...
The magic is performed using a simplified application of "expression
templates" [VELD1995]_, a technique originally developed for
optimization of high-performance matrix algebra expressions. The
essence is that instead of performing the computation immediately,
operators are overloaded to construct a type *representing* the
computation. In matrix algebra, dramatic optimizations are often
available when the structure of an entire expression can be taken into
account, rather than evaluating each operation "greedily".
Boost.Python uses the same technique to build an appropriate Python
method object based on expressions involving ``self``.
.. _NumPy: http://www.pfdubois.com/numpy/
Inheritance
===========
C++ inheritance relationships can be represented to Boost.Python by adding
an optional ``bases<...>`` argument to the ``class_<...>`` template
parameter list as follows::
class_<Derived, bases<Base1,Base2> >("Derived")
...
This has two effects:
1. When the ``class_<...>`` is created, Python type objects
corresponding to ``Base1`` and ``Base2`` are looked up in
Boost.Python's registry, and are used as bases for the new Python
``Derived`` type object, so methods exposed for the Python ``Base1``
and ``Base2`` types are automatically members of the ``Derived``
type. Because the registry is global, this works correctly even if
``Derived`` is exposed in a different module from either of its
bases.
2. C++ conversions from ``Derived`` to its bases are added to the
Boost.Python registry. Thus wrapped C++ methods expecting (a
pointer or reference to) an object of either base type can be
called with an object wrapping a ``Derived`` instance. Wrapped
member functions of class ``T`` are treated as though they have an
implicit first argument of ``T&``, so these conversions are
neccessary to allow the base class methods to be called for derived
objects.
Of course it's possible to derive new Python classes from wrapped C++
class instances. Because Boost.Python uses the new-style class
system, that works very much as for the Python built-in types. There
is one significant detail in which it differs: the built-in types
generally establish their invariants in their ``__new__`` function, so
that derived classes do not need to call ``__init__`` on the base
class before invoking its methods : ::
>>> class L(list):
... def __init__(self):
... pass
...
>>> L().reverse()
>>>
Because C++ object construction is a one-step operation, C++ instance
data cannot be constructed until the arguments are available, in the
``__init__`` function: ::
>>> class D(SomeBoostPythonClass):
... def __init__(self):
... pass
...
>>> D().some_boost_python_method()
Traceback (most recent call last):
File "<stdin>", line 1, in ?
TypeError: bad argument type for built-in operation
This happened because Boost.Python couldn't find instance data of type
``SomeBoostPythonClass`` within the ``D`` instance; ``D``'s ``__init__``
function masked construction of the base class. It could be corrected
by either removing ``D``'s ``__init__`` function or having it call
``SomeBoostPythonClass.__init__(...)`` explicitly.
Virtual Functions
=================
Deriving new types in Python from extension classes is not very
interesting unless they can be used polymorphically from C++. In
other words, Python method implementations should appear to override
the implementation of C++ virtual functions when called *through base
class pointers/references from C++*. Since the only way to alter the
behavior of a virtual function is to override it in a derived class,
the user must build a special derived class to dispatch a polymorphic
class' virtual functions::
//
// interface to wrap:
//
class Base
{
public:
virtual int f(std::string x) { return 42; }
virtual ~Base();
};
int calls_f(Base const& b, std::string x) { return b.f(x); }
//
// Wrapping Code
//
// Dispatcher class
struct BaseWrap : Base
{
// Store a pointer to the Python object
BaseWrap(PyObject* self_) : self(self_) {}
PyObject* self;
// Default implementation, for when f is not overridden
int f_default(std::string x) { return this->Base::f(x); }
// Dispatch implementation
int f(std::string x) { return call_method<int>(self, "f", x); }
};
...
def("calls_f", calls_f);
class_<Base, BaseWrap>("Base")
.def("f", &Base::f, &BaseWrap::f_default)
;
Now here's some Python code which demonstrates: ::
>>> class Derived(Base):
... def f(self, s):
... return len(s)
...
>>> calls_f(Base(), 'foo')
42
>>> calls_f(Derived(), 'forty-two')
9
Things to notice about the dispatcher class:
* The key element which allows overriding in Python is the
``call_method`` invocation, which uses the same global type
conversion registry as the C++ function wrapping does to convert its
arguments from C++ to Python and its return type from Python to C++.
* Any constructor signatures you wish to wrap must be replicated with
an initial ``PyObject*`` argument
* The dispatcher must store this argument so that it can be used to
invoke ``call_method``
* The ``f_default`` member function is needed when the function being
exposed is not pure virtual; there's no other way ``Base::f`` can be
called on an object of type ``BaseWrap``, since it overrides ``f``.
Deeper Reflection on the Horizon?
=================================
Admittedly, this formula is tedious to repeat, especially on a project
with many polymorphic classes. That it is neccessary reflects some
limitations in C++'s compile-time introspection capabilities: there's
no way to enumerate the members of a class and find out which are
virtual functions. At least one very promising project has been
started to write a front-end which can generate these dispatchers (and
other wrapping code) automatically from C++ headers.
Pyste builds on GCC_XML_, which generates an XML version of GCC's
internal program representation. Since GCC is a highly-conformant C++
compiler, this ensures correct handling of the most-sophisticated
template code and full access to the underlying type system. In
keeping with the Boost.Python philosophy, a Pyste interface
description is neither intrusive on the code being wrapped, nor
expressed in some unfamiliar language: instead it is a 100% pure
Python script. If Pyste is successful it will mark a move away from
wrapping everything directly in C++ for many of our users. We expect
that soon, not only our users but the Boost.Python developers
themselves will be "thinking hybrid" about their own code.
.. _`GCC_XML`: http://www.gccxml.org/HTML/Index.html
---------------
Serialization
---------------
*Serialization* is the process of converting objects in memory to a
form that can be stored on disk or sent over a network connection. The
serialized object (most often a plain string) can be retrieved and
converted back to the original object. A good serialization system will
automatically convert entire object hierarchies. Python's standard
``pickle`` module is just such a system. It leverages the language's strong
runtime introspection facilities for serializing practically arbitrary
user-defined objects. With a few simple and unintrusive provisions this
powerful machinery can be extended to also work for wrapped C++ objects.
Here is an example::
#include <string>
struct World
{
World(std::string a_msg) : msg(a_msg) {}
std::string greet() const { return msg; }
std::string msg;
};
#include <boost/python.hpp>
using namespace boost::python;
struct World_picklers : pickle_suite
{
static tuple
getinitargs(World const& w) { return make_tuple(w.greet()); }
};
BOOST_PYTHON_MODULE(hello)
{
class_<World>("World", init<std::string>())
.def("greet", &World::greet)
.def_pickle(World_picklers())
;
}
Now let's create a ``World`` object and put it to rest on disk::
>>> import hello
>>> import pickle
>>> a_world = hello.World("howdy")
>>> pickle.dump(a_world, open("my_world", "w"))
In a potentially *different script* on a potentially *different
computer* with a potentially *different operating system*::
>>> import pickle
>>> resurrected_world = pickle.load(open("my_world", "r"))
>>> resurrected_world.greet()
'howdy'
Of course the ``cPickle`` module can also be used for faster
processing.
Boost.Python's ``pickle_suite`` fully supports the ``pickle`` protocol
defined in the standard Python documentation. Like a __getinitargs__
function in Python, the pickle_suite's getinitargs() is responsible for
creating the argument tuple that will be use to reconstruct the pickled
object. The other elements of the Python pickling protocol,
__getstate__ and __setstate__ can be optionally provided via C++
getstate and setstate functions. C++'s static type system allows the
library to ensure at compile-time that nonsensical combinations of
functions (e.g. getstate without setstate) are not used.
Enabling serialization of more complex C++ objects requires a little
more work than is shown in the example above. Fortunately the
``object`` interface (see next section) greatly helps in keeping the
code manageable.
------------------
Object interface
------------------
Experienced 'C' language extension module authors will be familiar
with the ubiquitous ``PyObject*``, manual reference-counting, and the
need to remember which API calls return "new" (owned) references or
"borrowed" (raw) references. These constraints are not just
cumbersome but also a major source of errors, especially in the
presence of exceptions.
Boost.Python provides a class ``object`` which automates reference
counting and provides conversion to Python from C++ objects of
arbitrary type. This significantly reduces the learning effort for
prospective extension module writers.
Creating an ``object`` from any other type is extremely simple::
object s("hello, world"); // s manages a Python string
``object`` has templated interactions with all other types, with
automatic to-python conversions. It happens so naturally that it's
easily overlooked::
object ten_Os = 10 * s[4]; // -> "oooooooooo"
In the example above, ``4`` and ``10`` are converted to Python objects
before the indexing and multiplication operations are invoked.
The ``extract<T>`` class template can be used to convert Python objects
to C++ types::
double x = extract<double>(o);
If a conversion in either direction cannot be performed, an
appropriate exception is thrown at runtime.
The ``object`` type is accompanied by a set of derived types
that mirror the Python built-in types such as ``list``, ``dict``,
``tuple``, etc. as much as possible. This enables convenient
manipulation of these high-level types from C++::
dict d;
d["some"] = "thing";
d["lucky_number"] = 13;
list l = d.keys();
This almost looks and works like regular Python code, but it is pure
C++. Of course we can wrap C++ functions which accept or return
``object`` instances.
.. =====================
Development history
=====================
XXX Outline of development history to illustrate that the
library is mature. XXX
This can be postponed for the PyConDC paper
=================
Thinking hybrid
=================
Because of the practical and mental difficulties of combining
programming languages, it is common to settle a single language at the
outset of any development effort. For many applications, performance
considerations dictate the use of a compiled language for the core
algorithms. Unfortunately, due to the complexity of the static type
system, the price we pay for runtime performance is often a
significant increase in development time. Experience shows that
writing maintainable C++ code usually takes longer and requires *far*
more hard-earned working experience than developing comparable Python
code. Even when developers are comfortable working exclusively in
compiled languages, they often augment their systems by some type of
ad hoc scripting layer for the benefit of their users without ever
availing themselves of the same advantages.
Boost.Python enables us to *think hybrid*. Python can be used for
rapidly prototyping a new application; its ease of use and the large
pool of standard libraries give us a head start on the way to a
working system. If necessary, the working code can be used to
discover rate-limiting hotspots. To maximize performance these can
be reimplemented in C++, together with the Boost.Python bindings
needed to tie them back into the existing higher-level procedure.
Of course, this *top-down* approach is less attractive if it is clear
from the start that many algorithms will eventually have to be
implemented in C++. Fortunately Boost.Python also enables us to
pursue a *bottom-up* approach. We have used this approach very
successfully in the development of a toolbox for scientific
applications. The toolbox started out mainly as a library of C++
classes with Boost.Python bindings, and for a while the growth was
mainly concentrated on the C++ parts. However, as the toolbox is
becoming more complete, more and more newly added functionality can be
implemented in Python.
.. image:: python_cpp_mix.png
This figure shows the estimated ratio of newly added C++ and Python
code over time as new algorithms are implemented. We expect this
ratio to level out near 70% Python. Being able to solve new problems
mostly in Python rather than a more difficult statically typed
language is the return on our investment in Boost.Python. The ability
to access all of our code from Python allows a broader group of
developers to use it in the rapid development of new applications.
=============
Conclusions
=============
Boost.Python achieves seamless interoperability between two rich and
complimentary language environments. Because it leverages template
metaprogramming to introspect about types and functions, the user
never has to learn a third syntax: the interface definitions are
written in concise and maintainable C++. Also, the wrapping system
doesn't have to parse C++ headers or represent the type system: the
compiler does that work for us.
Computationally intensive tasks play to the strengths of C++ and are
often impossible to implement efficiently in pure Python, while jobs
like serialization that are trivial in Python can be very difficult in
pure C++. Given the luxury of building a hybrid software system from
the ground up, we can approach design with new confidence and power.
===========
Citations
===========
.. [VELD1995] T. Veldhuizen, "Expression Templates," C++ Report,
Vol. 7 No. 5 June 1995, pp. 26-31.
http://osl.iu.edu/~tveldhui/papers/Expression-Templates/exprtmpl.html

View File

@@ -1,908 +0,0 @@
.. This is a comment. Note how any initial comments are moved by
transforms to after the document title, subtitle, and docinfo.
.. Need intro and conclusion
.. Exposing classes
.. Constructors
.. Overloading
.. Properties and data members
.. Inheritance
.. Operators and Special Functions
.. Virtual Functions
.. Call Policies
++++++++++++++++++++++++++++++++++++++++++++++
Introducing Boost.Python (Extended Abstract)
++++++++++++++++++++++++++++++++++++++++++++++
.. bibliographic fields (which also require a transform):
:Author: David Abrahams
:Address: 45 Walnut Street
Somerville, MA 02143
:Contact: dave@boost-consulting.com
:organization: `Boost Consulting`_
:date: $Date$
:status: This is a "work in progress"
:version: 1
:copyright: Copyright David Abrahams 2002. All rights reserved
:Dedication:
For my girlfriend, wife, and partner Luann
:abstract:
This paper describes the Boost.Python library, a system for
C++/Python interoperability.
.. meta::
:keywords: Boost,python,Boost.Python,C++
:description lang=en: C++/Python interoperability with Boost.Python
.. contents:: Table of Contents
.. section-numbering::
.. _`Boost Consulting`: http://www.boost-consulting.com
==============
Introduction
==============
Python and C++ are in many ways as different as two languages could
be: while C++ is usually compiled to machine-code, Python is
interpreted. Python's dynamic type system is often cited as the
foundation of its flexibility, while in C++ static typing is the
cornerstone of its efficiency. C++ has an intricate and difficult
meta-language to support compile-time polymorphism, while Python is
a uniform language with convenient runtime polymorphism.
Yet for many programmers, these very differences mean that Python and
C++ complement one another perfectly. Performance bottlenecks in
Python programs can be rewritten in C++ for maximal speed, and
authors of powerful C++ libraries choose Python as a middleware
language for its flexible system integration capabilities.
Furthermore, the surface differences mask some strong similarities:
* 'C'-family control structures (if, while, for...)
* Support for object-orientation, functional programming, and generic
programming (these are both *multi-paradigm* programming languages.)
* Comprehensive operator overloading facilities, recognizing the
importance of syntactic variability for readability and
expressivity.
* High-level concepts such as collections and iterators.
* High-level encapsulation facilities (C++: namespaces, Python: modules)
to support the design of re-usable libraries.
* Exception-handling for effective management of error conditions.
* C++ idioms in common use, such as handle/body classes and
reference-counted smart pointers mirror Python reference semantics.
Python provides a rich 'C' API for writers of 'C' extension modules.
Unfortunately, using this API directly for exposing C++ type and
function interfaces to Python is much more tedious than it should be.
This is mainly due to the limitations of the 'C' language. Compared to
C++ and Python, 'C' has only very rudimentary abstraction facilities.
Support for exception-handling is completely missing. One important
undesirable consequence is that 'C' extension module writers are
required to manually manage Python reference counts. Another unpleasant
consequence is a very high degree of repetition of similar code in 'C'
extension modules. Of course highly redundant code does not only cause
frustration for the module writer, but is also very difficult to
maintain.
The limitations of the 'C' API have lead to the development of a
variety of wrapping systems. SWIG_ is probably the most popular package
for the integration of C/C++ and Python. A more recent development is
the SIP_ package, which is specifically designed for interfacing Python
with the Qt_ graphical user interface library. Both SWIG and SIP
introduce a new specialized language for defining the inter-language
bindings. Of course being able to use a specialized language has
advantages, but having to deal with three different languages (Python,
C/C++ and the interface language) also introduces practical and mental
difficulties. The CXX_ package demonstrates an interesting alternative.
It shows that at least some parts of Python's 'C' API can be wrapped
and presented through a much more user-friendly C++ interface. However,
unlike SWIG and SIP, CXX does not include support for wrapping C++
classes as new Python types. CXX is also no longer actively developed.
In some respects Boost.Python combines ideas from SWIG and SIP with
ideas from CXX. Like SWIG and SIP, Boost.Python is a system for
wrapping C++ classes as new Python "built-in" types, and C/C++
functions as Python functions. Like CXX, Boost.Python presents Python's
'C' API through a C++ interface. Boost.Python goes beyond the scope of
other systems with the unique support for C++ virtual functions that
are overrideable in Python, support for organizing extensions as Python
packages with a central registry for inter-language type conversions,
and a convenient mechanism for tying into Python's serialization engine
(pickle). Importantly, all this is achieved without introducing a new
syntax. Boost.Python leverages the power of C++ meta-programming
techniques to introspect about the C++ type system, and presents a
simple, IDL-like C++ interface for exposing C/C++ code in extension
modules. Boost.Python is a pure C++ library, the inter-language
bindings are defined in pure C++, and other than a C++ compiler only
Python itself is required to get started with Boost.Python. Last but
not least, Boost.Python is an unrestricted open source library. There
are no strings attached even for commercial applications.
.. _SWIG: http://www.swig.org/
.. _SIP: http://www.riverbankcomputing.co.uk/sip/index.php
.. _Qt: http://www.trolltech.com/
.. _CXX: http://cxx.sourceforge.net/
===========================
Boost.Python Design Goals
===========================
The primary goal of Boost.Python is to allow users to expose C++
classes and functions to Python using nothing more than a C++
compiler. In broad strokes, the user experience should be one of
directly manipulating C++ objects from Python.
However, it's also important not to translate all interfaces *too*
literally: the idioms of each language must be respected. For
example, though C++ and Python both have an iterator concept, they are
expressed very differently. Boost.Python has to be able to bridge the
interface gap.
It must be possible to insulate Python users from crashes resulting
from trivial misuses of C++ interfaces, such as accessing
already-deleted objects. By the same token the library should
insulate C++ users from low-level Python 'C' API, replacing
error-prone 'C' interfaces like manual reference-count management and
raw ``PyObject`` pointers with more-robust alternatives.
Support for component-based development is crucial, so that C++ types
exposed in one extension module can be passed to functions exposed in
another without loss of crucial information like C++ inheritance
relationships.
Finally, all wrapping must be *non-intrusive*, without modifying or
even seeing the original C++ source code. Existing C++ libraries have
to be wrappable by third parties who only have access to header files
and binaries.
==========================
Hello Boost.Python World
==========================
And now for a preview of Boost.Python, and how it improves on the raw
facilities offered by Python. Here's a function we might want to
expose::
char const* greet(unsigned x)
{
static char const* const msgs[] = { "hello", "Boost.Python", "world!" };
if (x > 2)
throw std::range_error("greet: index out of range");
return msgs[x];
}
To wrap this function in standard C++ using the Python 'C' API, we'd
need something like this::
extern "C" // all Python interactions use 'C' linkage and calling convention
{
// Wrapper to handle argument/result conversion and checking
PyObject* greet_wrap(PyObject* args, PyObject * keywords)
{
int x;
if (PyArg_ParseTuple(args, "i", &x)) // extract/check arguments
{
char const* result = greet(x); // invoke wrapped function
return PyString_FromString(result); // convert result to Python
}
return 0; // error occurred
}
// Table of wrapped functions to be exposed by the module
static PyMethodDef methods[] = {
{ "greet", greet_wrap, METH_VARARGS, "return one of 3 parts of a greeting" }
, { NULL, NULL, 0, NULL } // sentinel
};
// module initialization function
DL_EXPORT init_hello()
{
(void) Py_InitModule("hello", methods); // add the methods to the module
}
}
Now here's the wrapping code we'd use to expose it with Boost.Python::
#include <boost/python.hpp>
using namespace boost::python;
BOOST_PYTHON_MODULE(hello)
{
def("greet", greet, "return one of 3 parts of a greeting");
}
and here it is in action::
>>> import hello
>>> for x in range(3):
... print hello.greet(x)
...
hello
Boost.Python
world!
Aside from the fact that the 'C' API version is much more verbose than
the BPL one, it's worth noting that it doesn't handle a few things
correctly:
* The original function accepts an unsigned integer, and the Python
'C' API only gives us a way of extracting signed integers. The
Boost.Python version will raise a Python exception if we try to pass
a negative number to ``hello.greet``, but the other one will proceed
to do whatever the C++ implementation does when converting an
negative integer to unsigned (usually wrapping to some very large
number), and pass the incorrect translation on to the wrapped
function.
* That brings us to the second problem: if the C++ ``greet()``
function is called with a number greater than 2, it will throw an
exception. Typically, if a C++ exception propagates across the
boundary with code generated by a 'C' compiler, it will cause a
crash. As you can see in the first version, there's no C++
scaffolding there to prevent this from happening. Functions wrapped
by Boost.Python automatically include an exception-handling layer
which protects Python users by translating unhandled C++ exceptions
into a corresponding Python exception.
* A slightly more-subtle limitation is that the argument conversion
used in the Python 'C' API case can only get that integer ``x`` in
*one way*. PyArg_ParseTuple can't convert Python ``long`` objects
(arbitrary-precision integers) which happen to fit in an ``unsigned
int`` but not in a ``signed long``, nor will it ever handle a
wrapped C++ class with a user-defined implicit ``operator unsigned
int()`` conversion. The BPL's dynamic type conversion registry
allows users to add arbitrary conversion methods.
==================
Library Overview
==================
This section outlines some of the library's major features. Except as
neccessary to avoid confusion, details of library implementation are
omitted.
-------------------------------------------
The fundamental type-conversion mechanism
-------------------------------------------
XXX This needs to be rewritten.
Every argument of every wrapped function requires some kind of
extraction code to convert it from Python to C++. Likewise, the
function return value has to be converted from C++ to Python.
Appropriate Python exceptions must be raised if the conversion fails.
Argument and return types are part of the function's type, and much of
this tedium can be relieved if the wrapping system can extract that
information through introspection.
Passing a wrapped C++ derived class instance to a C++ function
accepting a pointer or reference to a base class requires knowledge of
the inheritance relationship and how to translate the address of a base
class into that of a derived class.
------------------
Exposing Classes
------------------
C++ classes and structs are exposed with a similarly-terse interface.
Given::
struct World
{
void set(std::string msg) { this->msg = msg; }
std::string greet() { return msg; }
std::string msg;
};
The following code will expose it in our extension module::
#include <boost/python.hpp>
BOOST_PYTHON_MODULE(hello)
{
class_<World>("World")
.def("greet", &World::greet)
.def("set", &World::set)
;
}
Although this code has a certain pythonic familiarity, people
sometimes find the syntax bit confusing because it doesn't look like
most of the C++ code they're used to. All the same, this is just
standard C++. Because of their flexible syntax and operator
overloading, C++ and Python are great for defining domain-specific
(sub)languages
(DSLs), and that's what we've done in BPL. To break it down::
class_<World>("World")
constructs an unnamed object of type ``class_<World>`` and passes
``"World"`` to its constructor. This creates a new-style Python class
called ``World`` in the extension module, and associates it with the
C++ type ``World`` in the BPL type conversion registry. We might have
also written::
class_<World> w("World");
but that would've been more verbose, since we'd have to name ``w``
again to invoke its ``def()`` member function::
w.def("greet", &World::greet)
There's nothing special about the location of the dot for member
access in the original example: C++ allows any amount of whitespace on
either side of a token, and placing the dot at the beginning of each
line allows us to chain as many successive calls to member functions
as we like with a uniform syntax. The other key fact that allows
chaining is that ``class_<>`` member functions all return a reference
to ``*this``.
So the example is equivalent to::
class_<World> w("World");
w.def("greet", &World::greet);
w.def("set", &World::set);
It's occasionally useful to be able to break down the components of a
Boost.Python class wrapper in this way, but the rest of this paper
will tend to stick to the terse syntax.
For completeness, here's the wrapped class in use:
>>> import hello
>>> planet = hello.World()
>>> planet.set('howdy')
>>> planet.greet()
'howdy'
Constructors
============
Since our ``World`` class is just a plain ``struct``, it has an
implicit no-argument (nullary) constructor. Boost.Python exposes the
nullary constructor by default, which is why we were able to write:
>>> planet = hello.World()
However, well-designed classes in any language may require constructor
arguments in order to establish their invariants. Unlike Python,
where ``__init__`` is just a specially-named method, In C++
constructors cannot be handled like ordinary member functions. In
particular, we can't take their address: ``&World::World`` is an
error. The library provides a different interface for specifying
constructors. Given::
struct World
{
World(std::string msg); // added constructor
...
we can modify our wrapping code as follows::
class_<World>("World", init<std::string>())
...
of course, a C++ class may have additional constructors, and we can
expose those as well by passing more instances of ``init<...>`` to
``def()``::
class_<World>("World", init<std::string>())
.def(init<double, double>())
...
Boost.Python allows wrapped functions, member functions, and
constructors to be overloaded to mirror C++ overloading.
Data Members and Properties
===========================
Any publicly-accessible data members in a C++ class can be easily
exposed as either ``readonly`` or ``readwrite`` attributes::
class_<World>("World", init<std::string>())
.def_readonly("msg", &World::msg)
...
and can be used directly in Python:
>>> planet = hello.World('howdy')
>>> planet.msg
'howdy'
This does *not* result in adding attributes to the ``World`` instance
``__dict__``, which can result in substantial memory savings when
wrapping large data structures. In fact, no instance ``__dict__``
will be created at all unless attributes are explicitly added from
Python. BPL owes this capability to the new Python 2.2 type system,
in particular the descriptor interface and ``property`` type.
In C++, publicly-accessible data members are considered a sign of poor
design because they break encapsulation, and style guides usually
dictate the use of "getter" and "setter" functions instead. In
Python, however, ``__getattr__``, ``__setattr__``, and since 2.2,
``property`` mean that attribute access is just one more
well-encapsulated syntactic tool at the programmer's disposal. BPL
bridges this idiomatic gap by making Python ``property`` creation
directly available to users. So if ``msg`` were private, we could
still expose it as attribute in Python as follows::
class_<World>("World", init<std::string>())
.add_property("msg", &World::greet, &World::set)
...
The example above mirrors the familiar usage of properties in Python
2.2+:
>>> class World(object):
... __init__(self, msg):
... self.__msg = msg
... def greet(self):
... return self.__msg
... def set(self, msg):
... self.__msg = msg
... msg = property(greet, set)
Operators and Special Functions
===============================
The ability to write arithmetic operators for user-defined types that
C++ and Python both allow the definition of has been a major factor in
the popularity of both languages for scientific computing. The
success of packages like NumPy attests to the power of exposing
operators in extension modules. In this example we'll wrap a class
representing a position in a large file::
class FilePos { /*...*/ };
// Linear offset
FilePos operator+(FilePos, int);
FilePos operator+(int, FilePos);
FilePos operator-(FilePos, int);
// Distance between two FilePos objects
int operator-(FilePos, FilePos);
// Offset with assignment
FilePos& operator+=(FilePos&, int);
FilePos& operator-=(FilePos&, int);
// Comparison
bool operator<(FilePos, FilePos);
The wrapping code looks like this::
class_<FilePos>("FilePos")
.def(self + int()) // __add__
.def(int() + self) // __radd__
.def(self - int()) // __sub__
.def(self - self) // __sub__
.def(self += int()) // __iadd__
.def(self -= int()) // __isub__
.def(self < self); // __lt__
;
The magic is performed using a simplified application of "expression
templates" [VELD1995]_, a technique originally developed by for
optimization of high-performance matrix algebra expressions. The
essence is that instead of performing the computation immediately,
operators are overloaded to construct a type *representing* the
computation. In matrix algebra, dramatic optimizations are often
available when the structure of an entire expression can be taken into
account, rather than processing each operation "greedily".
Boost.Python uses the same technique to build an appropriate Python
callable object based on an expression involving ``self``, which is
then added to the class.
Inheritance
===========
C++ inheritance relationships can be represented to Boost.Python by adding
an optional ``bases<...>`` argument to the ``class_<...>`` template
parameter list as follows::
class_<Derived, bases<Base1,Base2> >("Derived")
...
This has two effects:
1. When the ``class_<...>`` is created, Python type objects
corresponding to ``Base1`` and ``Base2`` are looked up in the BPL
registry, and are used as bases for the new Python ``Derived`` type
object [#mi]_, so methods exposed for the Python ``Base1`` and
``Base2`` types are automatically members of the ``Derived`` type.
Because the registry is global, this works correctly even if
``Derived`` is exposed in a different module from either of its
bases.
2. C++ conversions from ``Derived`` to its bases are added to the
Boost.Python registry. Thus wrapped C++ methods expecting (a
pointer or reference to) an object of either base type can be
called with an object wrapping a ``Derived`` instance. Wrapped
member functions of class ``T`` are treated as though they have an
implicit first argument of ``T&``, so these conversions are
neccessary to allow the base class methods to be called for derived
objects.
Of course it's possible to derive new Python classes from wrapped C++
class instances. Because Boost.Python uses the new-style class
system, that works very much as for the Python built-in types. There
is one significant detail in which it differs: the built-in types
generally establish their invariants in their ``__new__`` function, so
that derived classes do not need to call ``__init__`` on the base
class before invoking its methods :
>>> class L(list):
... def __init__(self):
... pass
...
>>> L().reverse()
>>>
Because C++ object construction is a one-step operation, C++ instance
data cannot be constructed until the arguments are available, in the
``__init__`` function:
>>> class D(SomeBPLClass):
... def __init__(self):
... pass
...
>>> D().some_bpl_method()
Traceback (most recent call last):
File "<stdin>", line 1, in ?
TypeError: bad argument type for built-in operation
This happened because Boost.Python couldn't find instance data of type
``SomeBPLClass`` within the ``D`` instance; ``D``'s ``__init__``
function masked construction of the base class. It could be corrected
by either removing ``D``'s ``__init__`` function or having it call
``SomeBPLClass.__init__(...)`` explicitly.
Virtual Functions
=================
Deriving new types in Python from extension classes is not very
interesting unless they can be used polymorphically from C++. In
other words, Python method implementations should appear to override
the implementation of C++ virtual functions when called *through base
class pointers/references from C++*. Since the only way to alter the
behavior of a virtual function is to override it in a derived class,
the user must build a special derived class to dispatch a polymorphic
class' virtual functions::
//
// interface to wrap:
//
class Base
{
public:
virtual int f(std::string x) { return 42; }
virtual ~Base();
};
int calls_f(Base const& b, std::string x) { return b.f(x); }
//
// Wrapping Code
//
// Dispatcher class
struct BaseWrap : Base
{
// Store a pointer to the Python object
BaseWrap(PyObject* self_) : self(self_) {}
PyObject* self;
// Default implementation, for when f is not overridden
int f_default(std::string x) { return this->Base::f(x); }
// Dispatch implementation
int f(std::string x) { return call_method<int>(self, "f", x); }
};
...
def("calls_f", calls_f);
class_<Base, BaseWrap>("Base")
.def("f", &Base::f, &BaseWrap::f_default)
;
Now here's some Python code which demonstrates:
>>> class Derived(Base):
... def f(self, s):
... return len(s)
...
>>> calls_f(Base(), 'foo')
42
>>> calls_f(Derived(), 'forty-two')
9
Things to notice about the dispatcher class:
* The key element which allows overriding in Python is the
``call_method`` invocation, which uses the same global type
conversion registry as the C++ function wrapping does to convert its
arguments from C++ to Python and its return type from Python to C++.
* Any constructor signatures you wish to wrap must be replicated with
an initial ``PyObject*`` argument
* The dispatcher must store this argument so that it can be used to
invoke ``call_method``
* The ``f_default`` member function is needed when the function being
exposed is not pure virtual; there's no other way ``Base::f`` can be
called on an object of type ``BaseWrap``, since it overrides ``f``.
Admittedly, this formula is tedious to repeat, especially on a project
with many polymorphic classes; that it is neccessary reflects
limitations in C++'s compile-time reflection capabilities. Several
efforts are underway to write front-ends for Boost.Python which can
generate these dispatchers (and other wrapping code) automatically.
If these are successful it will mark a move away from wrapping
everything directly in pure C++ for many of our users.
---------------
Serialization
---------------
*Serialization* is the process of converting objects in memory to a
form that can be stored on disk or sent over a network connection. The
serialized object (most often a plain string) can be retrieved and
converted back to the original object. A good serialization system will
automatically convert entire object hierarchies. Python's standard
``pickle`` module is such a system. It leverages the language's strong
runtime introspection facilities for serializing practically arbitrary
user-defined objects. With a few simple and unintrusive provisions this
powerful machinery can be extended to also work for wrapped C++ objects.
Here is an example::
#include <string>
struct World
{
World(std::string a_msg) : msg(a_msg) {}
std::string greet() const { return msg; }
std::string msg;
};
#include <boost/python.hpp>
using namespace boost::python;
struct World_picklers : pickle_suite
{
static tuple
getinitargs(World const& w) { return make_tuple(w.greet()); }
};
BOOST_PYTHON_MODULE(hello)
{
class_<World>("World", init<std::string>())
.def("greet", &World::greet)
.def_pickle(World_picklers())
;
}
Now let's create a ``World`` object and put it to rest on disk::
>>> import hello
>>> import pickle
>>> a_world = hello.World("howdy")
>>> pickle.dump(a_world, open("my_world", "w"))
In a potentially *different script* on a potentially *different
computer* with a potentially *different operating system*::
>>> import pickle
>>> resurrected_world = pickle.load(open("my_world", "r"))
>>> resurrected_world.greet()
'howdy'
Of course the ``cPickle`` module can also be used for faster
processing.
Boost.Python's ``pickle_suite`` fully supports the ``pickle`` protocol
defined in the standard Python documentation. There is a one-to-one
correspondence between the standard pickling methods (``__getinitargs__``,
``__getstate__``, ``__setstate__``) and the functions defined by the
user in the class derived from ``pickle_suite`` (``getinitargs``,
``getstate``, ``setstate``). The ``class_::def_pickle()`` member function
is used to establish the Python bindings for all user-defined functions
simultaneously. Correct signatures for these functions are enforced at
compile time. Non-sensical combinations of the three pickle functions
are also rejected at compile time. These measures are designed to
help the user in avoiding obvious errors.
Enabling serialization of more complex C++ objects requires a little
more work than is shown in the example above. Fortunately the
``object`` interface (see next section) greatly helps in keeping the
code manageable.
------------------
Object interface
------------------
Experienced extension module authors will be familiar with the 'C' view
of Python objects, the ubiquitous ``PyObject*``. Most if not all Python
'C' API functions involve ``PyObject*`` as arguments or return type. A
major complication is the raw reference counting interface presented to
the 'C' programmer. E.g. some API functions return *new references* and
others return *borrowed references*. It is up to the extension module
writer to properly increment and decrement reference counts. This
quickly becomes cumbersome and error prone, especially if there are
multiple execution paths.
Boost.Python provides a type ``object`` which is essentially a high
level wrapper around ``PyObject*``. ``object`` automates reference
counting as much as possible. It also provides the facilities for
converting arbitrary C++ types to Python objects and vice versa.
This significantly reduces the learning effort for prospective
extension module writers.
Creating an ``object`` from any other type is extremely simple::
object o(3);
``object`` has templated interactions with all other types, with
automatic to-python conversions. It happens so naturally that it's
easily overlooked.
The ``extract<T>`` class template can be used to convert Python objects
to C++ types::
double x = extract<double>(o);
All registered user-defined conversions are automatically accessible
through the ``object`` interface. With reference to the ``World`` class
defined in previous examples::
object as_python_object(World("howdy"));
World back_as_c_plus_plus_object = extract<World>(as_python_object);
If a C++ type cannot be converted to a Python object an appropriate
exception is thrown at runtime. Similarly, an appropriate exception is
thrown if a C++ type cannot be extracted from a Python object.
``extract<T>`` provides facilities for avoiding exceptions if this is
desired.
The ``object::attr()`` member function is available for accessing
and manipulating attributes of Python objects. For example::
object planet(World());
planet.attr("set")("howdy");
``planet.attr("set")`` returns a callable ``object``. ``"howdy"`` is
converted to a Python string object which is then passed as an argument
to the ``set`` method.
The ``object`` type is accompanied by a set of derived types
that mirror the Python built-in types such as ``list``, ``dict``,
``tuple``, etc. as much as possible. This enables convenient
manipulation of these high-level types from C++::
dict d;
d["some"] = "thing";
d["lucky_number"] = 13;
list l = d.keys();
This almost looks and works like regular Python code, but it is pure C++.
=================
Thinking hybrid
=================
For many applications runtime performance considerations are very
important. This is particularly true for most scientific applications.
Often the performance considerations dictate the use of a compiled
language for the core algorithms. Traditionally the decision to use a
particular programming language is an exclusive one. Because of the
practical and mental difficulties of combining different languages many
systems are written in just one language. This is quite unfortunate
because the price payed for runtime performance is typically a
significant overhead due to static typing. For example, our experience
shows that developing maintainable C++ code is typically much more
time-consuming and requires much more hard-earned working experience
than developing useful Python code. A related observation is that many
compiled packages are augmented by some type of rudimentary scripting
layer. These ad hoc solutions clearly show that many times a compiled
language alone does not get the job done. On the other hand it is also
clear that a pure Python implementation is too slow for numerically
intensive production code.
Boost.Python enables us to *think hybrid* when developing new
applications. Python can be used for rapidly prototyping a
new application. Python's ease of use and the large pool of standard
libraries give us a head start on the way to a first working system. If
necessary, the working procedure can be used to discover the
rate-limiting algorithms. To maximize performance these can be
reimplemented in C++, together with the Boost.Python bindings needed to
tie them back into the existing higher-level procedure.
Of course, this *top-down* approach is less attractive if it is clear
from the start that many algorithms will eventually have to be
implemented in a compiled language. Fortunately Boost.Python also
enables us to pursue a *bottom-up* approach. We have used this approach
very successfully in the development of a toolbox for scientific
applications (scitbx) that we will describe elsewhere. The toolbox
started out mainly as a library of C++ classes with Boost.Python
bindings, and for a while the growth was mainly concentrated on the C++
parts. However, as the toolbox is becoming more complete, more and more
newly added functionality can be implemented in Python. We expect this
trend to continue, as illustrated qualitatively in this figure:
.. image:: python_cpp_mix.png
This figure shows the ratio of newly added C++ and Python code over
time as new algorithms are implemented. We expect this ratio to level
out near 70% Python. The increasing ability to solve new problems
mostly with the easy-to-use Python language rather than a necessarily
more arcane statically typed language is the return on the investment
of learning how to use Boost.Python. The ability to solve some problems
entirely using only Python will enable a larger group of people to
participate in the rapid development of new applications.
=============
Conclusions
=============
The examples in this paper illustrate that Boost.Python enables
seamless interoperability between C++ and Python. Importantly, this is
achieved without introducing a third syntax: the Python/C++ interface
definitions are written in pure C++. This avoids any problems with
parsing the C++ code to be interfaced to Python, yet the interface
definitions are concise and maintainable. Freed from most of the
development-time penalties of crossing a language boundary, software
designers can take full advantage of two rich and complimentary
language environments. In practice it turns out that some things are
very difficult to do with pure Python/C (e.g. an efficient array
library with an intuitive interface in the compiled language) and
others are very difficult to do with pure C++ (e.g. serialization).
If one has the luxury of being able to design a software system as a
hybrid system from the ground up there are many new ways of avoiding
road blocks in one language or the other.
.. I'm not ready to give up on all of this quite yet
.. Perhaps one day we'll have a language with the simplicity and
expressive power of Python and the compile-time muscle of C++. Being
able to take advantage of all of these facilities without paying the
mental and development-time penalties of crossing a language barrier
would bring enormous benefits. Until then, interoperability tools
like Boost.Python can help lower the barrier and make the benefits of
both languages more accessible to both communities.
===========
Footnotes
===========
.. [#mi] For hard-core new-style class/extension module writers it is
worth noting that the normal requirement that all extension classes
with data form a layout-compatible single-inheritance chain is
lifted for Boost.Python extension classes. Clearly, either
``Base1`` or ``Base2`` has to occupy a different offset in the
``Derived`` class instance. This is possible because the wrapped
part of BPL extension class instances is never assumed to have a
fixed offset within the wrapper.
===========
Citations
===========
.. [VELD1995] T. Veldhuizen, "Expression Templates," C++ Report,
Vol. 7 No. 5 June 1995, pp. 26-31.
http://osl.iu.edu/~tveldhui/papers/Expression-Templates/exprtmpl.html

View File

@@ -1,188 +0,0 @@
/*
:Author: David Goodger
:Contact: goodger@users.sourceforge.net
:date: $Date$
:version: $Revision$
:copyright: This stylesheet has been placed in the public domain.
Default cascading style sheet for the HTML output of Docutils.
*/
.first {
margin-top: 0 }
.last {
margin-bottom: 0 }
a.toc-backref {
text-decoration: none ;
color: black }
dd {
margin-bottom: 0.5em }
div.abstract {
margin: 2em 5em }
div.abstract p.topic-title {
font-weight: bold ;
text-align: center }
div.attention, div.caution, div.danger, div.error, div.hint,
div.important, div.note, div.tip, div.warning {
margin: 2em ;
border: medium outset ;
padding: 1em }
div.attention p.admonition-title, div.caution p.admonition-title,
div.danger p.admonition-title, div.error p.admonition-title,
div.warning p.admonition-title {
color: red ;
font-weight: bold ;
font-family: sans-serif }
div.hint p.admonition-title, div.important p.admonition-title,
div.note p.admonition-title, div.tip p.admonition-title {
font-weight: bold ;
font-family: sans-serif }
div.dedication {
margin: 2em 5em ;
text-align: center ;
font-style: italic }
div.dedication p.topic-title {
font-weight: bold ;
font-style: normal }
div.figure {
margin-left: 2em }
div.footer, div.header {
font-size: smaller }
div.system-messages {
margin: 5em }
div.system-messages h1 {
color: red }
div.system-message {
border: medium outset ;
padding: 1em }
div.system-message p.system-message-title {
color: red ;
font-weight: bold }
div.topic {
margin: 2em }
h1.title {
text-align: center }
h2.subtitle {
text-align: center }
hr {
width: 75% }
ol.simple, ul.simple {
margin-bottom: 1em }
ol.arabic {
list-style: decimal }
ol.loweralpha {
list-style: lower-alpha }
ol.upperalpha {
list-style: upper-alpha }
ol.lowerroman {
list-style: lower-roman }
ol.upperroman {
list-style: upper-roman }
p.caption {
font-style: italic }
p.credits {
font-style: italic ;
font-size: smaller }
p.label {
white-space: nowrap }
p.topic-title {
font-weight: bold }
pre.address {
margin-bottom: 0 ;
margin-top: 0 ;
font-family: serif ;
font-size: 100% }
pre.line-block {
font-family: serif ;
font-size: 100% }
pre.literal-block, pre.doctest-block {
margin-left: 2em ;
margin-right: 2em ;
background-color: #eeeeee }
span.classifier {
font-family: sans-serif ;
font-style: oblique }
span.classifier-delimiter {
font-family: sans-serif ;
font-weight: bold }
span.interpreted {
font-family: sans-serif }
span.option-argument {
font-style: italic }
span.pre {
white-space: pre }
span.problematic {
color: red }
table {
margin-top: 0.5em ;
margin-bottom: 0.5em }
table.citation {
border-left: solid thin gray ;
padding-left: 0.5ex }
table.docinfo {
margin: 2em 4em }
table.footnote {
border-left: solid thin black ;
padding-left: 0.5ex }
td, th {
padding-left: 0.5em ;
padding-right: 0.5em ;
vertical-align: top }
th.docinfo-name, th.field-name {
font-weight: bold ;
text-align: left ;
white-space: nowrap }
h1 tt, h2 tt, h3 tt, h4 tt, h5 tt, h6 tt {
font-size: 100% }
tt {
background-color: #eeeeee }
ul.auto-toc {
list-style-type: none }

Binary file not shown.

Before

Width:  |  Height:  |  Size: 6.1 KiB

View File

@@ -1,59 +0,0 @@
H1
{
FONT-SIZE: 200%
COLOR: #00007f
}
H2
{
FONT-SIZE: 150%;
}
H3
{
FONT-SIZE: 125%;
}
H4
{
FONT-SIZE: 108%;
}
BODY
{
FONT-SIZE: 100%;
BACKGROUND-COLOR: #ffffff
}
PRE
{
MARGIN-LEFT: 2pc;
FONT-SIZE: 80%;
BACKGROUND-COLOR: #dfffff
}
CODE
{
FONT-SIZE: 95%;
white-space: pre
}
.index
{
TEXT-ALIGN: left
}
.page-index
{
TEXT-ALIGN: left
}
.definition
{
TEXT-ALIGN: left
}
.footnote
{
FONT-SIZE: 66%;
VERTICAL-ALIGN: super;
TEXT-DECORATION: none
}
.function-semantics
{
CLEAR: left
}
.metafunction-semantics
{
CLEAR: left
}

View File

@@ -1,402 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="boost.css">
<title>Boost.Python - Building and Testing</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="index.html">Boost.Python</a></h1>
<h2 align="center">Building and Testing</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="Reference">
<dt><a href="#requirements">Requirements</a></dt>
<dt><a href="#building">Building Boost.Python</a></dt>
<dd>
<dl class="index">
<dt><a href="#configuration">Configuration</a></dt>
<dt><a href="#results">Results</a></dt>
<dt><a href="#cygwin">Notes for Cygwin GCC Users</a></dt>
<dt><a href="#testing">Testing</a></dt>
</dl>
</dd>
<dt><a href="#building_ext">Building your Extension Module</a></dt>
<dd>
<dl>
<dt><a href="#easy">The Easy Way</a></dt>
<dt><a href="#outside">Building your module outside the Boost
project tree</a></dt>
</dl>
</dd>
<dt><a href="#variants">Build Variants</a></dt>
<dt><a href="#VisualStudio">Building Using the Microsoft Visual Studio
IDE</a></dt>
</dl>
<hr>
<h2><a name="requirements">Requirements</a></h2>
<b>Boost.Python</b> version 2 requires <a href=
"http://www.python.org/2.2">Python 2.2</a> <i>or <a href=
"http://www.python.org">newer</a></i>. An unsupported archive of
Boost.Python version 1, which works with versions of Python since 1.5.2,
is available <a href="../build/python_v1.zip">here</a>.
<h2><a name="building">Building Boost.Python</a></h2>
<p>Normally, Boost.Python extension modules must be linked with the
<code>boost_python</code> shared library. In special circumstances you
may want to link to a static version of the <code>boost_python</code>
library, but if multiple Boost.Pythone extension modules are used
together, it will prevent sharing of types across extension modules, and
consume extra code space. To build <code>boost_python</code>, use <a
href="../../../tools/build/index.html">Boost.Build</a> in the usual way
from the <code>libs/python/build</code> subdirectory of your boost
installation (if you have already built boost from the top level this may
have no effect, since the work is already done).</p>
<h3><a name="configuration">Configuration</a></h3>
You may need to configure the following variables to point Boost.Build at
your Python installation:
<table border="1" summary="build configuration variables">
<tr>
<th>Variable Name</th>
<th>Semantics</th>
<th>Default</th>
<th>Notes</th>
</tr>
<tr>
<td><code>PYTHON_ROOT</code></td>
<td>The root directory of your Python installation</td>
<td>Windows:&nbsp;<code>c:/tools/python</code>
Unix:&nbsp;<code>/usr/local</code></td>
<td>On Unix, this is the <code>--with-prefix=</code> directory used
to configure Python</td>
</tr>
<tr>
<td><code>PYTHON_VERSION</code></td>
<td>The The 2-part python Major.Minor version number</td>
<td><code>2.2</code></td>
<td>Be sure not to include a third number, e.g. <b>not</b>
"<code>2.2.1</code>", even if that's the version you have.</td>
</tr>
<tr>
<td><code>PYTHON_INCLUDES</code></td>
<td>path to Python <code>#include</code> directories</td>
<td>Autoconfigured from <code>PYTHON_ROOT</code></td>
</tr>
<tr>
<td><code>PYTHON_LIB_PATH</code></td>
<td>path to Python library object.</td>
<td>Autoconfigured from <code>PYTHON_ROOT</code></td>
</tr>
<tr>
<td><code>CYGWIN_PYTHON_[DEBUG_]VERSION</code></td>
<td>The version of python being used under Cygwin. </td>
<td>$(PYTHON_VERSION)
</td>
<td>Use only when building with <a href=
"http://www.cygwin.com">Cygwin</a> GCC. This and the following
settings are useful when building with multiple toolsets on
Windows, since Cygwin GCC requires a different build of
Python.</td> </tr>
<tr>
<td><code>CYGWIN_PYTHON_[DEBUG_]ROOT</code></td>
<td>unix-style path containing the <code>include/</code>
directory containing
<code>python$(CYGWIN_PYTHON_[DEBUG_]VERSION)/python.h</code>. </td>
<td>$(PYTHON_ROOT)
</td>
<td>Use only when building with <a href=
"http://www.cygwin.com">Cygwin</a> GCC.</td> </tr>
<tr>
<td><code>CYGWIN_PYTHON_[DEBUG_]LIB_PATH</code></td>
<td>path containing the user's Cygwin Python import lib
<code>libpython$(CYGWIN_PYTHON_[DEBUG_]VERSION).dll.a</code></td>
<td>Autoconfigured from <code>CYGWIN_PYTHON_ROOT</code></td>
<td>Use only when building with <a href=
"http://www.cygwin.com">Cygwin</a> GCC.</td> </tr>
<tr>
<td><code>CYGWIN_PYTHON_[DEBUG_]DLL_PATH</code></td>
<td>path containing the user's Cygwin Python dll
(<code>libpython$(CYGWIN_PYTHON_[DEBUG_]VERSION).dll</code>)</td>
<td><code>/bin</code></td>
<td>Use only when building with <a href=
"http://www.cygwin.com">Cygwin</a> GCC.</td> </tr>
</tr>
</table>
<h3><a name="cygwin">Notes for Cygwin GCC Users</a></h3>
<p>If you are using Cygwin GCC to build extension modules, you must use a
Cygwin build of Python. The regular Win32 Python installation that you
can download from <a href="http://www.python.org">python.org</a> will not
work with your compiler because the dynamic linking conventions are
different (you can use <a href="http://www.mingw.org/">MinGW</a> GCC if
you want to build extension modules which are compatible with a stock
Win32 Python). The Cygwin installer may be able to install an appropriate
version of Python, or you can follow the traditional <a href=
"http://www.python.org/download/download_source.html">Unix installation
process</a> to build Python from source.</p>
<p>The special build configuration variables listed above as "Cygwin
only" make it possible to use a regular Win32 build of bjam to build and
test Boost.Python and Boost.Python extensions using Cygwin GCC and
targeting a Cygwin build of Python.</p>
<h3><a name="results">Results</a></h3>
<p>The build process will create a
<code>libs/python/build/bin-stage</code> subdirectory of the boost root
(or of <code>$(ALL_LOCATE_TARGET)</code>, if you have set that variable),
containing the built libraries. The libraries are actually built to
unique directories for each toolset and variant elsewhere in the
filesystem, and copied to the <code>bin-stage</code> directory as a
convenience, so if you build with multiple toolsets at once, the product
of later toolsets will overwrite that of earlier toolsets in
<code>bin-stage</code>.</p>
<h3><a name="testing">Testing</a></h3>
<p>To build and test Boost.Python, start from the
<code>libs/python/test</code> directory and invoke</p>
<blockquote>
<pre>
bjam -sTOOLS=<i><a href=
"../../../tools/build/index.html#Tools">toolset</a></i> test
</pre>
</blockquote>
This will update all of the Boost.Python v1 test and example targets. The
tests are relatively quiet by default. To get more-verbose output, you
might try
<blockquote>
<pre>
bjam -sTOOLS=<i><a href=
"../../../tools/build/index.html#Tools">toolset</a></i> -sPYTHON_TEST_ARGS=-v test
</pre>
</blockquote>
which will print each test's Python code with the expected output as it
passes.
<h2><a name="building_ext">Building your Extension Module</a></h2>
Though there are other approaches, the best way to build an extension
module using Boost.Python is with Boost.Build. If you have to use another
build system, you should use Boost.Build at least once with the
"<code><b>-n</b></code>" option so you can see the command-lines it uses,
and replicate them. You are likely to run into compilation or linking
problems otherwise.
<h3><a name="easy">The Easy Way</a></h3>
Until Boost.Build v2 is released, cross-project build dependencies are
not supported, so it works most smoothly if you add a new subproject to
your boost installation. The <code>libs/python/example</code>
subdirectory of your boost installation contains a minimal example (along
with many extra sources). To copy the example subproject:
<ol>
<li>Create a new subdirectory in, <code>libs/python</code>, say
<code>libs/python/my_project</code>.</li>
<li>Copy <code><a href=
"../example/Jamfile">libs/python/example/Jamfile</a></code> to your new
directory.</li>
<li>Edit the Jamfile as appropriate for your project. You'll want to
change the "<code>subproject</code>" rule invocation at the top, and
the names of some of the source files and/or targets.</li>
</ol>
The instructions <a href="#testing">above</a> for testing Boost.Python
apply equally to your new extension modules in this subproject.
<h3><a name="outside">Building your module outside the Boost project
tree</a></h3>
If you can't (or don't wish to) modify your boost installation, the
alternative is to create your own Boost.Build project. A similar example
you can use as a starting point is available in <code><a href=
"../example/project.zip">this archive</a></code>. You'll need to edit the
Jamfile and Jamrules files, depending on the relative location of your
Boost installation and the new project. Note that automatic testing of
extension modules is not available in this configuration.
<h2><a name="variants">Build Variants</a></h2>
Three <a href=
"../../../tools/build/build_system.htm#variants">variant</a>
configurations of all python-related targets are supported, and can be
selected by setting the <code><a href=
"../../../tools/build/build_system.htm#user_globals">BUILD</a></code>
variable:
<ul>
<li><code>release</code> (optimization, <tt>-DNDEBUG</tt>)</li>
<li><code>debug</code> (no optimization <tt>-D_DEBUG</tt>)</li>
<li><code>debug-python</code> (no optimization, <tt>-D_DEBUG
-DBOOST_DEBUG_PYTHON</tt>)</li>
</ul>
<p>The first two variants of the <code>boost_python</code> library are
built by default, and are compatible with the default Python
distribution. The <code>debug-python</code> variant corresponds to a
specially-built debugging version of Python. On Unix platforms, this
python is built by adding <code>--with-pydebug</code> when configuring
the Python build. On Windows, the debugging version of Python is
generated by the "Win32 Debug" target of the <code>PCBuild.dsw</code>
Visual C++ 6.0 project in the <code>PCBuild</code> subdirectory of your
Python distribution. Extension modules built with Python debugging
enabled are <b>not link-compatible</b> with a non-debug build of Python.
Since few people actually have a debug build of Python (it doesn't come
with the standard distribution), the normal <code>debug</code> variant
builds modules which are compatible with ordinary Python.</p>
<p>On many windows compilers, when extension modules are built with
<tt>-D_DEBUG</tt>, Python defaults to <i>force</i> linking with a special
debugging version of the Python DLL. Since this debug DLL isn't supplied
with the default Python installation for Windows, Boost.Python uses
<tt><a href=
"../../../boost/python/detail/wrap_python.hpp">boost/python/detail/wrap_python.hpp</a></tt>
to temporarily undefine <tt>_DEBUG</tt> when <tt>Python.h</tt> is
<tt>#include</tt>d - unless <code>BOOST_DEBUG_PYTHON</code> is
defined.</p>
<p>If you want the extra runtime checks available with the debugging
version of the library, <tt>#define BOOST_DEBUG_PYTHON</tt> to re-enable
python debuggin, and link with the <code>debug-python</code> variant of
<tt>boost_python</tt>.</p>
<p>If you do not <tt>#define BOOST_DEBUG_PYTHON</tt>, be sure that any
source files in your extension module <tt>#include&nbsp;&lt;<a href=
"../../../boost/python/detail/wrap_python.hpp">boost/python/detail/wrap_python.hpp</a>&gt;</tt>
instead of the usual <tt>Python.h</tt>, or you will have link
incompatibilities.<br>
</p>
<h2><a name="VisualStudio">Building Using the Microsoft Visual Studio
IDE</a></h2>
<p>For the those of you who feel more comfortable in the IDE world, a
workspace and project file have been included in the <a href=
"../build/VisualStudio">libs/python/build/VisualStudio</a> subdirectory.
It builds release and debug versions of the Boost.Python libraries and
places them and the same directory as Jamfile build does, though the
intermediate object files are placed in a different directory. The files
have been created using Microsoft Visual C++ version 6, but they should
work for later versions as well. You will need to tell the IDE where to
find the Python <code>Include/</code> and <code>Libs/</code> directories.
Under <b>Tools&gt;Options&gt;Directories</b>, add an entry for the Python
include dir (i.e. <code>c:/Python22/Include</code>), and one for the Lib
(i.e. <code>c:/Python/Libs</code>. Make sure it is <code>Libs</code> with
an "<code>s</code>" and not just <code>Lib</code>).</p>
<h3>Using the IDE for you own projects</h3>
<p>Building your own projects using the IDE is slightly more complicated.
Firstly, you need to make sure that the project you create as the right
kind. It should be a "Win32 Dynamic-Link Library". The default one that
Visual Studio 6 creates needs some modifications: turn on RTTI, and
change the debug and release builds to use the respective debug and
release Multithreaded DLL versions. You should probably turn off
incremental linking too -- I believe it a bit flaky. If you do this, then
change the "Debug Info" to "Program Database" to get rid of the Edit and
Continue warning.</p>
<p>You'll need to add the Boost root directory under
<b>Tools&gt;Options&gt;Directories</b> to get your code compiling. To
make it link, add the above <code>boost_python.dsp</code> file to your
workspace, and make your project depend upon it (under
<b>Project&gt;Dependencies</b>). You should be able to build now.</p>
<p>Lastly, go to the <b>Project Settings&gt;Debug</b> Page and add the
<code>Python.exe</code> as the executable for the project. Set a startup
directory, and make sure that your current project's output dll, the
<code>boost_python.dll</code> and the <code>python22.dll</code> are on
the current <code>PATH</code>. If you have a python script that tests
your dll, then add it in the "Program Arguments". Now, if all went well,
you should be able to hit the Run (F5) button, and debug your code.</p>
<blockquote>
<em>The Visual Studio project files are graciously contributed and
maintained by <a href="mailto:brett.calcott@paradise.net.nz">Brett
Calcott</a></em>.
</blockquote>
<hr>
<p>&copy; Copyright David Abrahams 2002. Permission to copy, use, modify,
sell and distribute this document is granted provided this copyright
notice appears in all copies. This document is provided ``as is'' without
express or implied warranty, and with no claim as to its suitability for
any purpose.</p>
<p>Updated: 29 December, 2002 (David Abrahams)</p>
</body>
</html>

View File

@@ -1,108 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="boost.css">
<title>Boost.Python</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center">Boost.Python</h1>
<h2 align="center">Index</h2>
</td>
</tr>
</table>
<hr>
<h2>Synopsis</h2>
Welcome to version 2 of <b>Boost.Python</b>, a C++ library which enables
seamless interoperability between C++ and the <a href=
"http://www.python.org">Python</a> programming language. The new version
has been rewritten from the ground up, with a more convenient and
flexible interface, and many new capabilities, including support for:
<ul>
<li>References and Pointers</li>
<li>Globally Registered Type Coercions</li>
<li>Automatic Cross-Module Type Conversions</li>
<li>Efficient Function Overloading</li>
<li>C++ to Python Exception Translation</li>
<li>Default Arguments</li>
<li>Keyword Arguments</li>
<li>Manipulating Python objects in C++</li>
<li>Exporting C++ Iterators as Python Iterators</li>
<li>Documentation Strings</li>
</ul>
The development of these features was funded in part by grants to <a
href="http://www.boost-consulting.com">Boost Consulting</a> from the <a
href="http://www.llnl.gov/">Lawrence Livermore National Laboratories</a>
and by the <a href="http://cci.lbl.gov/">Computational Crystallography
Initiative</a> at Lawrence Berkeley National Laboratories.
<hr>
<h2>Contents</h2>
<dl class="index">
<dt><a href="tutorial/index.html">Tutorial Introduction</a></dt>
<dt><a href="building.html">Building and Testing</a></dt>
<dt><a href="v2/reference.html">Reference Manual</a></dt>
<dt><a href="v2/configuration.html">Configuration Information</a></dt>
<dt><a href="v2/platforms.html">Known Working Platforms and
Compilers</a></dt>
<dt><a href="v2/definitions.html">Definitions</a></dt>
<dt><a href="projects.html">Projects using Boost.Python</a></dt>
<dt><a href="support.html">Support Resources</a></dt>
<dt><a href="v2/faq.html">Frequently Asked Questions (FAQs)</a></dt>
<dt><a href="../pyste/index.html">Pyste (Boost.Python code generator)</a></dt>
<dt><a href="news.html">News/Change Log</a></dt>
<dt><a href="v2/progress_reports.html">LLNL Progress Reports</a></dt>
<dt><a href="v2/acknowledgments.html">Acknowledgments</a></dt>
</dl>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
17 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,328 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=windows-1252">
<title>A New Type Conversion Mechanism for Boost.Python</title>
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p><img border="0" src="../../../c++boost.gif" width="277" height="86"
alt="boost logo"></p>
<h1>A New Type Conversion Mechanism for Boost.Python</h1>
<p>By <a href="../../../people/dave_abrahams.htm">David Abrahams</a>.
<h2>Introduction</h2>
This document describes a redesign of the mechanism for automatically
converting objects between C++ and Python. The current implementation
uses two functions for any type <tt>T</tt>:
<blockquote><pre>
U from_python(PyObject*, type&lt;T&gt;);
void to_python(V);
</pre></blockquote>
where U is convertible to T and T is convertible to V. These functions
are at the heart of C++/Python interoperability in Boost.Python, so
why would we want to change them? There are many reasons:
<h3>Bugs</h3>
<p>Firstly, the current mechanism relies on a common C++ compiler
bug. This is not just embarrassing: as compilers get to be more
conformant, the library stops working. The issue, in detail, is the
use of inline friend functions in templates to generate
conversions. It is a very powerful, and legal technique as long as
it's used correctly:
<blockquote><pre>
template &lt;class Derived&gt;
struct add_some_functions
{
friend <i>return-type</i> some_function1(..., Derived <i>cv-*-&amp;-opt</i>, ...);
friend <i>return-type</i> some_function2(..., Derived <i>cv-*-&amp;-opt</i>, ...);
};
template &lt;class T&gt;
struct some_template : add_some_functions&lt;some_template&lt;T&gt; &gt;
{
};
</pre></blockquote>
The <tt>add_some_functions</tt> template generates free functions
which operate on <tt>Derived</tt>, or on related types. Strictly
speaking the related types are not just cv-qualified <tt>Derived</tt>
values, pointers and/or references. Section 3.4.2 in the standard
describes exactly which types you must use as parameters to these
functions if you want the functions to be found
(there is also a less-technical description in section 11.5.1 of
C++PL3 <a href="#ref_1">[1]</a>). Suffice it to say that
with the current design, the <tt>from_python</tt> and
<tt>to_python</tt> functions are not supposed to be callable under any
conditions!
<h3>Compilation and Linking Time</h3>
The conversion functions generated for each wrapped class using the
above technique are not function templates, but regular functions. The
upshot is that they must <i>all</i> be generated regardless of whether
they are actually used. Generating all of those functions can slow
down module compilation, and resolving the references can slow down
linking.
<h3>Efficiency</h3>
The conversion functions are primarily used in (member) function
wrappers to convert the arguments and return values. Being functions,
converters have no interface which allows us to ask &quot;will the
conversion succeed?&quot; without calling the function. Since the
return value of the function must be the object to be passed as an
argument, Boost.Python currently uses C++ exception-handling to detect
an unsuccessful conversion. It's not a particularly good use of
exception-handling, since the failure is not handled very far from
where it occurred. More importantly, it means that C++ exceptions are
thrown during overload resolution as we seek an overload that matches
the arguments passed. Depending on the implementation, this approach
can result in significant slowdowns.
<p>It is also unclear that the current library generates a minimal
amount of code for any type conversion. Many of the conversion
functions are nontrivial, and partly because of compiler limitations,
they are declared <tt>inline</tt>. Also, we could have done a better
job separating the type-specific conversion code from the code which
is type-independent.
<h3>Cross-module Support</h3>
The current strategy requires every module to contain the definition
of conversions it uses. In general, a new module can never supply
conversion code which is used by another module. Ralf Grosse-Kunstleve
designed a clever system which imports conversions directly from one
library into another using some explicit declarations, but it has some
disadvantages also:
<ol>
<li>The system Ullrich Koethe designed for implicit conversion between
wrapped classes related through inheritance does not currently work if
the classes are defined in separate modules.
<li>The writer of the importing module is required to know the name of
the module supplying the imported conversions.
<li>There can be only one way to extract any given C++ type from a
Python object in a given module.
</ol>
The first item might be addressed by moving Boost.Python into a shared
library, but the other two cannot. Ralf turned the limitation in item
two into a feature: the required module is loaded implicitly when a
conversion it defines is invoked. We will probably want to provide
that functionality anyway, but it's not clear that we should require
the declaration of all such conversions. The final item is a more
serious limitation. If, for example, new numeric types are defined in
separate modules, and these types can all be converted to
<tt>double</tt>s, we have to choose just one conversion method.
<h3>Ease-of-use</h3>
One persistent source of confusion for users of Boost.Python has been
the fact that conversions for a class are not be visible at
compile-time until the declaration of that class has been seen. When
the user tries to expose a (member) function operating on or returning
an instance of the class in question, compilation fails...even though
the user goes on to expose the class in the same translation unit!
<p>
The new system lifts all compile-time checks for the existence of
particular type conversions and replaces them with runtime checks, in
true Pythonic style. While this might seem cavalier, the compile-time
checks are actually not much use in the current system if many classes
are wrapped in separate modules, since the checks are based only on
the user's declaration that the conversions exist.
<h2>The New Design</h2>
<h3>Motivation</h3>
The new design was heavily influenced by a desire to generate as
little code as possible in extension modules. Some of Boost.Python's
clients are enormous projects where link time is proportional to the
amount of object code, and there are many Python extension modules. As
such, we try to keep type-specific conversion code out of modules
other than the one the converters are defined in, and rely as much as
possible on centralized control through a shared library.
<h3>The Basics</h3>
The library contains a <tt>registry</tt> which maps runtime type
identifiers (actually an extension of <tt>std::type_info</tt> which
preserves references and constness) to entries containing type
converters. An <tt>entry</tt> can contain only one converter from C++ to Python
(<tt>wrapper</tt>), but many converters from Python to C++
(<tt>unwrapper</tt>s). <font color="#ff0000">What should happen if
multiple modules try to register wrappers for the same type?</font>. Wrappers
and unwrappers are known as <tt>body</tt> objects, and are accessed
by the user and the library (in its function-wrapping code) through
corresponding <tt>handle</tt> (<tt>wrap&lt;T&gt;</tt> and
<tt>unwrap&lt;T&gt;</tt>) objects. The <tt>handle</tt> objects are
extremely lightweight, and delegate <i>all</i> of their operations to
the corresponding <tt>body</tt>.
<p>
When a <tt>handle</tt> object is constructed, it accesses the
registry to find a corresponding <tt>body</tt> that can convert the
handle's constructor argument. Actually the registry record for any
type
<tt>T</tt>used in a module is looked up only once and stored in a
static <tt>registration&lt;T&gt;</tt> object for efficiency. For
example, if the handle is an <tt>unwrap&lt;Foo&amp;&gt;</tt> object,
the <tt>entry</tt> for <tt>Foo&amp;</tt> is looked up in the
<tt>registry</tt>, and each <tt>unwrapper</tt> it contains is queried
to determine if it can convert the
<tt>PyObject*</tt> with which the <tt>unwrap</tt> was constructed. If
a body object which can perform the conversion is found, a pointer to
it is stored in the handle. A body object may at any point store
additional data in the handle to speed up the conversion process.
<p>
Now that the handle has been constructed, the user can ask it whether
the conversion can be performed. All handles can be tested as though
they were convertible to <tt>bool</tt>; a <tt>true</tt> value
indicates success. If the user forges ahead and tries to do the
conversion without checking when no conversion is possible, an
exception will be thrown as usual. The conversion itself is performed
by the body object.
<h3>Handling complex conversions</h3>
<p>Some conversions may require a dynamic allocation. For example,
when a Python tuple is converted to a <tt>std::vector&lt;double&gt;
const&amp;</tt>, we need some storage into which to construct the
vector so that a reference to it can be formed. Furthermore, multiple
conversions of the same type may need to be &quot;active&quot;
simultaneously, so we can't keep a single copy of the storage
anywhere. We could keep the storage in the <tt>body</tt> object, and
have the body clone itself in case the storage is used, but in that
case the storage in the body which lives in the registry is never
used. If the storage was actually an object of the target type (the
safest way in C++), we'd have to find a way to construct one for the
body in the registry, since it may not have a default constructor.
<p>
The most obvious way out of this quagmire is to allocate the object using a
<i>new-expression</i>, and store a pointer to it in the handle. Since
the <tt>body</tt> object knows everything about the data it needs to
allocate (if any), it is also given responsibility for destroying that
data. When the <tt>handle</tt> is destroyed it asks the <tt>body</tt>
object to tear down any data it may have stored there. In many ways,
you can think of the <tt>body</tt> as a &quot;dynamically-determined
vtable&quot; for the handle.
<h3>Eliminating Redundancy</h3>
If you look at the current Boost.Python code, you'll see that there
are an enormous number of conversion functions generated for each
wrapped class. For a given class <tt>T</tt>, functions are generated
to extract the following types <tt>from_python</tt>:
<blockquote><pre>
T*
T const*
T const* const&amp;
T* const&amp;
T&amp;
T const&amp;
T
std::auto_ptr&lt;T&gt;&amp;
std::auto_ptr&lt;T&gt;
std::auto_ptr&lt;T&gt; const&amp;
boost::shared_ptr&lt;T&gt;&amp;
boost::shared_ptr&lt;T&gt;
boost::shared_ptr&lt;T&gt; const&amp;
</pre></blockquote>
Most of these are implemented in terms of just a few conversions, and
<t>if you're lucky</t>, they will be inlined and cause no extra
overhead. In the new system, however, a significant amount of data
will be associated with each type that needs to be converted. We
certainly don't want to register a separate unwrapper object for all
of the above types.
<p>Fortunately, much of the redundancy can be eliminated. For example,
if we generate an unwrapper for <tt>T&</tt>, we don't need an
unwrapper for <tt>T const&</tt> or <tt>T</tt>. Accordingly, the user's
request to wrap/unwrap a given type is translated at compile-time into
a request which helps to eliminate redundancy. The rules used to
<tt>unwrap</tt> a type are:
<ol>
<li> Treat built-in types specially: when unwrapping a value or
constant reference to one of these, use a value for the target
type. It will bind to a const reference if neccessary, and more
importantly, avoids having to dynamically allocate room for
an lvalue of types which can be cheaply copied.
<li>
Reduce everything else to a reference to an un-cv-qualified type
where possible. Since cv-qualification is lost on Python
anyway, there's no point in trying to convert to a
<tt>const&amp;</tt>. <font color="#ff0000">What about conversions
to values like the tuple-&gt;vector example above? It seems to me
that we don't want to make a <tt>vector&lt;double&gt;&amp;</tt>
(non-const) converter available for that case. We may need to
rethink this slightly.</font>
</ol>
<p>To handle the problem described above in item 2, we modify the
procedure slightly. To unwrap any non-scalar <tt>T</tt>, we seek an
unwrapper for <tt>add_reference&lt;T&gt;::type</tt>. Unwrappers for
<tt>T&nbsp;const&amp;</tt> always return <tt>T&amp;</tt>, and are
registered under both <tt>T&nbsp;&amp;</tt> and
<tt>T&nbsp;const&amp;</tt>.
<p>For compilers not supporting partial specialization, unwrappers for
<tt>T&nbsp;const&amp;</tt> must return <tt>T&nbsp;const&amp;</tt>
(since constness can't be stripped), but a separate unwrapper object
need to be registered for <tt>T&nbsp;&amp;</tt> and
<tt>T&nbsp;const&amp;</tt> anyway, for the same reasons.
<font color="#ff0000">We may want to make it possible to compile as
though partial specialization were unavailable even on compilers where
it is available, in case modules could be compiled by different
compilers with compatible ABIs (e.g. Intel C++ and MSVC6).</font>
<h3>Efficient Argument Conversion</h3>
Since type conversions are primarily used in function wrappers, an
optimization is provided for the case where a group of conversions are
used together. Each <tt>handle</tt> class has a corresponding
&quot;<tt>_more</tt>&quot; class which does the same job, but has a
trivial destructor. Instead of asking each &quot;<tt>_more</tt>&quot;
handle to destroy its own body, it is linked into an endogenous list
managed by the first (ordinary) handle. The <tt>wrap</tt> and
<tt>unwrap</tt> destructors are responsible for traversing that list
and asking each <tt>body</tt> class to tear down its
<tt>handle</tt>. This mechanism is also used to determine if all of
the argument/return-value conversions can succeed with a single
function call in the function wrapping code. <font color="#ff0000">We
might need to handle return values in a separate step for Python
callbacks, since the availablility of a conversion won't be known
until the result object is retrieved.</font>
<br>
<hr>
<h2>References</h2>
<p><a name="ref_1">[1]</a>B. Stroustrup, The C++ Programming Language
Special Edition Addison-Wesley, ISBN 0-201-70073-5.
<hr>
<p>Revised <!--webbot bot="Timestamp" s-type="EDITED" s-format="%d %B %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="31283" --></p>
<p>© Copyright David Abrahams, 2001</p>
</body>
</html>

View File

@@ -1,111 +0,0 @@
This hierarchy contains converter handle classes.
+-------------+
| noncopyable |
+-------------+
^
| A common base class used so that
+--------+--------+ conversions can be linked into a
| conversion_base | chain for efficient argument
+-----------------+ conversion
^
|
+---------+-----------+
| |
+-----------+----+ +------+-------+ only used for
| unwrap_more<T> | | wrap_more<T> | chaining, and don't manage any
+----------------+ +--------------+ resources.
^ ^
| |
+-----+-----+ +-------+-+ These converters are what users
| unwrap<T> | | wrap<T> | actually touch, but they do so
+-----------+ +---------+ through a type generator which
minimizes the number of converters
that must be generated, so they
Each unwrap<T>, unwrap_more<T>, wrap<T>, wrap_more<T> converter holds
a reference to an appropriate converter object
This hierarchy contains converter body classes
Exposes use/release which
are needed in case the converter
+-----------+ in the registry needs to be
| converter | cloned. That occurs when a
+-----------+ unwrap target type is not
^ contained within the Python object.
|
+------------------+-----+
| |
+--------+-------+ Exposes |
| unwrapper_base | convertible() |
+----------------+ |
^ |
| |
+--------+----+ +-----+-----+
| unwrapper<T>| | wrapper<T>|
+-------------+ +-----------+
Exposes T convert(PyObject*) Exposes PyObject* convert(T)
unwrap:
constructed with a PyObject*, whose reference count is
incremented.
find the registry entry for the target type
look in the collection of converters for one which claims to be
able to convert the PyObject to the target type.
stick a pointer to the unwrapper in the unwrap object
when unwrap is queried for convertibility, it checks to see
if it has a pointer to an unwrapper.
on conversion, the unwrapper is asked to allocate an
implementation if the unwrap object isn't already holding
one. The unwrap object "takes ownership" of the unwrapper's
implementation. No memory allocation will actually take place
unless this is a value conversion.
on destruction, the unwrapper is asked to free any implementation
held by the unwrap object. No memory deallocation actually
takes place unless this is a value conversion
on destruction, the reference count on the held PyObject is
decremented.
We need to make sure that by default, you can't instantiate
callback<> for reference and pointer return types: although the
unwrappers may exist, they may convert by-value, which would cause
the referent to be destroyed upon return.
wrap:
find the registry entry for the source type
see if there is a converter. If found, stick a pointer to it in
the wrap object.
when queried for convertibility, it checks to see if it has a
pointer to a converter.
on conversion, a reference to the target PyObject is held by the
converter. Generally, the PyObject will have been created by the
converter, but in certain cases it may be a pre-existing object,
whose reference count will have been incremented.
when a wrap<T> x is used to return from a C++ function,
x.release() is returned so that x no longer holds a reference to
the PyObject when destroyed.
Otherwise, on destruction, any PyObject still held has its
reference-count decremented.
When a converter is created by the user, the appropriate element must
be added to the registry; when it is destroyed, it must be removed
from the registry.

View File

@@ -1,109 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="boost.css">
<title>Boost.Python - News/Change Log</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="index.html">Boost.Python</a></h1>
<h2 align="center">News/Change Log</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt>24 February 2003</dt>
<dd>Finished improved support
for <code>boost::shared_ptr</code>. Now any wrapped object of
C++ class <code>X</code> can be converted automatically
to <code>shared_ptr&lt;X&gt;</code>, regardless of how it was
wrapped. The <code>shared_ptr</code> will manage the lifetime
of the Python object which supplied the <code>X</code>, rather
than just the <code>X</code> object itself, and when such
a <code>shared_ptr</code> is converted back to Python, the
original Python object will be returned.</dd>
<dt>19 January 2003</dt>
<dd>Integrated <code>staticmethod</code> support from <a href=
"mailto:nickm-at-sitius.com">Nikolay Mladenov</a>. Thanks,
Nikolay!</dd>
<dt>29 December 2002</dt>
<dd>Added Visual Studio project file and instructions from Brett
Calcott. Thanks, Brett!</dd>
<dt>20 December 2002</dt>
<dd>Added automatic downcasting for pointers, references, and smart
pointers to polymorphic class types upon conversion to python</dd>
<dt>18 December 2002</dt>
<dd>Optimized from_python conversions for wrapped classes by putting
the conversion logic in the shared library instead of registering
separate converters for each class in each extension module</dd>
<dt>19 November 2002</dt>
<dd>Removed the need for users to cast base class member function
pointers when used as arguments to <a href=
"v2/class.html#class_-spec-modifiers">add_property</a></dd>
<dt>13 December 2002</dt>
<dd>Allow exporting of <a href=
"v2/enum.html#enum_-spec"><code>enum_</code></a> values into enclosing
<a href="v2/scope.html#scope-spec"><code>scope</code></a>.<br>
Fixed unsigned integer conversions to deal correctly with numbers that
are out-of-range of <code>signed long</code>.</dd>
<dt>14 November 2002</dt>
<dd>Auto-detection of class data members wrapped with <a href=
"v2/data_members.html#make_getter-spec"><code>make_getter</code></a></dd>
<dt>13 November 2002</dt>
<dd>Full Support for <code>std::auto_ptr&lt;&gt;</code> added.</dd>
<dt>October 2002</dt>
<dd>Ongoing updates and improvements to tutorial documentation</dd>
<dt>10 October 2002</dt>
<dd>Boost.Python V2 is released!</dd>
</dl>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
20 December, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i></p>
</body>
</html>

223
doc/pickle.html Normal file
View File

@@ -0,0 +1,223 @@
<html>
<head>
<title>BPL Pickle Support</title>
</head>
<body>
<img src="../../../c++boost.gif"
alt="c++boost.gif (8819 bytes)"
align="center"
width="277" height="86">
</body>
<hr>
<h1>BPL Pickle Support</h1>
Pickle is a Python module for object serialization, also known
as persistence, marshalling, or flattening.
<p>
It is often necessary to save and restore the contents of an object to
a file. One approach to this problem is to write a pair of functions
that read and write data from a file in a special format. A powerful
alternative approach is to use Python's pickle module. Exploiting
Python's ability for introspection, the pickle module recursively
converts nearly arbitrary Python objects into a stream of bytes that
can be written to a file.
<p>
The Boost Python Library supports the pickle module by emulating the
interface implemented by Jim Fulton's ExtensionClass module that is
included in the ZOPE distribution
(<a href="http://www.zope.org/">http://www.zope.org/</a>).
This interface is similar to that for regular Python classes as
described in detail in the Python Library Reference for pickle:
<blockquote>
<a href="http://www.python.org/doc/current/lib/module-pickle.html"
>http://www.python.org/doc/current/lib/module-pickle.html</a>
</blockquote>
<hr>
<h1>The BPL Pickle Interface</h1>
At the user level, the BPL pickle interface involves three special
methods:
<dl>
<dt>
<strong>__getinitargs__</strong>
<dd>
When an instance of a BPL extension class is pickled, the pickler
tests if the instance has a __getinitargs__ method. This method must
return a Python tuple. When the instance is restored by the
unpickler, the contents of this tuple are used as the arguments for
the class constructor.
<p>
If __getinitargs__ is not defined, the class constructor will be
called without arguments.
<p>
<dt>
<strong>__getstate__</strong>
<dd>
When an instance of a BPL extension class is pickled, the pickler
tests if the instance has a __getstate__ method. This method should
return a Python object representing the state of the instance.
<p>
If __getstate__ is not defined, the instance's __dict__ is pickled
(if it is not empty).
<p>
<dt>
<strong>__setstate__</strong>
<dd>
When an instance of a BPL extension class is restored by the
unpickler, it is first constructed using the result of
__getinitargs__ as arguments (see above). Subsequently the unpickler
tests if the new instance has a __setstate__ method. If so, this
method is called with the result of __getstate__ (a Python object) as
the argument.
<p>
If __setstate__ is not defined, the result of __getstate__ must be
a Python dictionary. The items of this dictionary are added to
the instance's __dict__.
</dl>
If both __getstate__ and __setstate__ are defined, the Python object
returned by __getstate__ need not be a dictionary. The __getstate__ and
__setstate__ methods can do what they want.
<hr>
<h1>Pitfalls and Safety Guards</h1>
In BPL extension modules with many extension classes, providing
complete pickle support for all classes would be a significant
overhead. In general complete pickle support should only be implemented
for extension classes that will eventually be pickled. However, the
author of a BPL extension module might not anticipate correctly which
classes need support for pickle. Unfortunately, the pickle protocol
described above has two important pitfalls that the end user of a BPL
extension module might not be aware of:
<dl>
<dt>
<strong>Pitfall 1:</strong>
Both __getinitargs__ and __getstate__ are not defined.
<dd>
In this situation the unpickler calls the class constructor without
arguments and then adds the __dict__ that was pickled by default to
that of the new instance.
<p>
However, most C++ classes wrapped with the BPL will have member data
that are not restored correctly by this procedure. To alert the user
to this problem, a safety guard is provided. If both __getinitargs__
and __getstate__ are not defined, the BPL tests if the class has an
attribute __dict_defines_state__. An exception is raised if this
attribute is not defined:
<pre>
RuntimeError: Incomplete pickle support (__dict_defines_state__ not set)
</pre>
In the rare cases where this is not the desired behavior, the safety
guard can deliberately be disabled. The corresponding C++ code for
this is, e.g.:
<pre>
class_builder<your_class> py_your_class(your_module, "your_class");
py_your_class.dict_defines_state();
</pre>
It is also possible to override the safety guard at the Python level.
E.g.:
<pre>
import your_bpl_module
class your_class(your_bpl_module.your_class):
__dict_defines_state__ = 1
</pre>
<p>
<dt>
<strong>Pitfall 2:</strong>
__getstate__ is defined and the instance's __dict__ is not empty.
<dd>
The author of a BPL extension class might provide a __getstate__
method without considering the possibilities that:
<p>
<ul>
<li>
his class is used as a base class. Most likely the __dict__ of
instances of the derived class needs to be pickled in order to
restore the instances correctly.
<p>
<li>
the user adds items to the instance's __dict__ directly. Again,
the __dict__ of the instance then needs to be pickled.
</ul>
<p>
To alert the user to this highly unobvious problem, a safety guard is
provided. If __getstate__ is defined and the instance's __dict__ is
not empty, the BPL tests if the class has an attribute
__getstate_manages_dict__. An exception is raised if this attribute
is not defined:
<pre>
RuntimeError: Incomplete pickle support (__getstate_manages_dict__ not set)
</pre>
To resolve this problem, it should first be established that the
__getstate__ and __setstate__ methods manage the instances's __dict__
correctly. Note that this can be done both at the C++ and the Python
level. Finally, the safety guard should intentionally be overridden.
E.g. in C++:
<pre>
class_builder<your_class> py_your_class(your_module, "your_class");
py_your_class.getstate_manages_dict();
</pre>
In Python:
<pre>
import your_bpl_module
class your_class(your_bpl_module.your_class):
__getstate_manages_dict__ = 1
def __getstate__(self):
# your code here
def __setstate__(self, state):
# your code here
</pre>
</dl>
<hr>
<h1>Practical Advice</h1>
<ul>
<li>
Avoid using __getstate__ if the instance can also be reconstructed
by way of __getinitargs__. This automatically avoids Pitfall 2.
<p>
<li>
If __getstate__ is required, include the instance's __dict__ in the
Python object that is returned.
</ul>
<hr>
<address>
Author: Ralf W. Grosse-Kunstleve, March 2001
</address>
</html>

View File

@@ -1,217 +0,0 @@
How Runtime Polymorphism is expressed in Boost.Python:
-----------------------------------------------------
struct A { virtual std::string f(); virtual ~A(); };
std::string call_f(A& x) { return x.f(); }
struct B { virtual std::string f() { return "B"; } };
struct Bcb : B
{
Bcb(PyObject* self) : m_self(self) {}
virtual std::string f() { return call_method<std::string>(m_sef, "f"); }
static std::string f_default(B& b) { return b.B::f(); }
PyObject* m_self;
};
struct C : B
{
virtual std::string f() { return "C"; }
};
>>> class D(B):
... def f():
... return 'D'
...
>>> class E(B): pass
...
When we write, "invokes B::f non-virtually", we mean:
void g(B& x) { x.B::f(); }
This will call B::f() regardless of the dynamic type of x. Any other
way of invoking B::f, including through a function pointer, is a
"virtual invocation", and will call the most-derived override of f().
Case studies
C++\Python class
\___A_____B_____C_____D____E___
|
A | 1
|
B | 2 3
|
Bcb | 4 5 6
|
C | 7 8
|
1. Simple case
2. Python A holds a B*. Probably won't happen once we have forced
downcasting.
Requires:
x.f() -> 'B'
call_f(x) -> 'B'
Implies: A.f invokes A::f() (virtually or otherwise)
3. Python B holds a B*.
Requires:
x.f() -> 'B'
call_f(x) -> 'B'
Implies: B.f invokes B::f (virtually or otherwise)
4. B constructed from Python
Requires:
x.f() -> 'B'
call_f(x) -> 'B'
Implies: B.f invokes B::f non-virtually. Bcb::f invokes B::f
non-virtually.
Question: Does it help if we arrange for Python B construction to
build a true B object? Then this case doesn't arise.
5. D is a Python class derived from B
Requires:
x.f() -> 'D'
call_f(x) -> 'D'
Implies: Bcb::f must invoke call_method to look up the Python
method override, otherwise call_f wouldn't work.
6. E is like D, but doesn't override f
Requires:
x.f() -> 'B'
call_f(x) -> 'B'
Implies: B.f invokes B::f non-virtually. If it were virtual, x.f()
would cause infinite recursion, because we've already
determined that Bcb::f must invoke call_method to look up
the Python method override.
7. Python B object holds a C*
Requires:
x.f() -> 'C'
call_f(x) -> 'C'
Implies: B.f invokes B::f virtually.
8. C object constructed from Python
Requires:
x.f() -> 'C'
call_f(x) -> 'C'
Implies: nothing new.
------
Total implications:
2: A.f invokes A::f() (virtually or otherwise)
3: B.f invokes B::f (virtually or otherwise)
4: B.f invokes B::f non-virtually. Bcb::f invokes B::f non-virtually
6: B.f invokes B::f non-virtually.
7: B.f invokes B::f virtually.
5: Bcb::f invokes call_method to look up the Python method
Though (4) is avoidable, clearly 6 and 7 are not, and they
conflict. The implication is that B.f must choose its behavior
according to the type of the contained C++ object. If it is Bcb, a
non-virtual call to B::f must occur. Otherwise, a virtual call to B::f
must occur. This is essentially the same scheme we had with
Boost.Python v1.
Note: in early versions of Boost.Python v1, we solved this problem by
introducing a new Python class in the hierarchy, so that D and E
actually derive from a B', and B'.f invokes B::f non-virtually, while
B.f invokes B::f virtually. However, people complained about the
artificial class in the hierarchy, which was revealed when they tried
to do normal kinds of Python introspection.
-------
Assumption: we will have a function which builds a virtual function
dispatch callable Python object.
make_virtual_function(pvmf, default_impl, call_policies, dispatch_type)
Pseudocode:
Get first argument from Python arg tuple
if it contains dispatch_type
call default_impl
else
call through pvmf
Open questions:
1. What about Python multiple inheritance? Do we have the right
check in the if clause above?
A: Not quite. The correct test looks like:
Deduce target type of pvmf, i.e. T in R(T::*)(A1...AN).
Find holder in first argument which holds T
if it holds dispatch_type...
2. Can we make this more efficient?
The current "returning" mechanism will look up a holder for T
again. I don't know if we know how to avoid that.
OK, the solution involves reworking the call mechanism. This is
neccesary anyway in order to enable wrapping of function objects.
It can result in a reduction in the overall amount of source code,
because returning<> won't need to be specialized for every
combination of function and member function... though it will still
need a void specialization. We will still need a way to dispatch to
member functions through a regular function interface. mem_fn is
almost the right tool, but it only goes up to 8
arguments. Forwarding is tricky if you don't want to incur copies.
I think the trick is to use arg_from_python<T>::result_type for each
argument to the forwarder.
Another option would be to use separate function, function object,
and member function dispatchers. Once you know you have a member
function, you don't need cv-qualified overloads to call it.
Hmm, while we're at this, maybe we should solve the write-back
converter problem. Can we do it? Maybe not. Ralf doesn't want to
write special write-back functions here, does he? He wants the
converter to do the work automatically. We could add
cleanup/destructor registration. That would relieve the client from
having accessible destructors for types which are being converted by
rvalue. I'm not sure that this will really save any code,
however. It rather depends on the linker, doesn't it? I wonder if
this can be done in a backwards-compatible fashion by generating the
delete function when it's not supplied?

View File

@@ -1,223 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="boost.css">
<title>Boost.Python - Projects using Boost.Python</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="index.html">Boost.Python</a></h1>
<h2 align="center">Projects using Boost.Python</h2>
</td>
</tr>
</table>
<hr>
<h2>Introduction</h2>
<p>This is a partial list of projects using Boost.Python. If you are
using Boost.Python as your Python/C++ binding solution, we'd be proud to
list your project on this page. Just <a href=
"mailto:c++-sig@python.org">post</a> a short description of your project
and how Boost.Python helps you get the job done, and we'll add it to this
page .</p>
<hr>
<h3>Enterprise Software</h3>
<dl class="page-index">
<dt><b><a href="http://openwbem.sourceforge.net">OpenWBEM</a></b></dt>
<dd>
The OpenWBEM project is an effort to develop an open-source
implementation of Web Based Enterprise Management suitable for
commercial and non-commercial application
<p><a href="mailto:dnuffer@sco.com">Dan Nuffer</a> writes:</p>
<blockquote>
I'm using Boost.Python to wrap the client API of OpenWBEM.This will
make it easier to do rapid prototyping, testing, and scripting when
developing management solutions that use WBEM.
</blockquote>
</dd>
</dl>
<h3>Financial Analysis</h3>
<dl class="page-index">
<dt><b>TSLib</b> - <a href="http://www.fortressinv.com">Fortress
Investment Group LLC</a></dt>
<dd>
Fortress Investment Group has contracted <a href=
"http://www.boost-consulting.com">Boost Consulting</a> to develop
core internal financial analysis tools in C++ and to prepare Python
bindings for them using Boost.Python.
<p>Tom Barket of Fortress writes:</p>
<blockquote>
We have a large C++ analytical library specialized for research in
finance and economics, built for speed and mission critical
stability. Yet Python offers us the flexibility to test out new
ideas quickly and increase the productivity of our time versus
working in C++. There are several key features which make Python
stand out. Its elegance, stability, and breadth of resources on the
web are all valuable, but the most important is its extensibility,
due to its open source transparency. Boost.Python makes Python
extensibility extremely simple and straightforward, yet preserves a
great deal of power and control.
</blockquote>
</dd>
</dl>
<h3>Graphics</h3>
<dl class="page-index">
<dt><b><a href=
"http://www.openscenegraph.org">OpenSceneGraph</a></b></dt>
<dd><a href="mailto:gideon@computer.org">Gideon May</a> has created a
set of bindings for OpenSceneGraph, a cross-platform C++/OpenGL library
for the real-time visualization. You can read the release announcement
at <a href="http://www.hypereyes.com">www.hypereyes.com</a>. <a href=
"mailto:gideon@computer.org">Contact Gideon</a> for more
information.<br>
&nbsp;</dd>
<dt><a href=
"http://pythonmagick.procoders.net/"><b>PythonMagick</b></a></dt>
<dd>PythonMagick binds the <a href=
"http://www.imagemagick.org">ImageMagick</a> image manipulation library
to Python.<br>
&nbsp;</dd>
<dt><b>HippoDraw</b> - <a href="http://www.slac.stanford.edu">Stanford
Linear Accelerator Center</a></dt>
<dd>
HippoDraw is a data analysis environment consisting of a canvas upon
which graphs such as histograms, scattter plots, etc, are prsented.
It has a highly interactive GUI interface, but some things you need
to do with scripts. HippoDraw can be run as Python extension module
so that all the manipulation can be done from either Python or the
GUI.
<p><a href="mailto:Paul_Kunz@SLAC.Stanford.EDU">Paul F. Kunz</a>
writes:</p>
<blockquote>
Don't have a web page for the project, but the organization's is <a
href=
"http://www.slac.stanford.edu">http://www.slac.stanford.edu</a>
(the first web server site in America, I installed it).
</blockquote>
Which was just too cool a piece of trivia to omit.<br>
&nbsp;
</dd>
</dl>
<h3>Scientific Computing</h3>
<dl class="page index">
<dt><a href="http://camfr.sourceforge.net"><b>CAMFR</b></a></dt>
<dd>
CAMFR is a photonics and electromagnetics modelling tool. Python is
used for computational steering.
<p><a href="mailto:Peter.Bienstman@rug.ac.be">Peter Bienstman</a>
writes:</p>
<blockquote>
Thanks for providing such a great tool!
</blockquote>
</dd>
<dt><a href="http://cctbx.sourceforge.net"><b>cctbx - Computational
Crystallography Toolbox</b></a></dt>
<dd>
Computational Crystallography is concerned with the derivation of
atomic models of crystal structures, given experimental X-ray
diffraction data. The cctbx is an open-source library of fundamental
algorithms for crystallographic computations. The core algorithms are
implemented in C++ and accessed through higher-level Python
interfaces.
<p>The cctbx grew together with Boost.Python and is designed from the
ground up as a hybrid Python/C++ system. With one minor exception,
run-time polymorphism is completely handled by Python. C++
compile-time polymorphism is used to implement performance critical
algorithms. The Python and C++ layers are seamlessly integrated using
Boost.Python.</p>
<p>The SourceForge cctbx project is organized in modules to
facilitate use in non-crystallographic applications. The scitbx
module implements a general purpose array family for scientific
applications and pure C++ ports of FFTPACK and the LBFGS conjugate
gradient minimizer.</p>
</dd>
<dt><a href="http://www.llnl.gov/CASC/emsolve"><b>EMSolve</b></a></dt>
<dd>EMSolve is a provably stable, charge conserving, and energy
conserving solver for Maxwell's equations.<br>
&nbsp;</dd>
<dt><b><a href="http://cern.ch/gaudi">Gaudi</a></b> and <b><a href=
"http://cern.ch/Gaudi/RootPython/">RootPython</a></b></dt>
<dd>
Gaudi is a framework for particle physics collision data processing
applications developed in the context of the LHCb and ATLAS
experiments at CERN.
<p><a href="mailto:Pere.Mato@cern.ch">Pere Mato Vila</a> writes:</p>
<blockquote>
We are using Boost.Python to provide scripting/interactive
capability to our framework. We have a module called "GaudiPython"
implemented using Boost.Python that allows the interaction with any
framework service or algorithm from python. RootPython also uses
Boost.Python to provide a generic "gateway" between the <a href=
"http://root.cern.ch">ROOT</a> framework and python
<p>Boost.Python is great. We managed very quickly to interface our
framework to python, which is great language. We are trying to
facilitate to our physicists (end-users) a rapid analysis
application development environment based on python. For that,
Boost.Python plays and essential role.</p>
</blockquote>
</dd>
</dl>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
16 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,68 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="boost.css">
<title>Boost.Python - Support Resources</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="index.html">Boost.Python</a></h1>
<h2 align="center">Support Resources</h2>
</td>
</tr>
</table>
<hr>
<h2>Synopsis</h2>
<p>This is a list of available resources for support with Boost.Python
problems and feature requests.</p>
<hr>
<dl class="page-index">
<dt><b><a href="http://www.boost-consulting.com">Boost
Consulting</a></b> - Commercial support, development, training, and
distribution for all the Boost libraries, from the people who brought
you Boost.Python.<br>
&nbsp;</dt>
<dt><b><a href="http://www.python.org/sigs/c++-sig/">The Python
C++-sig</a></b> mailing list is a forum for discussing Python/C++
interoperability, and Boost.Python in particular.<br>
&nbsp;</dt>
<dt>The <b>Boost.Python <a href=
"http://www.python.org/cgi-bin/moinmoin/boost_2epython">Wiki
Pages</a></b> established by Mike Rovner as part of the <a href=
"http://www.python.org/cgi-bin/moinmoin">PythonInfo Wiki</a> serves as
a forum to gather peoples' experience and as a cookbook.<br>
&nbsp;</dt>
</dl>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
17 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,112 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Auto-Overloading</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="default_arguments.html">
<link rel="next" href="object_interface.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Auto-Overloading</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="default_arguments.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="object_interface.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
It was mentioned in passing in the previous section that
<tt>BOOST_PYTHON_FUNCTION_OVERLOADS</tt> and <tt>BOOST_PYTHON_FUNCTION_OVERLOADS</tt>
can also be used for overloaded functions and member functions with a
common sequence of initial arguments. Here is an example:</p>
<code><pre>
<span class=keyword>void </span><span class=identifier>foo</span><span class=special>()
{
/*...*/
}
</span><span class=keyword>void </span><span class=identifier>foo</span><span class=special>(</span><span class=keyword>bool </span><span class=identifier>a</span><span class=special>)
{
/*...*/
}
</span><span class=keyword>void </span><span class=identifier>foo</span><span class=special>(</span><span class=keyword>bool </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>int </span><span class=identifier>b</span><span class=special>)
{
/*...*/
}
</span><span class=keyword>void </span><span class=identifier>foo</span><span class=special>(</span><span class=keyword>bool </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>int </span><span class=identifier>b</span><span class=special>, </span><span class=keyword>char </span><span class=identifier>c</span><span class=special>)
{
/*...*/
}
</span></pre></code>
<p>
Like in the previous section, we can generate thin wrappers for these
overloaded functions in one-shot:</p>
<code><pre>
<span class=identifier>BOOST_PYTHON_FUNCTION_OVERLOADS</span><span class=special>(</span><span class=identifier>foo_overloads</span><span class=special>, </span><span class=identifier>foo</span><span class=special>, </span><span class=number>0</span><span class=special>, </span><span class=number>3</span><span class=special>)
</span></pre></code>
<p>
Then...</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;foo&quot;</span><span class=special>, </span><span class=identifier>foo</span><span class=special>, </span><span class=identifier>foo_overloads</span><span class=special>());
</span></pre></code>
<p>
Notice though that we have a situation now where we have a minimum of zero
(0) arguments and a maximum of 3 arguments.</p>
<a name="manual_wrapping"></a><h2>Manual Wrapping</h2><p>
It is important to emphasize however that <b>the overloaded functions must
have a common sequence of initial arguments</b>. Otherwise, our scheme above
will not work. If this is not the case, we have to wrap our functions
<a href="overloading.html">
manually</a>.</p>
<p>
Actually, we can mix and match manual wrapping of overloaded functions and
automatic wrapping through <tt>BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS</tt> and
its sister, <tt>BOOST_PYTHON_FUNCTION_OVERLOADS</tt>. Following up on our example
presented in the section <a href="overloading.html">
on overloading</a>, since the
first 4 overload functins have a common sequence of initial arguments, we
can use <tt>BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS</tt> to automatically wrap the
first three of the <tt>def</tt>s and manually wrap just the last. Here's
how we'll do this:</p>
<code><pre>
<span class=identifier>BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS</span><span class=special>(</span><span class=identifier>xf_overloads</span><span class=special>, </span><span class=identifier>f</span><span class=special>, </span><span class=number>1</span><span class=special>, </span><span class=number>4</span><span class=special>)
</span></pre></code>
<p>
Create a member function pointers as above for both X::f overloads:</p>
<code><pre>
<span class=keyword>bool </span><span class=special>(</span><span class=identifier>X</span><span class=special>::*</span><span class=identifier>fx1</span><span class=special>)(</span><span class=keyword>int</span><span class=special>, </span><span class=keyword>double</span><span class=special>, </span><span class=keyword>char</span><span class=special>) = &amp;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>f</span><span class=special>;
</span><span class=keyword>int </span><span class=special>(</span><span class=identifier>X</span><span class=special>::*</span><span class=identifier>fx2</span><span class=special>)(</span><span class=keyword>int</span><span class=special>, </span><span class=keyword>int</span><span class=special>, </span><span class=keyword>int</span><span class=special>) = &amp;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>f</span><span class=special>;
</span></pre></code>
<p>
Then...</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>fx1</span><span class=special>, </span><span class=identifier>xf_overloads</span><span class=special>());
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>fx2</span><span class=special>)
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="default_arguments.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="object_interface.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,77 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Basic Interface</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="object_interface.html">
<link rel="next" href="derived_object_types.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Basic Interface</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="object_interface.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="derived_object_types.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Class <tt>object</tt> wraps <tt>PyObject*</tt>. All the intricacies of dealing with
<tt>PyObject</tt>s such as managing reference counting are handled by the
<tt>object</tt> class. C++ object interoperability is seamless. Boost.Python C++
<tt>object</tt>s can in fact be explicitly constructed from any C++ object.</p>
<p>
To illustrate, this Python code snippet:</p>
<code><pre>
<span class=identifier>def </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>x</span><span class=special>, </span><span class=identifier>y</span><span class=special>):
</span><span class=keyword>if </span><span class=special>(</span><span class=identifier>y </span><span class=special>== </span><span class=literal>'foo'</span><span class=special>):
</span><span class=identifier>x</span><span class=special>[</span><span class=number>3</span><span class=special>:</span><span class=number>7</span><span class=special>] = </span><span class=literal>'bar'
</span><span class=keyword>else</span><span class=special>:
</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>items </span><span class=special>+= </span><span class=identifier>y</span><span class=special>(</span><span class=number>3</span><span class=special>, </span><span class=identifier>x</span><span class=special>)
</span><span class=keyword>return </span><span class=identifier>x
</span><span class=identifier>def </span><span class=identifier>getfunc</span><span class=special>():
</span><span class=keyword>return </span><span class=identifier>f</span><span class=special>;
</span></pre></code>
<p>
Can be rewritten in C++ using Boost.Python facilities this way:</p>
<code><pre>
<span class=identifier>object </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>object </span><span class=identifier>x</span><span class=special>, </span><span class=identifier>object </span><span class=identifier>y</span><span class=special>) {
</span><span class=keyword>if </span><span class=special>(</span><span class=identifier>y </span><span class=special>== </span><span class=string>&quot;foo&quot;</span><span class=special>)
</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>slice</span><span class=special>(</span><span class=number>3</span><span class=special>,</span><span class=number>7</span><span class=special>) = </span><span class=string>&quot;bar&quot;</span><span class=special>;
</span><span class=keyword>else
</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;items&quot;</span><span class=special>) += </span><span class=identifier>y</span><span class=special>(</span><span class=number>3</span><span class=special>, </span><span class=identifier>x</span><span class=special>);
</span><span class=keyword>return </span><span class=identifier>x</span><span class=special>;
}
</span><span class=identifier>object </span><span class=identifier>getfunc</span><span class=special>() {
</span><span class=keyword>return </span><span class=identifier>object</span><span class=special>(</span><span class=identifier>f</span><span class=special>);
}
</span></pre></code>
<p>
Apart from cosmetic differences due to the fact that we are writing the
code in C++, the look and feel should be immediately apparent to the Python
coder.</p>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="object_interface.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="derived_object_types.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,191 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Building Hello World</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="quickstart.html">
<link rel="next" href="exposing_classes.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Building Hello World</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="quickstart.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="exposing_classes.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<a name="from_start_to_finish"></a><h2>From Start To Finish</h2><p>
Now the first thing you'd want to do is to build the Hello World module and
try it for yourself in Python. In this section, we shall outline the steps
necessary to achieve that. We shall use the build tool that comes bundled
with every boost distribution: <b>bjam</b>.</p>
<table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/lens.gif"></img> <b>Building without bjam</b><br><br>
Besides bjam, there are of course other ways to get your module built.
What's written here should not be taken as &quot;the one and only way&quot;.
There are of course other build tools apart from <tt>bjam</tt>.
</td>
</tr>
</table>
<p>
We shall skip over the details. Our objective will be to simply create the
hello world module and run it in Python. For a complete reference to
building Boost.Python, check out: <a href="../../building.html">
building.html</a>.
After this brief <i>bjam</i> tutorial, we should have built two DLLs:</p>
<ul><li>boost_python.dll</li><li>hello.pyd</li></ul><p>
if you are on Windows, and</p>
<ul><li>libboost_python.so</li><li>hello.so</li></ul><p>
if you are on Unix.</p>
<p>
The tutorial example can be found in the directory:
<tt>libs/python/example/tutorial</tt>. There, you can find:</p>
<ul><li>hello.cpp</li><li>Jamfile</li></ul><p>
The <tt>hello.cpp</tt> file is our C++ hello world example. The <tt>Jamfile</tt> is a
minimalist <i>bjam</i> script that builds the DLLs for us.</p>
<p>
Before anything else, you should have the bjam executable in your boost
directory or somewhere in your path such that <tt>bjam</tt> can be executed in
the command line. Pre-built Boost.Jam executables are available for most
platforms. For example, a pre-built Microsoft Windows bjam executable can
be downloaded <a href="http://boost.sourceforge.net/jam-executables/bin.ntx86/bjam.zip">
here</a>.
The complete list of bjam pre-built
executables can be found <a href="../../../../../tools/build/index.html#Jam">
here</a>.</p>
<a name="lets_jam_"></a><h2>Lets Jam!</h2><p>
<img src="theme/jam.png"></img></p>
<p>
Here is our minimalist Jamfile:</p>
<code><pre>
subproject libs/python/example/tutorial ;
SEARCH on python.jam = $(BOOST_BUILD_PATH) ;
include python.jam ;
extension hello # Declare a Python extension called hello
: hello.cpp # source
&lt;dll&gt;../../build/boost_python # dependencies
;
</pre></code><p>
First, we need to specify our location in the boost project hierarchy.
It so happens that the tutorial example is located in <tt>/libs/python/example/tutorial</tt>.
Thus:</p>
<code><pre>
subproject libs/python/example/tutorial ;
</pre></code><p>
Then we will include the definitions needed by Python modules:</p>
<code><pre>
SEARCH on python.jam = $(BOOST_BUILD_PATH) ;
include python.jam ;
</pre></code><p>
Finally we declare our <tt>hello</tt> extension:</p>
<code><pre>
extension hello # Declare a Python extension called hello
: hello.cpp # source
&lt;dll&gt;../../build/boost_python # dependencies
;
</pre></code><a name="running_bjam"></a><h2>Running bjam</h2><p>
<i>bjam</i> is run using your operating system's command line interpreter.</p>
<blockquote><p>Start it up.</p></blockquote><p>
Make sure that the environment is set so that we can invoke the C++
compiler. With MSVC, that would mean running the <tt>Vcvars32.bat</tt> batch
file. For instance:</p>
<code><pre>
<span class=identifier>C</span><span class=special>:\</span><span class=identifier>Program </span><span class=identifier>Files</span><span class=special>\</span><span class=identifier>Microsoft </span><span class=identifier>Visual </span><span class=identifier>Studio</span><span class=special>\</span><span class=identifier>VC98</span><span class=special>\</span><span class=identifier>bin</span><span class=special>\</span><span class=identifier>Vcvars32</span><span class=special>.</span><span class=identifier>bat
</span></pre></code>
<p>
Some environment variables will have to be setup for proper building of our
Python modules. Example:</p>
<code><pre>
<span class=identifier>set </span><span class=identifier>PYTHON_ROOT</span><span class=special>=</span><span class=identifier>c</span><span class=special>:/</span><span class=identifier>dev</span><span class=special>/</span><span class=identifier>tools</span><span class=special>/</span><span class=identifier>python
</span><span class=identifier>set </span><span class=identifier>PYTHON_VERSION</span><span class=special>=</span><span class=number>2.2
</span></pre></code>
<p>
The above assumes that the Python installation is in <tt>c:/dev/tools/python</tt>
and that we are using Python version 2.2. You'll have to tweak this path
appropriately. <img src="theme/note.gif"></img> Be sure not to include a third number, e.g. <b>not</b> &quot;2.2.1&quot;,
even if that's the version you have.</p>
<p>
Now we are ready... Be sure to <tt>cd</tt> to <tt>libs/python/example/tutorial</tt>
where the tutorial <tt>&quot;hello.cpp&quot;</tt> and the <tt>&quot;Jamfile&quot;</tt> is situated.</p>
<p>
Finally:</p>
<code><pre>
<span class=identifier>bjam </span><span class=special>-</span><span class=identifier>sTOOLS</span><span class=special>=</span><span class=identifier>msvc
</span></pre></code>
<p>
We are again assuming that we are using Microsoft Visual C++ version 6. If
not, then you will have to specify the appropriate tool. See
<a href="../../../../../tools/build/index.html">
Building Boost Libraries</a> for
further details.</p>
<p>
It should be building now:</p>
<code><pre>
cd C:\dev\boost\libs\python\example\tutorial
bjam -sTOOLS=msvc
...patience...
...found 1703 targets...
...updating 40 targets...
</pre></code><p>
And so on... Finally:</p>
<code><pre>
vc-C++ ..\..\..\..\libs\python\example\tutorial\bin\hello.pyd\msvc\debug\
runtime-link-dynamic\hello.obj
hello.cpp
vc-Link ..\..\..\..\libs\python\example\tutorial\bin\hello.pyd\msvc\debug\
runtime-link-dynamic\hello.pyd ..\..\..\..\libs\python\example\tutorial\bin\
hello.pyd\msvc\debug\runtime-link-dynamic\hello.lib
Creating library ..\..\..\..\libs\python\example\tutorial\bin\hello.pyd\
msvc\debug\runtime-link-dynamic\hello.lib and object ..\..\..\..\libs\python\
example\tutorial\bin\hello.pyd\msvc\debug\runtime-link-dynamic\hello.exp
...updated 40 targets...
</pre></code><p>
If all is well, you should now have:</p>
<ul><li>boost_python.dll</li><li>hello.pyd</li></ul><p>
if you are on Windows, and</p>
<ul><li>libboost_python.so</li><li>hello.so</li></ul><p>
if you are on Unix.</p>
<p>
<tt>boost_python.dll</tt> can be found somewhere in <tt>libs\python\build\bin</tt>
while <tt>hello.pyd</tt> can be found somewhere in
<tt>libs\python\example\tutorial\bin</tt>. After a successful build, you can just
link in these DLLs with the Python interpreter. In Windows for example, you
can simply put these libraries inside the directory where the Python
executable is.</p>
<p>
You may now fire up Python and run our hello module:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>import </span><span class=identifier>hello
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>print </span><span class=identifier>hello</span><span class=special>.</span><span class=identifier>greet</span><span class=special>()
</span><span class=identifier>hello</span><span class=special>, </span><span class=identifier>world
</span></pre></code>
<blockquote><p><b>There you go... Have fun!</b></p></blockquote><table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="quickstart.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="exposing_classes.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,169 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Call Policies</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="functions.html">
<link rel="next" href="overloading.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Call Policies</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="functions.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="overloading.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
In C++, we often deal with arguments and return types such as pointers
and references. Such primitive types are rather, ummmm, low level and
they really don't tell us much. At the very least, we don't know the
owner of the pointer or the referenced object. No wonder languages
such as Java and Python never deal with such low level entities. In
C++, it's usually considered a good practice to use smart pointers
which exactly describe ownership semantics. Still, even good C++
interfaces use raw references and pointers sometimes, so Boost.Python
must deal with them. To do this, it may need your help. Consider the
following C++ function:</p>
<code><pre>
<span class=identifier>X</span><span class=special>&amp; </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>Y</span><span class=special>&amp; </span><span class=identifier>y</span><span class=special>, </span><span class=identifier>Z</span><span class=special>* </span><span class=identifier>z</span><span class=special>);
</span></pre></code>
<p>
How should the library wrap this function? A naive approach builds a
Python X object around result reference. This strategy might or might
not work out. Here's an example where it didn't</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>x </span><span class=special>= </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>y</span><span class=special>, </span><span class=identifier>z</span><span class=special>) </span>##<span class=identifier>x </span><span class=identifier>refers </span><span class=identifier>to </span><span class=identifier>some </span><span class=identifier>C</span><span class=special>++ </span><span class=identifier>X
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>del </span><span class=identifier>y
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>some_method</span><span class=special>() </span>##<span class=identifier>CRASH</span><span class=special>!
</span></pre></code>
<p>
What's the problem?</p>
<p>
Well, what if f() was implemented as shown below:</p>
<code><pre>
<span class=identifier>X</span><span class=special>&amp; </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>Y</span><span class=special>&amp; </span><span class=identifier>y</span><span class=special>, </span><span class=identifier>Z</span><span class=special>* </span><span class=identifier>z</span><span class=special>)
{
</span><span class=identifier>y</span><span class=special>.</span><span class=identifier>z </span><span class=special>= </span><span class=identifier>z</span><span class=special>;
</span><span class=keyword>return </span><span class=identifier>y</span><span class=special>.</span><span class=identifier>x</span><span class=special>;
}
</span></pre></code>
<p>
The problem is that the lifetime of result X&amp; is tied to the lifetime
of y, because the f() returns a reference to a member of the y
object. This idiom is is not uncommon and perfectly acceptable in the
context of C++. However, Python users should not be able to crash the
system just by using our C++ interface. In this case deleting y will
invalidate the reference to X. We have a dangling reference.</p>
<p>
Here's what's happening:</p>
<ol><li><tt>f</tt> is called passing in a reference to <tt>y</tt> and a pointer to <tt>z</tt></li><li>A reference to <tt>y.x</tt> is returned</li><li><tt>y</tt> is deleted. <tt>x</tt> is a dangling reference</li><li><tt>x.some_method()</tt> is called</li><li><b>BOOM!</b></li></ol><p>
We could copy result into a new object:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>y</span><span class=special>, </span><span class=identifier>z</span><span class=special>).</span><span class=identifier>set</span><span class=special>(</span><span class=number>42</span><span class=special>) </span>##<span class=identifier>Result </span><span class=identifier>disappears
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>y</span><span class=special>.</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>get</span><span class=special>() </span>##<span class=identifier>No </span><span class=identifier>crash</span><span class=special>, </span><span class=identifier>but </span><span class=identifier>still </span><span class=identifier>bad
</span><span class=number>3.14
</span></pre></code>
<p>
This is not really our intent of our C++ interface. We've broken our
promise that the Python interface should reflect the C++ interface as
closely as possible.</p>
<p>
Our problems do not end there. Suppose Y is implemented as follows:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Y
</span><span class=special>{
</span><span class=identifier>X </span><span class=identifier>x</span><span class=special>; </span><span class=identifier>Z</span><span class=special>* </span><span class=identifier>z</span><span class=special>;
</span><span class=keyword>int </span><span class=identifier>z_value</span><span class=special>() { </span><span class=keyword>return </span><span class=identifier>z</span><span class=special>-&gt;</span><span class=identifier>value</span><span class=special>(); }
};
</span></pre></code>
<p>
Notice that the data member <tt>z</tt> is held by class Y using a raw
pointer. Now we have a potential dangling pointer problem inside Y:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>x </span><span class=special>= </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>y</span><span class=special>, </span><span class=identifier>z</span><span class=special>) </span>##<span class=identifier>y </span><span class=identifier>refers </span><span class=identifier>to </span><span class=identifier>z
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>del </span><span class=identifier>z </span>##<span class=identifier>Kill </span><span class=identifier>the </span><span class=identifier>z </span><span class=identifier>object
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>y</span><span class=special>.</span><span class=identifier>z_value</span><span class=special>() </span>##<span class=identifier>CRASH</span><span class=special>!
</span></pre></code>
<p>
For reference, here's the implementation of <tt>f</tt> again:</p>
<code><pre>
<span class=identifier>X</span><span class=special>&amp; </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>Y</span><span class=special>&amp; </span><span class=identifier>y</span><span class=special>, </span><span class=identifier>Z</span><span class=special>* </span><span class=identifier>z</span><span class=special>)
{
</span><span class=identifier>y</span><span class=special>.</span><span class=identifier>z </span><span class=special>= </span><span class=identifier>z</span><span class=special>;
</span><span class=keyword>return </span><span class=identifier>y</span><span class=special>.</span><span class=identifier>x</span><span class=special>;
}
</span></pre></code>
<p>
Here's what's happening:</p>
<ol><li><tt>f</tt> is called passing in a reference to <tt>y</tt> and a pointer to <tt>z</tt></li><li>A pointer to <tt>z</tt> is held by <tt>y</tt></li><li>A reference to <tt>y.x</tt> is returned</li><li><tt>z</tt> is deleted. <tt>y.z</tt> is a dangling pointer</li><li><tt>y.z_value()</tt> is called</li><li><tt>z-&gt;value()</tt> is called</li><li><b>BOOM!</b></li></ol><a name="call_policies"></a><h2>Call Policies</h2><p>
Call Policies may be used in situations such as the example detailed above.
In our example, <tt>return_internal_reference</tt> and <tt>with_custodian_and_ward</tt>
are our friends:</p>
<code><pre>
<span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>f</span><span class=special>,
</span><span class=identifier>return_internal_reference</span><span class=special>&lt;</span><span class=number>1</span><span class=special>,
</span><span class=identifier>with_custodian_and_ward</span><span class=special>&lt;</span><span class=number>1</span><span class=special>, </span><span class=number>2</span><span class=special>&gt; &gt;());
</span></pre></code>
<p>
What are the <tt>1</tt> and <tt>2</tt> parameters, you ask?</p>
<code><pre>
<span class=identifier>return_internal_reference</span><span class=special>&lt;</span><span class=number>1
</span></pre></code>
<p>
Informs Boost.Python that the first argument, in our case <tt>Y&amp; y</tt>, is the
owner of the returned reference: <tt>X&amp;</tt>. The &quot;<tt>1</tt>&quot; simply specifies the
first argument. In short: &quot;return an internal reference <tt>X&amp;</tt> owned by the
1st argument <tt>Y&amp; y</tt>&quot;.</p>
<code><pre>
<span class=identifier>with_custodian_and_ward</span><span class=special>&lt;</span><span class=number>1</span><span class=special>, </span><span class=number>2</span><span class=special>&gt;
</span></pre></code>
<p>
Informs Boost.Python that the lifetime of the argument indicated by ward
(i.e. the 2nd argument: <tt>Z* z</tt>) is dependent on the lifetime of the
argument indicated by custodian (i.e. the 1st argument: <tt>Y&amp; y</tt>).</p>
<p>
It is also important to note that we have defined two policies above. Two
or more policies can be composed by chaining. Here's the general syntax:</p>
<code><pre>
<span class=identifier>policy1</span><span class=special>&lt;</span><span class=identifier>args</span><span class=special>...,
</span><span class=identifier>policy2</span><span class=special>&lt;</span><span class=identifier>args</span><span class=special>...,
</span><span class=identifier>policy3</span><span class=special>&lt;</span><span class=identifier>args</span><span class=special>...&gt; &gt; &gt;
</span></pre></code>
<p>
Here is the list of predefined call policies. A complete reference detailing
these can be found <a href="../../v2/reference.html#models_of_call_policies">
here</a>.</p>
<ul><li><b>with_custodian_and_ward</b><br> Ties lifetimes of the arguments</li><li><b>with_custodian_and_ward_postcall</b><br> Ties lifetimes of the arguments and results</li><li><b>return_internal_reference</b><br> Ties lifetime of one argument to that of result</li><li><b>return_value_policy&lt;T&gt; with T one of:</b><br></li><li><b>reference_existing_object</b><br>naïve (dangerous) approach</li><li><b>copy_const_reference</b><br>Boost.Python v1 approach</li><li><b>copy_non_const_reference</b><br></li><li><b>manage_new_object</b><br> Adopt a pointer and hold the instance</li></ul><table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/smiley.gif"></img> <b>Remember the Zen, Luke:</b><br><br>
&quot;Explicit is better than implicit&quot;<br>
&quot;In the face of ambiguity, refuse the temptation to guess&quot;<br> </td>
</tr>
</table>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="functions.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="overloading.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,78 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Class Data Members</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="constructors.html">
<link rel="next" href="class_properties.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Class Data Members</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="constructors.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_properties.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Data members may also be exposed to Python so that they can be
accessed as attributes of the corresponding Python class. Each data
member that we wish to be exposed may be regarded as <b>read-only</b> or
<b>read-write</b>. Consider this class <tt>Var</tt>:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Var
</span><span class=special>{
</span><span class=identifier>Var</span><span class=special>(</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>name</span><span class=special>) : </span><span class=identifier>name</span><span class=special>(</span><span class=identifier>name</span><span class=special>), </span><span class=identifier>value</span><span class=special>() {}
</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=keyword>const </span><span class=identifier>name</span><span class=special>;
</span><span class=keyword>float </span><span class=identifier>value</span><span class=special>;
};
</span></pre></code>
<p>
Our C++ <tt>Var</tt> class and its data members can be exposed to Python:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Var</span><span class=special>&gt;(</span><span class=string>&quot;Var&quot;</span><span class=special>, </span><span class=identifier>init</span><span class=special>&lt;</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string</span><span class=special>&gt;())
.</span><span class=identifier>def_readonly</span><span class=special>(</span><span class=string>&quot;name&quot;</span><span class=special>, &amp;</span><span class=identifier>Var</span><span class=special>::</span><span class=identifier>name</span><span class=special>)
.</span><span class=identifier>def_readwrite</span><span class=special>(</span><span class=string>&quot;value&quot;</span><span class=special>, &amp;</span><span class=identifier>Var</span><span class=special>::</span><span class=identifier>value</span><span class=special>);
</span></pre></code>
<p>
Then, in Python, assuming we have placed our Var class inside the namespace
hello as we did before:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>x </span><span class=special>= </span><span class=identifier>hello</span><span class=special>.</span><span class=identifier>Var</span><span class=special>(</span><span class=literal>'pi'</span><span class=special>)
&gt;&gt;&gt; </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>value </span><span class=special>= </span><span class=number>3.14
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>print </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>name</span><span class=special>, </span><span class=literal>'is around'</span><span class=special>, </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>value
</span><span class=identifier>pi </span><span class=identifier>is </span><span class=identifier>around </span><span class=number>3.14
</span></pre></code>
<p>
Note that <tt>name</tt> is exposed as <b>read-only</b> while <tt>value</tt> is exposed
as <b>read-write</b>.</p>
<code><pre>
&gt;&gt;&gt; x.name = 'e' # can't change name
Traceback (most recent call last):
File &quot;&lt;stdin&gt;&quot;, line 1, in ?
AttributeError: can't set attribute
</pre></code><table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="constructors.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_properties.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,109 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Class Operators/Special Functions</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="virtual_functions_with_default_implementations.html">
<link rel="next" href="functions.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Class Operators/Special Functions</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="virtual_functions_with_default_implementations.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="functions.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<a name="python_operators"></a><h2>Python Operators</h2><p>
C is well known for the abundance of operators. C++ extends this to the
extremes by allowing operator overloading. Boost.Python takes advantage of
this and makes it easy to wrap C++ operator-powered classes.</p>
<p>
Consider a file position class <tt>FilePos</tt> and a set of operators that take
on FilePos instances:</p>
<code><pre>
<span class=keyword>class </span><span class=identifier>FilePos </span><span class=special>{ /*...*/ };
</span><span class=identifier>FilePos </span><span class=keyword>operator</span><span class=special>+(</span><span class=identifier>FilePos</span><span class=special>, </span><span class=keyword>int</span><span class=special>);
</span><span class=identifier>FilePos </span><span class=keyword>operator</span><span class=special>+(</span><span class=keyword>int</span><span class=special>, </span><span class=identifier>FilePos</span><span class=special>);
</span><span class=keyword>int </span><span class=keyword>operator</span><span class=special>-(</span><span class=identifier>FilePos</span><span class=special>, </span><span class=identifier>FilePos</span><span class=special>);
</span><span class=identifier>FilePos </span><span class=keyword>operator</span><span class=special>-(</span><span class=identifier>FilePos</span><span class=special>, </span><span class=keyword>int</span><span class=special>);
</span><span class=identifier>FilePos</span><span class=special>&amp; </span><span class=keyword>operator</span><span class=special>+=(</span><span class=identifier>FilePos</span><span class=special>&amp;, </span><span class=keyword>int</span><span class=special>);
</span><span class=identifier>FilePos</span><span class=special>&amp; </span><span class=keyword>operator</span><span class=special>-=(</span><span class=identifier>FilePos</span><span class=special>&amp;, </span><span class=keyword>int</span><span class=special>);
</span><span class=keyword>bool </span><span class=keyword>operator</span><span class=special>&lt;(</span><span class=identifier>FilePos</span><span class=special>, </span><span class=identifier>FilePos</span><span class=special>);
</span></pre></code>
<p>
The class and the various operators can be mapped to Python rather easily
and intuitively:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>FilePos</span><span class=special>&gt;(</span><span class=string>&quot;FilePos&quot;</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>self </span><span class=special>+ </span><span class=keyword>int</span><span class=special>()) // </span><span class=identifier>__add__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=keyword>int</span><span class=special>() + </span><span class=identifier>self</span><span class=special>) // </span><span class=identifier>__radd__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>self </span><span class=special>- </span><span class=identifier>self</span><span class=special>) // </span><span class=identifier>__sub__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>self </span><span class=special>- </span><span class=keyword>int</span><span class=special>()) // </span><span class=identifier>__sub__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>self </span><span class=special>+= </span><span class=keyword>int</span><span class=special>()) // </span><span class=identifier>__iadd__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>self </span><span class=special>-= </span><span class=identifier>other</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt;())
.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>self </span><span class=special>&lt; </span><span class=identifier>self</span><span class=special>); // </span><span class=identifier>__lt__
</span></pre></code>
<p>
The code snippet above is very clear and needs almost no explanation at
all. It is virtually the same as the operators' signatures. Just take
note that <tt>self</tt> refers to FilePos object. Also, not every class <tt>T</tt> that
you might need to interact with in an operator expression is (cheaply)
default-constructible. You can use <tt>other&lt;T&gt;()</tt> in place of an actual
<tt>T</tt> instance when writing &quot;self expressions&quot;.</p>
<a name="special_methods"></a><h2>Special Methods</h2><p>
Python has a few more <i>Special Methods</i>. Boost.Python supports all of the
standard special method names supported by real Python class instances. A
similar set of intuitive interfaces can also be used to wrap C++ functions
that correspond to these Python <i>special functions</i>. Example:</p>
<code><pre>
<span class=keyword>class </span><span class=identifier>Rational
</span><span class=special>{ </span><span class=keyword>operator </span><span class=keyword>double</span><span class=special>() </span><span class=keyword>const</span><span class=special>; };
</span><span class=identifier>Rational </span><span class=identifier>pow</span><span class=special>(</span><span class=identifier>Rational</span><span class=special>, </span><span class=identifier>Rational</span><span class=special>);
</span><span class=identifier>Rational </span><span class=identifier>abs</span><span class=special>(</span><span class=identifier>Rational</span><span class=special>);
</span><span class=identifier>ostream</span><span class=special>&amp; </span><span class=keyword>operator</span><span class=special>&lt;&lt;(</span><span class=identifier>ostream</span><span class=special>&amp;,</span><span class=identifier>Rational</span><span class=special>);
</span><span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Rational</span><span class=special>&gt;()
.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>float_</span><span class=special>(</span><span class=identifier>self</span><span class=special>)) // </span><span class=identifier>__float__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>pow</span><span class=special>(</span><span class=identifier>self</span><span class=special>, </span><span class=identifier>other</span><span class=special>&lt;</span><span class=identifier>Rational</span><span class=special>&gt;)) // </span><span class=identifier>__pow__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>abs</span><span class=special>(</span><span class=identifier>self</span><span class=special>)) // </span><span class=identifier>__abs__
</span><span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>str</span><span class=special>(</span><span class=identifier>self</span><span class=special>)) // </span><span class=identifier>__str__
</span><span class=special>;
</span></pre></code>
<p>
Need we say more?</p>
<table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/lens.gif"></img> What is the business of <tt>operator&lt;&lt;</tt> <tt>.def(str(self))</tt>?
Well, the method <tt>str</tt> requires the <tt>operator&lt;&lt;</tt> to do its work (i.e.
<tt>operator&lt;&lt;</tt> is used by the method defined by def(str(self)). </td>
</tr>
</table>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="virtual_functions_with_default_implementations.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="functions.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,81 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Class Properties</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="class_data_members.html">
<link rel="next" href="inheritance.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Class Properties</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_data_members.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="inheritance.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
In C++, classes with public data members are usually frowned
upon. Well designed classes that take advantage of encapsulation hide
the class' data members. The only way to access the class' data is
through access (getter/setter) functions. Access functions expose class
properties. Here's an example:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Num
</span><span class=special>{
</span><span class=identifier>Num</span><span class=special>();
</span><span class=keyword>float </span><span class=identifier>get</span><span class=special>() </span><span class=keyword>const</span><span class=special>;
</span><span class=keyword>void </span><span class=identifier>set</span><span class=special>(</span><span class=keyword>float </span><span class=identifier>value</span><span class=special>);
...
};
</span></pre></code>
<p>
However, in Python attribute access is fine; it doesn't neccessarily break
encapsulation to let users handle attributes directly, because the
attributes can just be a different syntax for a method call. Wrapping our
<tt>Num</tt> class using Boost.Python:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Num</span><span class=special>&gt;(</span><span class=string>&quot;Num&quot;</span><span class=special>)
.</span><span class=identifier>add_property</span><span class=special>(</span><span class=string>&quot;rovalue&quot;</span><span class=special>, &amp;</span><span class=identifier>Num</span><span class=special>::</span><span class=identifier>get</span><span class=special>)
.</span><span class=identifier>add_property</span><span class=special>(</span><span class=string>&quot;value&quot;</span><span class=special>, &amp;</span><span class=identifier>Num</span><span class=special>::</span><span class=identifier>get</span><span class=special>, &amp;</span><span class=identifier>Num</span><span class=special>::</span><span class=identifier>set</span><span class=special>);
</span></pre></code>
<p>
And at last, in Python:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>x </span><span class=special>= </span><span class=identifier>Num</span><span class=special>()
&gt;&gt;&gt; </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>value </span><span class=special>= </span><span class=number>3.14
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>value</span><span class=special>, </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>rovalue
</span><span class=special>(</span><span class=number>3.14</span><span class=special>, </span><span class=number>3.14</span><span class=special>)
&gt;&gt;&gt; </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>rovalue </span><span class=special>= </span><span class=number>2.17 </span>##<span class=identifier>error</span><span class=special>!
</span></pre></code>
<p>
Take note that the class property <tt>rovalue</tt> is exposed as <b>read-only</b>
since the <tt>rovalue</tt> setter member function is not passed in:</p>
<code><pre>
<span class=special>.</span><span class=identifier>add_property</span><span class=special>(</span><span class=string>&quot;rovalue&quot;</span><span class=special>, &amp;</span><span class=identifier>Num</span><span class=special>::</span><span class=identifier>get</span><span class=special>)
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_data_members.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="inheritance.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,129 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Class Virtual Functions</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="inheritance.html">
<link rel="next" href="deriving_a_python_class.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Class Virtual Functions</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="inheritance.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="deriving_a_python_class.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
In this section, we shall learn how to make functions behave
polymorphically through virtual functions. Continuing our example, let us
add a virtual function to our <tt>Base</tt> class:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Base
</span><span class=special>{
</span><span class=keyword>virtual </span><span class=keyword>int </span><span class=identifier>f</span><span class=special>() = </span><span class=number>0</span><span class=special>;
};
</span></pre></code>
<p>
Since <tt>f</tt> is a pure virtual function, <tt>Base</tt> is now an abstract
class. Given an instance of our class, the free function <tt>call_f</tt>
calls some implementation of this virtual function in a concrete
derived class:</p>
<code><pre>
<span class=keyword>int </span><span class=identifier>call_f</span><span class=special>(</span><span class=identifier>Base</span><span class=special>&amp; </span><span class=identifier>b</span><span class=special>) { </span><span class=keyword>return </span><span class=identifier>b</span><span class=special>.</span><span class=identifier>f</span><span class=special>(); }
</span></pre></code>
<p>
To allow this function to be implemented in a Python derived class, we
need to create a class wrapper:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>BaseWrap </span><span class=special>: </span><span class=identifier>Base
</span><span class=special>{
</span><span class=identifier>BaseWrap</span><span class=special>(</span><span class=identifier>PyObject</span><span class=special>* </span><span class=identifier>self_</span><span class=special>)
: </span><span class=identifier>self</span><span class=special>(</span><span class=identifier>self_</span><span class=special>) {}
</span><span class=keyword>int </span><span class=identifier>f</span><span class=special>() { </span><span class=keyword>return </span><span class=identifier>call_method</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt;(</span><span class=identifier>self</span><span class=special>, </span><span class=string>&quot;f&quot;</span><span class=special>); }
</span><span class=identifier>PyObject</span><span class=special>* </span><span class=identifier>self</span><span class=special>;
};
</span></pre></code>
<table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/lens.gif"></img> <b>member function and methods</b><br><br> Python, like
many object oriented languages uses the term <b>methods</b>. Methods
correspond roughly to C++'s <b>member functions</b> </td>
</tr>
</table>
<p>
Our class wrapper <tt>BaseWrap</tt> is derived from <tt>Base</tt>. Its overridden
virtual member function <tt>f</tt> in effect calls the corresponding method
of the Python object <tt>self</tt>, which is a pointer back to the Python
<tt>Base</tt> object holding our <tt>BaseWrap</tt> instance.</p>
<table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/note.gif"></img> <b>Why do we need BaseWrap?</b><br><br>
<i>You may ask</i>, &quot;Why do we need the <tt>BaseWrap</tt> derived class? This could
have been designed so that everything gets done right inside of
Base.&quot;<br><br>
One of the goals of Boost.Python is to be minimally intrusive on an
existing C++ design. In principle, it should be possible to expose the
interface for a 3rd party library without changing it. To unintrusively
hook into the virtual functions so that a Python override may be called, we
must use a derived class.<br><br>
Note however that you don't need to do this to get methods overridden
in Python to behave virtually when called <i>from</i> <b>Python</b>. The only
time you need to do the <tt>BaseWrap</tt> dance is when you have a virtual
function that's going to be overridden in Python and called
polymorphically <i>from</i> <b>C++</b>. </td>
</tr>
</table>
<p>
Wrapping <tt>Base</tt> and the free function <tt>call_f</tt>:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Base</span><span class=special>, </span><span class=identifier>BaseWrap</span><span class=special>, </span><span class=identifier>boost</span><span class=special>::</span><span class=identifier>noncopyable</span><span class=special>&gt;(</span><span class=string>&quot;Base&quot;</span><span class=special>, </span><span class=identifier>no_init</span><span class=special>)
;
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;call_f&quot;</span><span class=special>, </span><span class=identifier>call_f</span><span class=special>);
</span></pre></code>
<p>
Notice that we parameterized the <tt>class_</tt> template with <tt>BaseWrap</tt> as the
second parameter. What is <tt>noncopyable</tt>? Without it, the library will try
to create code for converting Base return values of wrapped functions to
Python. To do that, it needs Base's copy constructor... which isn't
available, since Base is an abstract class.</p>
<p>
In Python, let us try to instantiate our <tt>Base</tt> class:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>base </span><span class=special>= </span><span class=identifier>Base</span><span class=special>()
</span><span class=identifier>RuntimeError</span><span class=special>: </span><span class=identifier>This </span><span class=keyword>class </span><span class=identifier>cannot </span><span class=identifier>be </span><span class=identifier>instantiated </span><span class=identifier>from </span><span class=identifier>Python
</span></pre></code>
<p>
Why is it an error? <tt>Base</tt> is an abstract class. As such it is advisable
to define the Python wrapper with <tt>no_init</tt> as we have done above. Doing
so will disallow abstract base classes such as <tt>Base</tt> to be instantiated.</p>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="inheritance.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="deriving_a_python_class.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,102 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Constructors</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="exposing_classes.html">
<link rel="next" href="class_data_members.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Constructors</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="exposing_classes.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_data_members.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Our previous example didn't have any explicit constructors.
Since <tt>World</tt> is declared as a plain struct, it has an implicit default
constructor. Boost.Python exposes the default constructor by default,
which is why we were able to write</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>planet </span><span class=special>= </span><span class=identifier>hello</span><span class=special>.</span><span class=identifier>World</span><span class=special>()
</span></pre></code>
<p>
We may wish to wrap a class with a non-default constructor. Let us
build on our previous example:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>World
</span><span class=special>{
</span><span class=identifier>World</span><span class=special>(</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>msg</span><span class=special>): </span><span class=identifier>msg</span><span class=special>(</span><span class=identifier>msg</span><span class=special>) {} // </span><span class=identifier>added </span><span class=identifier>constructor
</span><span class=keyword>void </span><span class=identifier>set</span><span class=special>(</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>msg</span><span class=special>) { </span><span class=keyword>this</span><span class=special>-&gt;</span><span class=identifier>msg </span><span class=special>= </span><span class=identifier>msg</span><span class=special>; }
</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>greet</span><span class=special>() { </span><span class=keyword>return </span><span class=identifier>msg</span><span class=special>; }
</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>msg</span><span class=special>;
};
</span></pre></code>
<p>
This time <tt>World</tt> has no default constructor; our previous
wrapping code would fail to compile when the library tried to expose
it. We have to tell <tt>class_&lt;World&gt;</tt> about the constructor we want to
expose instead.</p>
<code><pre>
<span class=preprocessor>#include </span><span class=special>&lt;</span><span class=identifier>boost</span><span class=special>/</span><span class=identifier>python</span><span class=special>.</span><span class=identifier>hpp</span><span class=special>&gt;
</span><span class=keyword>using </span><span class=keyword>namespace </span><span class=identifier>boost</span><span class=special>::</span><span class=identifier>python</span><span class=special>;
</span><span class=identifier>BOOST_PYTHON_MODULE</span><span class=special>(</span><span class=identifier>hello</span><span class=special>)
{
</span><span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>World</span><span class=special>&gt;(</span><span class=string>&quot;World&quot;</span><span class=special>, </span><span class=identifier>init</span><span class=special>&lt;</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string</span><span class=special>&gt;())
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;greet&quot;</span><span class=special>, &amp;</span><span class=identifier>World</span><span class=special>::</span><span class=identifier>greet</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;set&quot;</span><span class=special>, &amp;</span><span class=identifier>World</span><span class=special>::</span><span class=identifier>set</span><span class=special>)
;
}
</span></pre></code>
<p>
<tt>init&lt;std::string&gt;()</tt> exposes the constructor taking in a
<tt>std::string</tt> (in Python, constructors are spelled
&quot;<tt>&quot;__init__&quot;</tt>&quot;).</p>
<p>
We can expose additional constructors by passing more <tt>init&lt;...&gt;</tt>s to
the <tt>def()</tt> member function. Say for example we have another World
constructor taking in two doubles:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>World</span><span class=special>&gt;(</span><span class=string>&quot;World&quot;</span><span class=special>, </span><span class=identifier>init</span><span class=special>&lt;</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string</span><span class=special>&gt;())
.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>init</span><span class=special>&lt;</span><span class=keyword>double</span><span class=special>, </span><span class=keyword>double</span><span class=special>&gt;())
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;greet&quot;</span><span class=special>, &amp;</span><span class=identifier>World</span><span class=special>::</span><span class=identifier>greet</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;set&quot;</span><span class=special>, &amp;</span><span class=identifier>World</span><span class=special>::</span><span class=identifier>set</span><span class=special>)
;
</span></pre></code>
<p>
On the other hand, if we do not wish to expose any constructors at
all, we may use <tt>no_init</tt> instead:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Abstract</span><span class=special>&gt;(</span><span class=string>&quot;Abstract&quot;</span><span class=special>, </span><span class=identifier>no_init</span><span class=special>)
</span></pre></code>
<p>
This actually adds an <tt>__init__</tt> method which always raises a
Python RuntimeError exception.</p>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="exposing_classes.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_data_members.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,158 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Default Arguments</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="overloading.html">
<link rel="next" href="auto_overloading.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Default Arguments</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="overloading.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="auto_overloading.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Boost.Python wraps (member) function pointers. Unfortunately, C++ function
pointers carry no default argument info. Take a function <tt>f</tt> with default
arguments:</p>
<code><pre>
<span class=keyword>int </span><span class=identifier>f</span><span class=special>(</span><span class=keyword>int</span><span class=special>, </span><span class=keyword>double </span><span class=special>= </span><span class=number>3.14</span><span class=special>, </span><span class=keyword>char </span><span class=keyword>const</span><span class=special>* = </span><span class=string>&quot;hello&quot;</span><span class=special>);
</span></pre></code>
<p>
But the type of a pointer to the function <tt>f</tt> has no information
about its default arguments:</p>
<code><pre>
<span class=keyword>int</span><span class=special>(*</span><span class=identifier>g</span><span class=special>)(</span><span class=keyword>int</span><span class=special>,</span><span class=keyword>double</span><span class=special>,</span><span class=keyword>char </span><span class=keyword>const</span><span class=special>*) = </span><span class=identifier>f</span><span class=special>; // </span><span class=identifier>defaults </span><span class=identifier>lost</span><span class=special>!
</span></pre></code>
<p>
When we pass this function pointer to the <tt>def</tt> function, there is no way
to retrieve the default arguments:</p>
<code><pre>
<span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>f</span><span class=special>); // </span><span class=identifier>defaults </span><span class=identifier>lost</span><span class=special>!
</span></pre></code>
<p>
Because of this, when wrapping C++ code, we had to resort to manual
wrapping as outlined in the <a href="overloading.html">
previous section</a>, or
writing thin wrappers:</p>
<code><pre>
<span class=comment>// write &quot;thin wrappers&quot;
</span><span class=keyword>int </span><span class=identifier>f1</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>x</span><span class=special>) { </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>x</span><span class=special>); }
</span><span class=keyword>int </span><span class=identifier>f2</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>x</span><span class=special>, </span><span class=keyword>double </span><span class=identifier>y</span><span class=special>) { </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>x</span><span class=special>,</span><span class=identifier>y</span><span class=special>); }
/*...*/
// </span><span class=identifier>in </span><span class=identifier>module </span><span class=identifier>init
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>f</span><span class=special>); // </span><span class=identifier>all </span><span class=identifier>arguments
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>f2</span><span class=special>); // </span><span class=identifier>two </span><span class=identifier>arguments
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>f1</span><span class=special>); // </span><span class=identifier>one </span><span class=identifier>argument
</span></pre></code>
<p>
When you want to wrap functions (or member functions) that either:</p>
<ul><li>have default arguments, or</li><li>are overloaded with a common sequence of initial arguments</li></ul><a name="boost_python_function_overloads"></a><h2>BOOST_PYTHON_FUNCTION_OVERLOADS</h2><p>
Boost.Python now has a way to make it easier. For instance, given a function:</p>
<code><pre>
<span class=keyword>int </span><span class=identifier>foo</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>char </span><span class=identifier>b </span><span class=special>= </span><span class=number>1</span><span class=special>, </span><span class=keyword>unsigned </span><span class=identifier>c </span><span class=special>= </span><span class=number>2</span><span class=special>, </span><span class=keyword>double </span><span class=identifier>d </span><span class=special>= </span><span class=number>3</span><span class=special>)
{
/*...*/
}
</span></pre></code>
<p>
The macro invocation:</p>
<code><pre>
<span class=identifier>BOOST_PYTHON_FUNCTION_OVERLOADS</span><span class=special>(</span><span class=identifier>foo_overloads</span><span class=special>, </span><span class=identifier>foo</span><span class=special>, </span><span class=number>1</span><span class=special>, </span><span class=number>4</span><span class=special>)
</span></pre></code>
<p>
will automatically create the thin wrappers for us. This macro will create
a class <tt>foo_overloads</tt> that can be passed on to <tt>def(...)</tt>. The third
and fourth macro argument are the minimum arguments and maximum arguments,
respectively. In our <tt>foo</tt> function the minimum number of arguments is 1
and the maximum number of arguments is 4. The <tt>def(...)</tt> function will
automatically add all the foo variants for us:</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;foo&quot;</span><span class=special>, </span><span class=identifier>foo</span><span class=special>, </span><span class=identifier>foo_overloads</span><span class=special>());
</span></pre></code>
<a name="boost_python_member_function_overloads"></a><h2>BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS</h2><p>
Objects here, objects there, objects here there everywhere. More frequently
than anything else, we need to expose member functions of our classes to
Python. Then again, we have the same inconveniences as before when default
arguments or overloads with a common sequence of initial arguments come
into play. Another macro is provided to make this a breeze.</p>
<p>
Like <tt>BOOST_PYTHON_FUNCTION_OVERLOADS</tt>,
<tt>BOOST_PYTHON_FUNCTION_OVERLOADS</tt> may be used to automatically create
the thin wrappers for wrapping member functions. Let's have an example:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>george
</span><span class=special>{
</span><span class=keyword>void
</span><span class=identifier>wack_em</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>int </span><span class=identifier>b </span><span class=special>= </span><span class=number>0</span><span class=special>, </span><span class=keyword>char </span><span class=identifier>c </span><span class=special>= </span><span class=literal>'x'</span><span class=special>)
{
/*...*/
}
};
</span></pre></code>
<p>
The macro invocation:</p>
<code><pre>
<span class=identifier>BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS</span><span class=special>(</span><span class=identifier>george_overloads</span><span class=special>, </span><span class=identifier>wack_em</span><span class=special>, </span><span class=number>1</span><span class=special>, </span><span class=number>3</span><span class=special>)
</span></pre></code>
<p>
will generate a set of thin wrappers for george's <tt>wack_em</tt> member function
accepting a minimum of 1 and a maximum of 3 arguments (i.e. the third and
fourth macro argument). The thin wrappers are all enclosed in a class named
<tt>george_overloads</tt> that can then be used as an argument to <tt>def(...)</tt>:</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;wack_em&quot;</span><span class=special>, &amp;</span><span class=identifier>george</span><span class=special>::</span><span class=identifier>wack_em</span><span class=special>, </span><span class=identifier>george_overloads</span><span class=special>());
</span></pre></code>
<p>
See the <a href="../../v2/overloads.html#BOOST_PYTHON_FUNCTION_OVERLOADS-spec">
overloads reference</a>
for details.</p>
<a name="init_and_optional"></a><h2>init and optional</h2><p>
A similar facility is provided for class constructors, again, with
default arguments or a sequence of overloads. Remember <tt>init&lt;...&gt;</tt>? For example,
given a class X with a constructor:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>X
</span><span class=special>{
</span><span class=identifier>X</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>char </span><span class=identifier>b </span><span class=special>= </span><span class=literal>'D'</span><span class=special>, </span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>c </span><span class=special>= </span><span class=string>&quot;constructor&quot;</span><span class=special>, </span><span class=keyword>double </span><span class=identifier>d </span><span class=special>= </span><span class=number>0.0</span><span class=special>);
/*...*/
}
</span></pre></code>
<p>
You can easily add this constructor to Boost.Python in one shot:</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=identifier>init</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>, </span><span class=identifier>optional</span><span class=special>&lt;</span><span class=keyword>char</span><span class=special>, </span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string</span><span class=special>, </span><span class=keyword>double</span><span class=special>&gt; &gt;())
</span></pre></code>
<p>
Notice the use of <tt>init&lt;...&gt;</tt> and <tt>optional&lt;...&gt;</tt> to signify the default
(optional arguments).</p>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="overloading.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="auto_overloading.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,117 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Derived Object types</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="basic_interface.html">
<link rel="next" href="extracting_c___objects.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Derived Object types</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="basic_interface.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="extracting_c___objects.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Boost.Python comes with a set of derived <tt>object</tt> types corresponding to
that of Python's:</p>
<ul><li>list</li><li>dict</li><li>tuple</li><li>str</li><li>long_</li><li>enum</li></ul><p>
These derived <tt>object</tt> types act like real Python types. For instance:</p>
<code><pre>
<span class=identifier>str</span><span class=special>(</span><span class=number>1</span><span class=special>) ==&gt; </span><span class=string>&quot;1&quot;
</span></pre></code>
<p>
Wherever appropriate, a particular derived <tt>object</tt> has corresponding
Python type's methods. For instance, <tt>dict</tt> has a <tt>keys()</tt> method:</p>
<code><pre>
<span class=identifier>d</span><span class=special>.</span><span class=identifier>keys</span><span class=special>()
</span></pre></code>
<p>
<tt>make_tuple</tt> is provided for declaring <i>tuple literals</i>. Example:</p>
<code><pre>
<span class=identifier>make_tuple</span><span class=special>(</span><span class=number>123</span><span class=special>, </span><span class=literal>'D'</span><span class=special>, </span><span class=string>&quot;Hello, World&quot;</span><span class=special>, </span><span class=number>0.0</span><span class=special>);
</span></pre></code>
<p>
In C++, when Boost.Python <tt>object</tt>s are used as arguments to functions,
subtype matching is required. For example, when a function <tt>f</tt>, as
declared below, is wrapped, it will only accept instances of Python's
<tt>str</tt> type and subtypes.</p>
<code><pre>
<span class=keyword>void </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>str </span><span class=identifier>name</span><span class=special>)
{
</span><span class=identifier>object </span><span class=identifier>n2 </span><span class=special>= </span><span class=identifier>name</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;upper&quot;</span><span class=special>)(); // </span><span class=identifier>NAME </span><span class=special>= </span><span class=identifier>name</span><span class=special>.</span><span class=identifier>upper</span><span class=special>()
</span><span class=identifier>str </span><span class=identifier>NAME </span><span class=special>= </span><span class=identifier>name</span><span class=special>.</span><span class=identifier>upper</span><span class=special>(); // </span><span class=identifier>better
</span><span class=identifier>object </span><span class=identifier>msg </span><span class=special>= </span><span class=string>&quot;%s is bigger than %s&quot; </span><span class=special>% </span><span class=identifier>make_tuple</span><span class=special>(</span><span class=identifier>NAME</span><span class=special>,</span><span class=identifier>name</span><span class=special>);
}
</span></pre></code>
<p>
In finer detail:</p>
<code><pre>
<span class=identifier>str </span><span class=identifier>NAME </span><span class=special>= </span><span class=identifier>name</span><span class=special>.</span><span class=identifier>upper</span><span class=special>();
</span></pre></code>
<p>
Illustrates that we provide versions of the str type's methods as C++
member functions.</p>
<code><pre>
<span class=identifier>object </span><span class=identifier>msg </span><span class=special>= </span><span class=string>&quot;%s is bigger than %s&quot; </span><span class=special>% </span><span class=identifier>make_tuple</span><span class=special>(</span><span class=identifier>NAME</span><span class=special>,</span><span class=identifier>name</span><span class=special>);
</span></pre></code>
<p>
Demonstrates that you can write the C++ equivalent of <tt>&quot;format&quot; % x,y,z</tt>
in Python, which is useful since there's no easy way to do that in std C++.</p>
<p>
<img src="theme/alert.gif"></img> <b>Beware</b> the common pitfall of forgetting that the constructors
of most of Python's mutable types make copies, just as in Python.</p>
<p>
Python:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>d </span><span class=special>= </span><span class=identifier>dict</span><span class=special>(</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>__dict__</span><span class=special>) </span>##<span class=identifier>copies </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>__dict__
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>d</span><span class=special>[</span><span class=literal>'whatever'</span><span class=special>] </span>##<span class=identifier>modifies </span><span class=identifier>the </span><span class=identifier>copy
</span></pre></code>
<p>
C++:</p>
<code><pre>
<span class=identifier>dict </span><span class=identifier>d</span><span class=special>(</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;__dict__&quot;</span><span class=special>)); </span>##<span class=identifier>copies </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>__dict__
</span><span class=identifier>d</span><span class=special>[</span><span class=literal>'whatever'</span><span class=special>] = </span><span class=number>3</span><span class=special>; </span>##<span class=identifier>modifies </span><span class=identifier>the </span><span class=identifier>copy
</span></pre></code>
<a name="class__lt_t_gt__as_objects"></a><h2>class_&lt;T&gt; as objects</h2><p>
Due to the dynamic nature of Boost.Python objects, any <tt>class_&lt;T&gt;</tt> may
also be one of these types! The following code snippet wraps the class
(type) object.</p>
<p>
We can use this to create wrapped instances. Example:</p>
<code><pre>
<span class=identifier>object </span><span class=identifier>vec345 </span><span class=special>= (
</span><span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Vec2</span><span class=special>&gt;(</span><span class=string>&quot;Vec2&quot;</span><span class=special>, </span><span class=identifier>init</span><span class=special>&lt;</span><span class=keyword>double</span><span class=special>, </span><span class=keyword>double</span><span class=special>&gt;())
.</span><span class=identifier>def_readonly</span><span class=special>(</span><span class=string>&quot;length&quot;</span><span class=special>, &amp;</span><span class=identifier>Point</span><span class=special>::</span><span class=identifier>length</span><span class=special>)
.</span><span class=identifier>def_readonly</span><span class=special>(</span><span class=string>&quot;angle&quot;</span><span class=special>, &amp;</span><span class=identifier>Point</span><span class=special>::</span><span class=identifier>angle</span><span class=special>)
)(</span><span class=number>3.0</span><span class=special>, </span><span class=number>4.0</span><span class=special>);
</span><span class=identifier>assert</span><span class=special>(</span><span class=identifier>vec345</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;length&quot;</span><span class=special>) == </span><span class=number>5.0</span><span class=special>);
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="basic_interface.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="extracting_c___objects.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,83 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Deriving a Python Class</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="class_virtual_functions.html">
<link rel="next" href="virtual_functions_with_default_implementations.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Deriving a Python Class</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_virtual_functions.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="virtual_functions_with_default_implementations.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Continuing, we can derive from our base class Base in Python and override
the virtual function in Python. Before we can do that, we have to set up
our <tt>class_</tt> wrapper as:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Base</span><span class=special>, </span><span class=identifier>BaseWrap</span><span class=special>, </span><span class=identifier>boost</span><span class=special>::</span><span class=identifier>noncopyable</span><span class=special>&gt;(</span><span class=string>&quot;Base&quot;</span><span class=special>)
;
</span></pre></code>
<p>
Otherwise, we have to suppress the Base class' <tt>no_init</tt> by adding an
<tt>__init__()</tt> method to all our derived classes. <tt>no_init</tt> actually adds
an <tt>__init__</tt> method that raises a Python RuntimeError exception.</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=keyword>class </span><span class=identifier>Derived</span><span class=special>(</span><span class=identifier>Base</span><span class=special>):
... </span><span class=identifier>def </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>self</span><span class=special>):
... </span><span class=keyword>return </span><span class=number>42
</span><span class=special>...
</span></pre></code>
<p>
Cool eh? A Python class deriving from a C++ class!</p>
<p>
Let's now make an instance of our Python class <tt>Derived</tt>:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>derived </span><span class=special>= </span><span class=identifier>Derived</span><span class=special>()
</span></pre></code>
<p>
Calling <tt>derived.f()</tt>:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>derived</span><span class=special>.</span><span class=identifier>f</span><span class=special>()
</span><span class=number>42
</span></pre></code>
<p>
Will yield the expected result. Finally, calling calling the free function
<tt>call_f</tt> with <tt>derived</tt> as argument:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>call_f</span><span class=special>(</span><span class=identifier>derived</span><span class=special>)
</span><span class=number>42
</span></pre></code>
<p>
Will also yield the expected result.</p>
<p>
Here's what's happening:</p>
<ol><li><tt>call_f(derived)</tt> is called in Python</li><li>This corresponds to <tt>def(&quot;call_f&quot;, call_f);</tt>. Boost.Python dispatches this call.</li><li><tt>int call_f(Base&amp; b) { return b.f(); }</tt> accepts the call.</li><li>The overridden virtual function <tt>f</tt> of <tt>BaseWrap</tt> is called.</li><li><tt>call_method&lt;int&gt;(self, &quot;f&quot;);</tt> dispatches the call back to Python.</li><li><tt>def f(self): return 42</tt> is finally called.</li></ol><table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_virtual_functions.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="virtual_functions_with_default_implementations.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,97 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Embedding</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="enums.html">
<link rel="next" href="using_the_interpreter.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Embedding</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="enums.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="using_the_interpreter.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
By now you should know how to use Boost.Python to call your C++ code from
Python. However, sometimes you may need to do the reverse: call Python code
from the C++-side. This requires you to <i>embed</i> the Python interpreter
into your C++ program.</p>
<p>
Currently, Boost.Python does not directly support everything you'll need
when embedding. Therefore you'll need to use the
<a href="http://www.python.org/doc/current/api/api.html">
Python/C API</a> to fill in
the gaps. However, Boost.Python already makes embedding a lot easier and,
in a future version, it may become unnecessary to touch the Python/C API at
all. So stay tuned... <img src="theme/smiley.gif"></img></p>
<a name="building_embedded_programs"></a><h2>Building embedded programs</h2><p>
To be able to use embedding in your programs, they have to be linked to
both Boost.Python's and Python's static link library.</p>
<p>
Boost.Python's static link library comes in two variants. Both are located
in Boost's <tt>/libs/python/build/bin-stage</tt> subdirectory. On Windows, the
variants are called <tt>boost_python.lib</tt> (for release builds) and
<tt>boost_python_debug.lib</tt> (for debugging). If you can't find the
libraries, you probably haven't built Boost.Python yet. See <a
href="../../building.html">Building and Testing</a> on how to do
this.</p>
<p>
Python's static link library can be found in the <tt>/libs</tt> subdirectory of
your Python directory. On Windows it is called pythonXY.lib where X.Y is
your major Python version number.</p>
<p>
Additionally, Python's <tt>/include</tt> subdirectory has to be added to your
include path.</p>
<p>
In a Jamfile, all the above boils down to:</p>
<code><pre>
projectroot c:\projects\embedded_program ; # location of the program
# bring in the rules for python
SEARCH on python.jam = $(BOOST_BUILD_PATH) ;
include python.jam ;
exe embedded_program # name of the executable
: #sources
embedded_program.cpp
: # requirements
&lt;find-library&gt;boost_python &lt;library-path&gt;c:\boost\libs\python
$(PYTHON_PROPERTIES)
&lt;library-path&gt;$(PYTHON_LIB_PATH)
&lt;find-library&gt;$(PYTHON_EMBEDDED_LIBRARY) ;
</pre></code><a name="getting_started"></a><h2>Getting started</h2><p>
Being able to build is nice, but there is nothing to build yet. Embedding
the Python interpreter into one of your C++ programs requires these 4
steps:</p>
<ol><li>#include <tt>&lt;boost/python.hpp&gt;</tt><br><br></li><li>Call <a href="http://www.python.org/doc/current/api/initialization.html#l2h-652">
Py_Initialize</a>() to start the interpreter and create the <tt>__main__</tt> module.<br><br></li><li>Call other Python C API routines to use the interpreter.<br><br></li><li>Call <a href="http://www.python.org/doc/current/api/initialization.html#l2h-656">
Py_Finalize</a>() to stop the interpreter and release its resources.</li></ol><p>
(Of course, there can be other C++ code between all of these steps.)</p>
<blockquote><p><i><b>Now that we can embed the interpreter in our programs, lets see how to put it to use...</b></i></p></blockquote><table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="enums.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="using_the_interpreter.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 Dirk Gerrits<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,95 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Enums</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="extracting_c___objects.html">
<link rel="next" href="embedding.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Enums</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="extracting_c___objects.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="embedding.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Boost.Python has a nifty facility to capture and wrap C++ enums. While
Python has no <tt>enum</tt> type, we'll often want to expose our C++ enums to
Python as an <tt>int</tt>. Boost.Python's enum facility makes this easy while
taking care of the proper conversions from Python's dynamic typing to C++'s
strong static typing (in C++, ints cannot be implicitly converted to
enums). To illustrate, given a C++ enum:</p>
<code><pre>
<span class=keyword>enum </span><span class=identifier>choice </span><span class=special>{ </span><span class=identifier>red</span><span class=special>, </span><span class=identifier>blue </span><span class=special>};
</span></pre></code>
<p>
the construct:</p>
<code><pre>
<span class=identifier>enum_</span><span class=special>&lt;</span><span class=identifier>choice</span><span class=special>&gt;(</span><span class=string>&quot;choice&quot;</span><span class=special>)
.</span><span class=identifier>value</span><span class=special>(</span><span class=string>&quot;red&quot;</span><span class=special>, </span><span class=identifier>red</span><span class=special>)
.</span><span class=identifier>value</span><span class=special>(</span><span class=string>&quot;blue&quot;</span><span class=special>, </span><span class=identifier>blue</span><span class=special>)
;
</span></pre></code>
<p>
can be used to expose to Python. The new enum type is created in the
current <tt>scope()</tt>, which is usually the current module. The snippet above
creates a Python class derived from Python's <tt>int</tt> type which is
associated with the C++ type passed as its first parameter.</p>
<table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/lens.gif"></img> <b>what is a scope?</b><br><br> The scope is a class that has an
associated global Python object which controls the Python namespace in
which new extension classes and wrapped functions will be defined as
attributes. Details can be found <a href="../../v2/scope.html">
here</a>. </td>
</tr>
</table>
<p>
You can access those values in Python as</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>my_module</span><span class=special>.</span><span class=identifier>choice</span><span class=special>.</span><span class=identifier>red
</span><span class=identifier>my_module</span><span class=special>.</span><span class=identifier>choice</span><span class=special>.</span><span class=identifier>red
</span></pre></code>
<p>
where my_module is the module where the enum is declared. You can also
create a new scope around a class:</p>
<code><pre>
<span class=identifier>scope </span><span class=identifier>in_X </span><span class=special>= </span><span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>X</span><span class=special>&gt;(</span><span class=string>&quot;X&quot;</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>( ... )
.</span><span class=identifier>def</span><span class=special>( ... )
;
// </span><span class=identifier>Expose </span><span class=identifier>X</span><span class=special>::</span><span class=identifier>nested </span><span class=identifier>as </span><span class=identifier>X</span><span class=special>.</span><span class=identifier>nested
</span><span class=identifier>enum_</span><span class=special>&lt;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>nested</span><span class=special>&gt;(</span><span class=string>&quot;nested&quot;</span><span class=special>)
.</span><span class=identifier>value</span><span class=special>(</span><span class=string>&quot;red&quot;</span><span class=special>, </span><span class=identifier>red</span><span class=special>)
.</span><span class=identifier>value</span><span class=special>(</span><span class=string>&quot;blue&quot;</span><span class=special>, </span><span class=identifier>blue</span><span class=special>)
;
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="extracting_c___objects.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="embedding.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,60 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Exception Translation</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="iterators.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Exception Translation</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="iterators.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><img src="theme/r_arr_disabled.gif" border="0"></td>
</tr>
</table>
<p>
All C++ exceptions must be caught at the boundary with Python code. This
boundary is the point where C++ meets Python. Boost.Python provides a
default exception handler that translates selected standard exceptions,
then gives up:</p>
<code><pre>
<span class=identifier>raise </span><span class=identifier>RuntimeError</span><span class=special>, </span><span class=literal>'unidentifiable C++ Exception'
</span></pre></code>
<p>
Users may provide custom translation. Here's an example:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>PodBayDoorException</span><span class=special>;
</span><span class=keyword>void </span><span class=identifier>translator</span><span class=special>(</span><span class=identifier>PodBayDoorException </span><span class=keyword>const</span><span class=special>&amp; </span><span class=identifier>x</span><span class=special>) {
</span><span class=identifier>PyErr_SetString</span><span class=special>(</span><span class=identifier>PyExc_UserWarning</span><span class=special>, </span><span class=string>&quot;I'm sorry Dave...&quot;</span><span class=special>);
}
</span><span class=identifier>BOOST_PYTHON_MODULE</span><span class=special>(</span><span class=identifier>kubrick</span><span class=special>) {
</span><span class=identifier>register_exception_translator</span><span class=special>&lt;
</span><span class=identifier>PodBayDoorException</span><span class=special>&gt;(</span><span class=identifier>translator</span><span class=special>);
...
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="iterators.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><img src="theme/r_arr_disabled.gif" border="0"></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,79 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Exposing Classes</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="building_hello_world.html">
<link rel="next" href="constructors.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Exposing Classes</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="building_hello_world.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="constructors.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Now let's expose a C++ class to Python.</p>
<p>
Consider a C++ class/struct that we want to expose to Python:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>World
</span><span class=special>{
</span><span class=keyword>void </span><span class=identifier>set</span><span class=special>(</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>msg</span><span class=special>) { </span><span class=keyword>this</span><span class=special>-&gt;</span><span class=identifier>msg </span><span class=special>= </span><span class=identifier>msg</span><span class=special>; }
</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>greet</span><span class=special>() { </span><span class=keyword>return </span><span class=identifier>msg</span><span class=special>; }
</span><span class=identifier>std</span><span class=special>::</span><span class=identifier>string </span><span class=identifier>msg</span><span class=special>;
};
</span></pre></code>
<p>
We can expose this to Python by writing a corresponding Boost.Python
C++ Wrapper:</p>
<code><pre>
<span class=preprocessor>#include </span><span class=special>&lt;</span><span class=identifier>boost</span><span class=special>/</span><span class=identifier>python</span><span class=special>.</span><span class=identifier>hpp</span><span class=special>&gt;
</span><span class=keyword>using </span><span class=keyword>namespace </span><span class=identifier>boost</span><span class=special>::</span><span class=identifier>python</span><span class=special>;
</span><span class=identifier>BOOST_PYTHON_MODULE</span><span class=special>(</span><span class=identifier>hello</span><span class=special>)
{
</span><span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>World</span><span class=special>&gt;(</span><span class=string>&quot;World&quot;</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;greet&quot;</span><span class=special>, &amp;</span><span class=identifier>World</span><span class=special>::</span><span class=identifier>greet</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;set&quot;</span><span class=special>, &amp;</span><span class=identifier>World</span><span class=special>::</span><span class=identifier>set</span><span class=special>)
;
}
</span></pre></code>
<p>
Here, we wrote a C++ class wrapper that exposes the member functions
<tt>greet</tt> and <tt>set</tt>. Now, after building our module as a shared library, we
may use our class <tt>World</tt> in Python. Here's a sample Python session:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>import </span><span class=identifier>hello
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>planet </span><span class=special>= </span><span class=identifier>hello</span><span class=special>.</span><span class=identifier>World</span><span class=special>()
&gt;&gt;&gt; </span><span class=identifier>planet</span><span class=special>.</span><span class=identifier>set</span><span class=special>(</span><span class=literal>'howdy'</span><span class=special>)
&gt;&gt;&gt; </span><span class=identifier>planet</span><span class=special>.</span><span class=identifier>greet</span><span class=special>()
</span><span class=literal>'howdy'
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="building_hello_world.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="constructors.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,79 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Extracting C++ objects</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="derived_object_types.html">
<link rel="next" href="enums.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Extracting C++ objects</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="derived_object_types.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="enums.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
At some point, we will need to get C++ values out of object instances. This
can be achieved with the <tt>extract&lt;T&gt;</tt> function. Consider the following:</p>
<code><pre>
<span class=keyword>double </span><span class=identifier>x </span><span class=special>= </span><span class=identifier>o</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;length&quot;</span><span class=special>); // </span><span class=identifier>compile </span><span class=identifier>error
</span></pre></code>
<p>
In the code above, we got a compiler error because Boost.Python
<tt>object</tt> can't be implicitly converted to <tt>double</tt>s. Instead, what
we wanted to do above can be achieved by writing:</p>
<code><pre>
<span class=keyword>double </span><span class=identifier>l </span><span class=special>= </span><span class=identifier>extract</span><span class=special>&lt;</span><span class=keyword>double</span><span class=special>&gt;(</span><span class=identifier>o</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;length&quot;</span><span class=special>));
</span><span class=identifier>Vec2</span><span class=special>&amp; </span><span class=identifier>v </span><span class=special>= </span><span class=identifier>extract</span><span class=special>&lt;</span><span class=identifier>Vec2</span><span class=special>&amp;&gt;(</span><span class=identifier>o</span><span class=special>);
</span><span class=identifier>assert</span><span class=special>(</span><span class=identifier>l </span><span class=special>== </span><span class=identifier>v</span><span class=special>.</span><span class=identifier>length</span><span class=special>());
</span></pre></code>
<p>
The first line attempts to extract the &quot;length&quot; attribute of the
Boost.Python <tt>object</tt> <tt>o</tt>. The second line attempts to <i>extract</i> the
<tt>Vec2</tt> object from held by the Boost.Python <tt>object</tt> <tt>o</tt>.</p>
<p>
Take note that we said &quot;attempt to&quot; above. What if the Boost.Python
<tt>object</tt> <tt>o</tt> does not really hold a <tt>Vec2</tt> type? This is certainly
a possibility considering the dynamic nature of Python <tt>object</tt>s. To
be on the safe side, if the C++ type can't be extracted, an
appropriate exception is thrown. To avoid an exception, we need to
test for extractibility:</p>
<code><pre>
<span class=identifier>extract</span><span class=special>&lt;</span><span class=identifier>Vec2</span><span class=special>&amp;&gt; </span><span class=identifier>x</span><span class=special>(</span><span class=identifier>o</span><span class=special>);
</span><span class=keyword>if </span><span class=special>(</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>check</span><span class=special>()) {
</span><span class=identifier>Vec2</span><span class=special>&amp; </span><span class=identifier>v </span><span class=special>= </span><span class=identifier>x</span><span class=special>(); ...
</span></pre></code>
<p>
<img src="theme/bulb.gif"></img> The astute reader might have noticed that the <tt>extract&lt;T&gt;</tt>
facility in fact solves the mutable copying problem:</p>
<code><pre>
<span class=identifier>dict </span><span class=identifier>d </span><span class=special>= </span><span class=identifier>extract</span><span class=special>&lt;</span><span class=identifier>dict</span><span class=special>&gt;(</span><span class=identifier>x</span><span class=special>.</span><span class=identifier>attr</span><span class=special>(</span><span class=string>&quot;__dict__&quot;</span><span class=special>));
</span><span class=identifier>d</span><span class=special>[</span><span class=literal>'whatever'</span><span class=special>] = </span><span class=number>3</span><span class=special>; </span>##<span class=identifier>modifies </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>__dict__ </span><span class=special>!
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="derived_object_types.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="enums.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,73 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Functions</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="class_operators_special_functions.html">
<link rel="next" href="call_policies.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Functions</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_operators_special_functions.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="call_policies.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
In this chapter, we'll look at Boost.Python powered functions in closer
detail. We shall see some facilities to make exposing C++ functions to
Python safe from potential pifalls such as dangling pointers and
references. We shall also see facilities that will make it even easier for
us to expose C++ functions that take advantage of C++ features such as
overloading and default arguments.</p>
<blockquote><p><i>Read on...</i></p></blockquote><p>
But before you do, you might want to fire up Python 2.2 or later and type
<tt>&gt;&gt;&gt; import this</tt>.</p>
<code><pre>
&gt;&gt;&gt; import this
The Zen of Python, by Tim Peters
Beautiful is better than ugly.
Explicit is better than implicit.
Simple is better than complex.
Complex is better than complicated.
Flat is better than nested.
Sparse is better than dense.
Readability counts.
Special cases aren't special enough to break the rules.
Although practicality beats purity.
Errors should never pass silently.
Unless explicitly silenced.
In the face of ambiguity, refuse the temptation to guess.
There should be one-- and preferably only one --obvious way to do it
Although that way may not be obvious at first unless you're Dutch.
Now is better than never.
Although never is often better than *right* now.
If the implementation is hard to explain, it's a bad idea.
If the implementation is easy to explain, it may be a good idea.
Namespaces are one honking great idea -- let's do more of those!
</pre></code><table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_operators_special_functions.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="call_policies.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,98 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Inheritance</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="class_properties.html">
<link rel="next" href="class_virtual_functions.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Inheritance</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_properties.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_virtual_functions.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
In the previous examples, we dealt with classes that are not polymorphic.
This is not often the case. Much of the time, we will be wrapping
polymorphic classes and class hierarchies related by inheritance. We will
often have to write Boost.Python wrappers for classes that are derived from
abstract base classes.</p>
<p>
Consider this trivial inheritance structure:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Base </span><span class=special>{ </span><span class=keyword>virtual </span><span class=special>~</span><span class=identifier>Base</span><span class=special>(); };
</span><span class=keyword>struct </span><span class=identifier>Derived </span><span class=special>: </span><span class=identifier>Base </span><span class=special>{};
</span></pre></code>
<p>
And a set of C++ functions operating on <tt>Base</tt> and <tt>Derived</tt> object
instances:</p>
<code><pre>
<span class=keyword>void </span><span class=identifier>b</span><span class=special>(</span><span class=identifier>Base</span><span class=special>*);
</span><span class=keyword>void </span><span class=identifier>d</span><span class=special>(</span><span class=identifier>Derived</span><span class=special>*);
</span><span class=identifier>Base</span><span class=special>* </span><span class=identifier>factory</span><span class=special>() { </span><span class=keyword>return </span><span class=keyword>new </span><span class=identifier>Derived</span><span class=special>; }
</span></pre></code>
<p>
We've seen how we can wrap the base class <tt>Base</tt>:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Base</span><span class=special>&gt;(</span><span class=string>&quot;Base&quot;</span><span class=special>)
/*...*/
;
</span></pre></code>
<p>
Now we can inform Boost.Python of the inheritance relationship between
<tt>Derived</tt> and its base class <tt>Base</tt>. Thus:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Derived</span><span class=special>, </span><span class=identifier>bases</span><span class=special>&lt;</span><span class=identifier>Base</span><span class=special>&gt; &gt;(</span><span class=string>&quot;Derived&quot;</span><span class=special>)
/*...*/
;
</span></pre></code>
<p>
Doing so, we get some things for free:</p>
<ol><li>Derived automatically inherits all of Base's Python methods (wrapped C++ member functions)</li><li><b>If</b> Base is polymorphic, <tt>Derived</tt> objects which have been passed to Python via a pointer or reference to <tt>Base</tt> can be passed where a pointer or reference to <tt>Derived</tt> is expected.</li></ol><p>
Now, we shall expose the C++ free functions <tt>b</tt> and <tt>d</tt> and <tt>factory</tt>:</p>
<code><pre>
<span class=identifier>def</span><span class=special>(</span><span class=string>&quot;b&quot;</span><span class=special>, </span><span class=identifier>b</span><span class=special>);
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;d&quot;</span><span class=special>, </span><span class=identifier>d</span><span class=special>);
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;factory&quot;</span><span class=special>, </span><span class=identifier>factory</span><span class=special>);
</span></pre></code>
<p>
Note that free function <tt>factory</tt> is being used to generate new
instances of class <tt>Derived</tt>. In such cases, we use
<tt>return_value_policy&lt;manage_new_object&gt;</tt> to instruct Python to adopt
the pointer to <tt>Base</tt> and hold the instance in a new Python <tt>Base</tt>
object until the the Python object is destroyed. We shall see more of
Boost.Python <a href="call_policies.html">
call policies</a> later.</p>
<code><pre>
<span class=comment>// Tell Python to take ownership of factory's result
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;factory&quot;</span><span class=special>, </span><span class=identifier>factory</span><span class=special>,
</span><span class=identifier>return_value_policy</span><span class=special>&lt;</span><span class=identifier>manage_new_object</span><span class=special>&gt;());
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="class_properties.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_virtual_functions.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,101 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Iterators</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="using_the_interpreter.html">
<link rel="next" href="exception_translation.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Iterators</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="using_the_interpreter.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="exception_translation.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
In C++, and STL in particular, we see iterators everywhere. Python also has
iterators, but these are two very different beasts.</p>
<p>
<b>C++ iterators:</b></p>
<ul><li>C++ has 5 type categories (random-access, bidirectional, forward, input, output)</li><li>There are 2 Operation categories: reposition, access</li><li>A pair of iterators is needed to represent a (first/last) range.</li></ul><p>
<b>Python Iterators:</b></p>
<ul><li>1 category (forward)</li><li>1 operation category (next())</li><li>Raises StopIteration exception at end</li></ul><p>
The typical Python iteration protocol: <tt><b>for y in x...</b></tt> is as follows:</p>
<code><pre>
<span class=identifier>iter </span><span class=special>= </span><span class=identifier>x</span><span class=special>.</span><span class=identifier>__iter__</span><span class=special>() </span>##<span class=identifier>get </span><span class=identifier>iterator
</span><span class=keyword>try</span><span class=special>:
</span><span class=keyword>while </span><span class=number>1</span><span class=special>:
</span><span class=identifier>y </span><span class=special>= </span><span class=identifier>iter</span><span class=special>.</span><span class=identifier>next</span><span class=special>() </span>##<span class=identifier>get </span><span class=identifier>each </span><span class=identifier>item
</span><span class=special>... </span>##<span class=identifier>process </span><span class=identifier>y
</span><span class=identifier>except </span><span class=identifier>StopIteration</span><span class=special>: </span><span class=identifier>pass </span>##<span class=identifier>iterator </span><span class=identifier>exhausted
</span></pre></code>
<p>
Boost.Python provides some mechanisms to make C++ iterators play along
nicely as Python iterators. What we need to do is to produce
appropriate __iter__ function from C++ iterators that is compatible
with the Python iteration protocol. For example:</p>
<code><pre>
<span class=identifier>object </span><span class=identifier>get_iterator </span><span class=special>= </span><span class=identifier>iterator</span><span class=special>&lt;</span><span class=identifier>vector</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt; &gt;();
</span><span class=identifier>object </span><span class=identifier>iter </span><span class=special>= </span><span class=identifier>get_iterator</span><span class=special>(</span><span class=identifier>v</span><span class=special>);
</span><span class=identifier>object </span><span class=identifier>first </span><span class=special>= </span><span class=identifier>iter</span><span class=special>.</span><span class=identifier>next</span><span class=special>();
</span></pre></code>
<p>
Or for use in class_&lt;&gt;:</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;__iter__&quot;</span><span class=special>, </span><span class=identifier>iterator</span><span class=special>&lt;</span><span class=identifier>vector</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt; &gt;())
</span></pre></code>
<p>
<b>range</b></p>
<p>
We can create a Python savvy iterator using the range function:</p>
<ul><li>range(start, finish)</li><li>range&lt;Policies,Target&gt;(start, finish)</li></ul><p>
Here, start/finish may be one of:</p>
<ul><li>member data pointers</li><li>member function pointers</li><li>adaptable function object (use Target parameter)</li></ul><p>
<b>iterator</b></p>
<ul><li>iterator&lt;T, Policies&gt;()</li></ul><p>
Given a container <tt>T</tt>, iterator is a shortcut that simply calls <tt>range</tt>
with &amp;T::begin, &amp;T::end.</p>
<p>
Let's put this into action... Here's an example from some hypothetical
bogon Particle accelerator code:</p>
<code><pre>
<span class=identifier>f </span><span class=special>= </span><span class=identifier>Field</span><span class=special>()
</span><span class=keyword>for </span><span class=identifier>x </span><span class=identifier>in </span><span class=identifier>f</span><span class=special>.</span><span class=identifier>pions</span><span class=special>:
</span><span class=identifier>smash</span><span class=special>(</span><span class=identifier>x</span><span class=special>)
</span><span class=keyword>for </span><span class=identifier>y </span><span class=identifier>in </span><span class=identifier>f</span><span class=special>.</span><span class=identifier>bogons</span><span class=special>:
</span><span class=identifier>count</span><span class=special>(</span><span class=identifier>y</span><span class=special>)
</span></pre></code>
<p>
Now, our C++ Wrapper:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>F</span><span class=special>&gt;(</span><span class=string>&quot;Field&quot;</span><span class=special>)
.</span><span class=identifier>property</span><span class=special>(</span><span class=string>&quot;pions&quot;</span><span class=special>, </span><span class=identifier>range</span><span class=special>(&amp;</span><span class=identifier>F</span><span class=special>::</span><span class=identifier>p_begin</span><span class=special>, &amp;</span><span class=identifier>F</span><span class=special>::</span><span class=identifier>p_end</span><span class=special>))
.</span><span class=identifier>property</span><span class=special>(</span><span class=string>&quot;bogons&quot;</span><span class=special>, </span><span class=identifier>range</span><span class=special>(&amp;</span><span class=identifier>F</span><span class=special>::</span><span class=identifier>b_begin</span><span class=special>, &amp;</span><span class=identifier>F</span><span class=special>::</span><span class=identifier>b_end</span><span class=special>));
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="using_the_interpreter.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="exception_translation.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,54 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Object Interface</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="auto_overloading.html">
<link rel="next" href="basic_interface.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Object Interface</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="auto_overloading.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="basic_interface.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Python is dynamically typed, unlike C++ which is statically typed. Python
variables may hold an integer, a float, list, dict, tuple, str, long etc.,
among other things. In the viewpoint of Boost.Python and C++, these
Pythonic variables are just instances of class <tt>object</tt>. We shall see in
this chapter how to deal with Python objects.</p>
<p>
As mentioned, one of the goals of Boost.Python is to provide a
bidirectional mapping between C++ and Python while maintaining the Python
feel. Boost.Python C++ <tt>object</tt>s are as close as possible to Python. This
should minimize the learning curve significantly.</p>
<p>
<img src="theme/python.png"></img></p>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="auto_overloading.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="basic_interface.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,88 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Overloading</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="call_policies.html">
<link rel="next" href="default_arguments.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Overloading</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="call_policies.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="default_arguments.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
The following illustrates a scheme for manually wrapping an overloaded
member functions. Of course, the same technique can be applied to wrapping
overloaded non-member functions.</p>
<p>
We have here our C++ class:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>X
</span><span class=special>{
</span><span class=keyword>bool </span><span class=identifier>f</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>)
{
</span><span class=keyword>return </span><span class=keyword>true</span><span class=special>;
}
</span><span class=keyword>bool </span><span class=identifier>f</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>double </span><span class=identifier>b</span><span class=special>)
{
</span><span class=keyword>return </span><span class=keyword>true</span><span class=special>;
}
</span><span class=keyword>bool </span><span class=identifier>f</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>double </span><span class=identifier>b</span><span class=special>, </span><span class=keyword>char </span><span class=identifier>c</span><span class=special>)
{
</span><span class=keyword>return </span><span class=keyword>true</span><span class=special>;
}
</span><span class=keyword>int </span><span class=identifier>f</span><span class=special>(</span><span class=keyword>int </span><span class=identifier>a</span><span class=special>, </span><span class=keyword>int </span><span class=identifier>b</span><span class=special>, </span><span class=keyword>int </span><span class=identifier>c</span><span class=special>)
{
</span><span class=keyword>return </span><span class=identifier>a </span><span class=special>+ </span><span class=identifier>b </span><span class=special>+ </span><span class=identifier>c</span><span class=special>;
};
};
</span></pre></code>
<p>
Class X has 4 overloaded functions. We shall start by introducing some
member function pointer variables:</p>
<code><pre>
<span class=keyword>bool </span><span class=special>(</span><span class=identifier>X</span><span class=special>::*</span><span class=identifier>fx1</span><span class=special>)(</span><span class=keyword>int</span><span class=special>) = &amp;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>f</span><span class=special>;
</span><span class=keyword>bool </span><span class=special>(</span><span class=identifier>X</span><span class=special>::*</span><span class=identifier>fx2</span><span class=special>)(</span><span class=keyword>int</span><span class=special>, </span><span class=keyword>double</span><span class=special>) = &amp;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>f</span><span class=special>;
</span><span class=keyword>bool </span><span class=special>(</span><span class=identifier>X</span><span class=special>::*</span><span class=identifier>fx3</span><span class=special>)(</span><span class=keyword>int</span><span class=special>, </span><span class=keyword>double</span><span class=special>, </span><span class=keyword>char</span><span class=special>)= &amp;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>f</span><span class=special>;
</span><span class=keyword>int </span><span class=special>(</span><span class=identifier>X</span><span class=special>::*</span><span class=identifier>fx4</span><span class=special>)(</span><span class=keyword>int</span><span class=special>, </span><span class=keyword>int</span><span class=special>, </span><span class=keyword>int</span><span class=special>) = &amp;</span><span class=identifier>X</span><span class=special>::</span><span class=identifier>f</span><span class=special>;
</span></pre></code>
<p>
With these in hand, we can proceed to define and wrap this for Python:</p>
<code><pre>
<span class=special>.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>fx1</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>fx2</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>fx3</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, </span><span class=identifier>fx4</span><span class=special>)
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="call_policies.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="default_arguments.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,79 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>QuickStart</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="next" href="building_hello_world.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>QuickStart</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><img src="theme/l_arr_disabled.gif" border="0"></td>
<td width="20"><a href="building_hello_world.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
The Boost Python Library is a framework for interfacing Python and
C++. It allows you to quickly and seamlessly expose C++ classes
functions and objects to Python, and vice-versa, using no special
tools -- just your C++ compiler. It is designed to wrap C++ interfaces
non-intrusively, so that you should not have to change the C++ code at
all in order to wrap it, making Boost.Python ideal for exposing
3rd-party libraries to Python. The library's use of advanced
metaprogramming techniques simplifies its syntax for users, so that
wrapping code takes on the look of a kind of declarative interface
definition language (IDL).</p>
<a name="hello_world"></a><h2>Hello World</h2><p>
Following C/C++ tradition, let's start with the &quot;hello, world&quot;. A C++
Function:</p>
<code><pre>
<span class=keyword>char </span><span class=keyword>const</span><span class=special>* </span><span class=identifier>greet</span><span class=special>()
{
</span><span class=keyword>return </span><span class=string>&quot;hello, world&quot;</span><span class=special>;
}
</span></pre></code>
<p>
can be exposed to Python by writing a Boost.Python wrapper:</p>
<code><pre>
<span class=preprocessor>#include </span><span class=special>&lt;</span><span class=identifier>boost</span><span class=special>/</span><span class=identifier>python</span><span class=special>.</span><span class=identifier>hpp</span><span class=special>&gt;
</span><span class=keyword>using </span><span class=keyword>namespace </span><span class=identifier>boost</span><span class=special>::</span><span class=identifier>python</span><span class=special>;
</span><span class=identifier>BOOST_PYTHON_MODULE</span><span class=special>(</span><span class=identifier>hello</span><span class=special>)
{
</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;greet&quot;</span><span class=special>, </span><span class=identifier>greet</span><span class=special>);
}
</span></pre></code>
<p>
That's it. We're done. We can now build this as a shared library. The
resulting DLL is now visible to Python. Here's a sample Python session:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>import </span><span class=identifier>hello
</span><span class=special>&gt;&gt;&gt; </span><span class=identifier>print </span><span class=identifier>hello</span><span class=special>.</span><span class=identifier>greet</span><span class=special>()
</span><span class=identifier>hello</span><span class=special>, </span><span class=identifier>world
</span></pre></code>
<blockquote><p><i><b>Next stop... Building your Hello World module from start to finish...</b></i></p></blockquote><table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><img src="theme/l_arr_disabled.gif" border="0"></td>
<td width="20"><a href="building_hello_world.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

File diff suppressed because it is too large Load Diff

Binary file not shown.

Before

Width:  |  Height:  |  Size: 577 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 70 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 1.3 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 2.5 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 944 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 152 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 8.6 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 3.8 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 147 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 91 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 897 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 151 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 14 KiB

Binary file not shown.

Before

Width:  |  Height:  |  Size: 147 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 91 B

Binary file not shown.

Before

Width:  |  Height:  |  Size: 879 B

View File

@@ -1,170 +0,0 @@
body
{
background-image: url(bkd.gif);
background-color: #FFFFFF;
margin: 1em 2em 1em 2em;
}
h1 { font-family: Verdana, Arial, Helvetica, sans-serif; font-weight: bold; text-align: left; }
h2 { font: 140% sans-serif; font-weight: bold; text-align: left; }
h3 { font: 120% sans-serif; font-weight: bold; text-align: left; }
h4 { font: bold 100% sans-serif; font-weight: bold; text-align: left; }
h5 { font: italic 100% sans-serif; font-weight: bold; text-align: left; }
h6 { font: small-caps 100% sans-serif; font-weight: bold; text-align: left; }
pre
{
border-top: gray 1pt solid;
border-right: gray 1pt solid;
border-left: gray 1pt solid;
border-bottom: gray 1pt solid;
padding-top: 2pt;
padding-right: 2pt;
padding-left: 2pt;
padding-bottom: 2pt;
display: block;
font-family: "courier new", courier, mono;
background-color: #eeeeee; font-size: small
}
code
{
font-family: "Courier New", Courier, mono;
font-size: small
}
tt
{
display: inline;
font-family: "Courier New", Courier, mono;
color: #000099;
font-size: small
}
p
{
text-align: justify;
font-family: Georgia, "Times New Roman", Times, serif
}
ul
{
list-style-image: url(bullet.gif);
font-family: Georgia, "Times New Roman", Times, serif
}
ol
{
font-family: Georgia, "Times New Roman", Times, serif
}
a
{
font-weight: bold;
color: #003366;
text-decoration: none;
}
a:hover { color: #8080FF; }
.literal { color: #666666; font-style: italic}
.keyword { color: #000099}
.identifier {}
.comment { font-style: italic; color: #990000}
.special { color: #800040}
.preprocessor { color: #FF0000}
.string { font-style: italic; color: #666666}
.copyright { color: #666666; font-size: small}
.white_bkd { background-color: #FFFFFF}
.dk_grey_bkd { background-color: #999999}
.quotes { color: #666666; font-style: italic; font-weight: bold}
.note_box
{
display: block;
border-top: gray 1pt solid;
border-right: gray 1pt solid;
border-left: gray 1pt solid;
border-bottom: gray 1pt solid;
padding-right: 12pt;
padding-left: 12pt;
padding-bottom: 12pt;
padding-top: 12pt;
font-family: Arial, Helvetica, sans-serif;
background-color: #E2E9EF;
font-size: small; text-align: justify
}
.table_title
{
background-color: #648CCA;
font-family: Verdana, Arial, Helvetica, sans-serif; color: #FFFFFF;
font-weight: bold
; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 4px
}
.table_cells
{
background-color: #E2E9EF;
font-family: Geneva, Arial, Helvetica, san-serif;
font-size: small
; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 4px
}
.toc
{
DISPLAY: block;
background-color: #E2E9EF
font-family: Arial, Helvetica, sans-serif;
border-top: gray 1pt solid;
border-left: gray 1pt solid;
border-bottom: gray 1pt solid;
border-right: gray 1pt solid;
padding-top: 24pt;
padding-right: 24pt;
padding-left: 24pt;
padding-bottom: 24pt;
}
.toc_title
{
background-color: #648CCA;
padding-top: 4px;
padding-right: 4px;
padding-bottom: 4px;
padding-left: 4px;
font-family: Geneva, Arial, Helvetica, san-serif;
color: #FFFFFF;
font-weight: bold
}
.toc_cells
{
background-color: #E2E9EF;
padding-top: 4px;
padding-right: 4px;
padding-bottom: 4px;
padding-left: 4px;
font-family: Geneva, Arial, Helvetica, san-serif;
font-size: small
}
div.logo
{
float: right;
}
.toc_cells_L0 { background-color: #E2E9EF; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 4px; font-family: Geneva, Arial, Helvetica, san-serif; font-size: small }
.toc_cells_L1 { background-color: #E2E9EF; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 44px; font-family: Geneva, Arial, Helvetica, san-serif; font-size: small }
.toc_cells_L2 { background-color: #E2E9EF; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 88px; font-family: Geneva, Arial, Helvetica, san-serif; font-size: small }
.toc_cells_L3 { background-color: #E2E9EF; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 122px; font-family: Geneva, Arial, Helvetica, san-serif; font-size: small }
.toc_cells_L4 { background-color: #E2E9EF; padding-top: 4px; padding-right: 4px; padding-bottom: 4px; padding-left: 166px; font-family: Geneva, Arial, Helvetica, san-serif; font-size: small }

Binary file not shown.

Before

Width:  |  Height:  |  Size: 170 B

View File

@@ -1,236 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Using the interpreter</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="embedding.html">
<link rel="next" href="iterators.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Using the interpreter</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="embedding.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="iterators.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
As you probably already know, objects in Python are reference-counted.
Naturally, the <tt>PyObject</tt>s of the Python/C API are also reference-counted.
There is a difference however. While the reference-counting is fully
automatic in Python, the Python/C API requires you to do it
<a href="http://www.python.org/doc/current/api/refcounts.html">
by hand</a>. This is
messy and especially hard to get right in the presence of C++ exceptions.
Fortunately Boost.Python provides the <a href="../../v2/handle.html">
handle</a> class
template to automate the process.</p>
<a name="reference_counting_handles"></a><h2>Reference-counting handles</h2><p>
There are two ways in which a function in the Python/C API can return a
<tt>PyObject*</tt>: as a <i>borrowed reference</i> or as a <i>new reference</i>. Which of
these a function uses, is listed in that function's documentation. The two
require slightely different approaches to reference-counting but both can
be 'handled' by Boost.Python.</p>
<p>
For a function returning a <i>borrowed reference</i> we'll have to tell the
<tt>handle</tt> that the <tt>PyObject*</tt> is borrowed with the aptly named
<a href="../../v2/handle.html#borrowed-spec">
borrowed</a> function. Two functions
returning borrowed references are <a href="http://www.python.org/doc/current/api/importing.html#l2h-125">
PyImport_AddModule</a> and <a href="http://www.python.org/doc/current/api/moduleObjects.html#l2h-594">
PyModule_GetDict</a>.
The former returns a reference to an already imported module, the latter
retrieves a module's namespace dictionary. Let's use them to retrieve the
namespace of the <tt>__main__</tt> module:</p>
<code><pre>
<span class=identifier>handle</span><span class=special>&lt;&gt; </span><span class=identifier>main_module</span><span class=special>(</span><span class=identifier>borrowed</span><span class=special>( </span><span class=identifier>PyImport_AddModule</span><span class=special>(</span><span class=string>&quot;__main__&quot;</span><span class=special>) ));
</span><span class=identifier>handle</span><span class=special>&lt;&gt; </span><span class=identifier>main_namespace</span><span class=special>(</span><span class=identifier>borrowed</span><span class=special>( </span><span class=identifier>PyModule_GetDict</span><span class=special>(</span><span class=identifier>main_module</span><span class=special>.</span><span class=identifier>get</span><span class=special>()) ));
</span></pre></code>
<p>
Because the Python/C API doesn't know anything about <tt>handle</tt>s, we used
the <a href="../../v2/handle.html#handle-spec-observers">
get</a> member function to
retrieve the <tt>PyObject*</tt> from which the <tt>handle</tt> was constructed.</p>
<p>
For a function returning a <i>new reference</i> we can just create a <tt>handle</tt>
out of the raw <tt>PyObject*</tt> without wrapping it in a call to borrowed. One
such function that returns a new reference is <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-55">
PyRun_String</a> which we'll
discuss in the next section.</p>
<table width="80%" border="0" align="center">
<tr>
<td class="note_box">
<img src="theme/lens.gif"></img> <b>Handle is a class <i>template</i>, so why haven't we been using any template parameters?</b><br>
<br>
<tt>handle</tt> has a single template parameter specifying the type of the managed object. This type is <tt>PyObject</tt> 99% of the time, so the parameter was defaulted to <tt>PyObject</tt> for convenience. Therefore we can use the shorthand <tt>handle&lt;&gt;</tt> instead of the longer, but equivalent, <tt>handle&lt;PyObject&gt;</tt>.
</td>
</tr>
</table>
<a name="running_python_code"></a><h2>Running Python code</h2><p>
To run Python code from C++ there is a family of functions in the API
starting with the PyRun prefix. You can find the full list of these
functions <a href="http://www.python.org/doc/current/api/veryhigh.html">
here</a>. They
all work similarly so we will look at only one of them, namely:</p>
<code><pre>
<span class=identifier>PyObject</span><span class=special>* </span><span class=identifier>PyRun_String</span><span class=special>(</span><span class=keyword>char </span><span class=special>*</span><span class=identifier>str</span><span class=special>, </span><span class=keyword>int </span><span class=identifier>start</span><span class=special>, </span><span class=identifier>PyObject </span><span class=special>*</span><span class=identifier>globals</span><span class=special>, </span><span class=identifier>PyObject </span><span class=special>*</span><span class=identifier>locals</span><span class=special>)
</span></pre></code>
<p>
<a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-55">
PyRun_String</a> takes the code to execute as a null-terminated (C-style)
string in its <tt>str</tt> parameter. The function returns a new reference to a
Python object. Which object is returned depends on the <tt>start</tt> paramater.</p>
<p>
The <tt>start</tt> parameter is the start symbol from the Python grammar to use
for interpreting the code. The possible values are:</p>
<table width="90%" border="0" align="center"> <tr>
<td class="table_title" colspan="6">
Start symbols </td>
</tr>
<tr><tr><td class="table_cells"><a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-58">
Py_eval_input</a></td><td class="table_cells">for interpreting isolated expressions</td></tr><td class="table_cells"><a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-59">
Py_file_input</a></td><td class="table_cells">for interpreting sequences of statements</td></tr><td class="table_cells"><a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-60">
Py_single_input</a></td><td class="table_cells">for interpreting a single statement</td></tr></table>
<p>
When using <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-58">
Py_eval_input</a>, the input string must contain a single expression
and its result is returned. When using <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-59">
Py_file_input</a>, the string can
contain an abitrary number of statements and None is returned.
<a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-60">
Py_single_input</a> works in the same way as <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-59">
Py_file_input</a> but only accepts a
single statement.</p>
<p>
Lastly, the <tt>globals</tt> and <tt>locals</tt> parameters are Python dictionaries
containing the globals and locals of the context in which to run the code.
For most intents and purposes you can use the namespace dictionary of the
<tt>__main__</tt> module for both parameters.</p>
<p>
We have already seen how to get the <tt>__main__</tt> module's namespace so let's
run some Python code in it:</p>
<code><pre>
<span class=identifier>handle</span><span class=special>&lt;&gt; </span><span class=identifier>main_module</span><span class=special>(</span><span class=identifier>borrowed</span><span class=special>( </span><span class=identifier>PyImport_AddModule</span><span class=special>(</span><span class=string>&quot;__main__&quot;</span><span class=special>) ));
</span><span class=identifier>handle</span><span class=special>&lt;&gt; </span><span class=identifier>main_namespace</span><span class=special>(</span><span class=identifier>borrowed</span><span class=special>( </span><span class=identifier>PyModule_GetDict</span><span class=special>(</span><span class=identifier>main_module</span><span class=special>.</span><span class=identifier>get</span><span class=special>()) ));
</span><span class=identifier>handle</span><span class=special>&lt;&gt;( </span><span class=identifier>PyRun_String</span><span class=special>(</span><span class=string>&quot;hello = file('hello.txt', 'w')\n&quot;
</span><span class=string>&quot;hello.write('Hello world!')\n&quot;
</span><span class=string>&quot;hello.close()&quot;</span><span class=special>, </span><span class=identifier>Py_file_input</span><span class=special>,
</span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>get</span><span class=special>(), </span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>get</span><span class=special>()) );
</span></pre></code>
<p>
This should create a file called 'hello.txt' in the current directory
containing a phrase that is well-known in programming circles.</p>
<p>
<img src="theme/note.gif"></img> <b>Note</b> that we wrap the return value of <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-55">
PyRun_String</a> in a
(nameless) <tt>handle</tt> even though we are not interested in it. If we didn't
do this, the the returned object would be kept alive unnecessarily. Unless
you want to be a Dr. Frankenstein, always wrap <tt>PyObject*</tt>s in <tt>handle</tt>s.</p>
<a name="beyond_handles"></a><h2>Beyond handles</h2><p>
It's nice that <tt>handle</tt> manages the reference counting details for us, but
other than that it doesn't do much. Often we'd like to have a more useful
class to manipulate Python objects. But we have already seen such a class
in the <a href="object_interface.html">
previous section</a>: the aptly named <tt>object</tt>
class and it's derivatives. What we haven't seen, is that they can be
constructed from a <tt>handle</tt>. The following examples should illustrate this
fact:</p>
<code><pre>
<span class=identifier>handle</span><span class=special>&lt;&gt; </span><span class=identifier>main_module</span><span class=special>(</span><span class=identifier>borrowed</span><span class=special>( </span><span class=identifier>PyImport_AddModule</span><span class=special>(</span><span class=string>&quot;__main__&quot;</span><span class=special>) ));
</span><span class=identifier>main_namespace </span><span class=identifier>dict</span><span class=special>(</span><span class=identifier>handle</span><span class=special>&lt;&gt;(</span><span class=identifier>borrowed</span><span class=special>( </span><span class=identifier>PyModule_GetDict</span><span class=special>(</span><span class=identifier>main_module</span><span class=special>.</span><span class=identifier>get</span><span class=special>()) )));
</span><span class=identifier>handle</span><span class=special>&lt;&gt;( </span><span class=identifier>PyRun_String</span><span class=special>(</span><span class=string>&quot;result = 5 ** 2&quot;</span><span class=special>, </span><span class=identifier>Py_file_input</span><span class=special>,
</span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>(), </span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>()) );
</span><span class=keyword>int </span><span class=identifier>five_squared </span><span class=special>= </span><span class=identifier>extract</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt;( </span><span class=identifier>main_namespace</span><span class=special>[</span><span class=string>&quot;result&quot;</span><span class=special>] );
</span></pre></code>
<p>
Here we create a dictionary object for the <tt>__main__</tt> module's namespace.
Then we assign 5 squared to the result variable and read this variable from
the dictionary. Another way to achieve the same result is to let
<a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-55">
PyRun_String</a> return the result directly with <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-58">
Py_eval_input</a>:</p>
<code><pre>
<span class=identifier>object </span><span class=identifier>result</span><span class=special>(</span><span class=identifier>handle</span><span class=special>&lt;&gt;( </span><span class=identifier>PyRun_String</span><span class=special>(</span><span class=string>&quot;5 ** 2&quot;</span><span class=special>, </span><span class=identifier>Py_eval_input</span><span class=special>,
</span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>(), </span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>()) ));
</span><span class=keyword>int </span><span class=identifier>five_squared </span><span class=special>= </span><span class=identifier>extract</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt;(</span><span class=identifier>result</span><span class=special>);
</span></pre></code>
<p>
<img src="theme/note.gif"></img> <b>Note</b> that <tt>object</tt>'s member function to return the wrapped
<tt>PyObject*</tt> is called <tt>ptr</tt> instead of <tt>get</tt>. This makes sense if you
take into account the different functions that <tt>object</tt> and <tt>handle</tt>
perform.</p>
<a name="exception_handling"></a><h2>Exception handling</h2><p>
If an exception occurs in the execution of some Python code, the <a href="http://www.python.org/doc/current/api/veryhigh.html#l2h-55">
PyRun_String</a> function returns a null pointer. Constructing a <tt>handle</tt> out of this null pointer throws <a href="../../v2/errors.html#error_already_set-spec">
error_already_set</a>, so basically, the Python exception is automatically translated into a C++ exception when using <tt>handle</tt>:</p>
<code><pre>
<span class=keyword>try
</span><span class=special>{
</span><span class=identifier>object </span><span class=identifier>result</span><span class=special>(</span><span class=identifier>handle</span><span class=special>&lt;&gt;( </span><span class=identifier>PyRun_String</span><span class=special>(</span><span class=string>&quot;5/0&quot;</span><span class=special>, </span><span class=identifier>Py_eval_input</span><span class=special>,
</span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>(), </span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>()) ));
// </span><span class=identifier>execution </span><span class=identifier>will </span><span class=identifier>never </span><span class=identifier>get </span><span class=identifier>here</span><span class=special>:
</span><span class=keyword>int </span><span class=identifier>five_divided_by_zero </span><span class=special>= </span><span class=identifier>extract</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt;(</span><span class=identifier>result</span><span class=special>);
}
</span><span class=keyword>catch</span><span class=special>(</span><span class=identifier>error_already_set</span><span class=special>)
{
// </span><span class=identifier>handle </span><span class=identifier>the </span><span class=identifier>exception </span><span class=identifier>in </span><span class=identifier>some </span><span class=identifier>way
</span><span class=special>}
</span></pre></code>
<p>
The <tt>error_already_set</tt> exception class doesn't carry any information in itself. To find out more about the Python exception that occurred, you need to use the <a href="http://www.python.org/doc/api/exceptionHandling.html">
exception handling functions</a> of the Python/C API in your catch-statement. This can be as simple as calling <a href="http://www.python.org/doc/api/exceptionHandling.html#l2h-70">
PyErr_Print()</a> to print the exception's traceback to the console, or comparing the type of the exception with those of the <a href="http://www.python.org/doc/api/standardExceptions.html">
standard exceptions</a>:</p>
<code><pre>
<span class=keyword>catch</span><span class=special>(</span><span class=identifier>error_already_set</span><span class=special>)
{
</span><span class=keyword>if </span><span class=special>(</span><span class=identifier>PyErr_ExceptionMatches</span><span class=special>(</span><span class=identifier>PyExc_ZeroDivisionError</span><span class=special>))
{
// </span><span class=identifier>handle </span><span class=identifier>ZeroDivisionError </span><span class=identifier>specially
</span><span class=special>}
</span><span class=keyword>else
</span><span class=special>{
// </span><span class=identifier>print </span><span class=identifier>all </span><span class=identifier>other </span><span class=identifier>errors </span><span class=identifier>to </span><span class=identifier>stderr
</span><span class=identifier>PyErr_Print</span><span class=special>();
}
}
</span></pre></code>
<p>
(To retrieve even more information from the exception you can use some of the other exception handling functions listed <a href="http://www.python.org/doc/api/exceptionHandling.html">
here</a>.)</p>
<p>
If you'd rather not have <tt>handle</tt> throw a C++ exception when it is constructed, you can use the <a href="../../v2/handle.html#allow_null-spec">
allow_null</a> function in the same way you'd use borrowed:</p>
<code><pre>
<span class=identifier>handle</span><span class=special>&lt;&gt; </span><span class=identifier>result</span><span class=special>(</span><span class=identifier>allow_null</span><span class=special>( </span><span class=identifier>PyRun_String</span><span class=special>(</span><span class=string>&quot;5/0&quot;</span><span class=special>, </span><span class=identifier>Py_eval_input</span><span class=special>,
</span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>(), </span><span class=identifier>main_namespace</span><span class=special>.</span><span class=identifier>ptr</span><span class=special>()) ));
</span><span class=keyword>if </span><span class=special>(!</span><span class=identifier>result</span><span class=special>)
// </span><span class=identifier>Python </span><span class=identifier>exception </span><span class=identifier>occurred
</span><span class=keyword>else
</span><span class=comment>// everything went okay, it's safe to use the result
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="embedding.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="iterators.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 Dirk Gerrits<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,122 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Virtual Functions with Default Implementations</title>
<link rel="stylesheet" href="theme/style.css" type="text/css">
<link rel="prev" href="deriving_a_python_class.html">
<link rel="next" href="class_operators_special_functions.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Virtual Functions with Default Implementations</b></font>
</td>
</tr>
</table>
<br>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="deriving_a_python_class.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_operators_special_functions.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<p>
Recall that in the <a href="class_virtual_functions.html">
previous section</a>, we
wrapped a class with a pure virtual function that we then implemented in
C++ or Python classes derived from it. Our base class:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Base
</span><span class=special>{
</span><span class=keyword>virtual </span><span class=keyword>int </span><span class=identifier>f</span><span class=special>() = </span><span class=number>0</span><span class=special>;
};
</span></pre></code>
<p>
had a pure virtual function <tt>f</tt>. If, however, its member function <tt>f</tt> was
not declared as pure virtual:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>Base
</span><span class=special>{
</span><span class=keyword>virtual </span><span class=keyword>int </span><span class=identifier>f</span><span class=special>() { </span><span class=keyword>return </span><span class=number>0</span><span class=special>; }
};
</span></pre></code>
<p>
and instead had a default implementation that returns <tt>0</tt>, as shown above,
we need to add a forwarding function that calls the <tt>Base</tt> default virtual
function <tt>f</tt> implementation:</p>
<code><pre>
<span class=keyword>struct </span><span class=identifier>BaseWrap </span><span class=special>: </span><span class=identifier>Base
</span><span class=special>{
</span><span class=identifier>BaseWrap</span><span class=special>(</span><span class=identifier>PyObject</span><span class=special>* </span><span class=identifier>self_</span><span class=special>)
: </span><span class=identifier>self</span><span class=special>(</span><span class=identifier>self_</span><span class=special>) {}
</span><span class=keyword>int </span><span class=identifier>f</span><span class=special>() { </span><span class=keyword>return </span><span class=identifier>call_method</span><span class=special>&lt;</span><span class=keyword>int</span><span class=special>&gt;(</span><span class=identifier>self</span><span class=special>, </span><span class=string>&quot;f&quot;</span><span class=special>); }
</span><span class=keyword>int </span><span class=identifier>default_f</span><span class=special>() { </span><span class=keyword>return </span><span class=identifier>Base</span><span class=special>::</span><span class=identifier>f</span><span class=special>(); } // &lt;&lt;=== ***</span><span class=identifier>ADDED</span><span class=special>***
</span><span class=identifier>PyObject</span><span class=special>* </span><span class=identifier>self</span><span class=special>;
};
</span></pre></code>
<p>
Then, Boost.Python needs to keep track of 1) the dispatch function <tt>f</tt> and
2) the forwarding function to its default implementation <tt>default_f</tt>.
There's a special <tt>def</tt> function for this purpose. Here's how it is
applied to our example above:</p>
<code><pre>
<span class=identifier>class_</span><span class=special>&lt;</span><span class=identifier>Base</span><span class=special>, </span><span class=identifier>BaseWrap</span><span class=special>&gt;(</span><span class=string>&quot;Base&quot;</span><span class=special>)
.</span><span class=identifier>def</span><span class=special>(</span><span class=string>&quot;f&quot;</span><span class=special>, &amp;</span><span class=identifier>Base</span><span class=special>::</span><span class=identifier>f</span><span class=special>, &amp;</span><span class=identifier>BaseWrap</span><span class=special>::</span><span class=identifier>default_f</span><span class=special>)
</span></pre></code>
<p>
Note that we are allowing <tt>Base</tt> objects to be instantiated this time,
unlike before where we specifically defined the <tt>class_&lt;Base&gt;</tt> with
<tt>no_init</tt>.</p>
<p>
In Python, the results would be as expected:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>base </span><span class=special>= </span><span class=identifier>Base</span><span class=special>()
&gt;&gt;&gt; </span><span class=keyword>class </span><span class=identifier>Derived</span><span class=special>(</span><span class=identifier>Base</span><span class=special>):
... </span><span class=identifier>def </span><span class=identifier>f</span><span class=special>(</span><span class=identifier>self</span><span class=special>):
... </span><span class=keyword>return </span><span class=number>42
</span><span class=special>...
&gt;&gt;&gt; </span><span class=identifier>derived </span><span class=special>= </span><span class=identifier>Derived</span><span class=special>()
</span></pre></code>
<p>
Calling <tt>base.f()</tt>:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>base</span><span class=special>.</span><span class=identifier>f</span><span class=special>()
</span><span class=number>0
</span></pre></code>
<p>
Calling <tt>derived.f()</tt>:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>derived</span><span class=special>.</span><span class=identifier>f</span><span class=special>()
</span><span class=number>42
</span></pre></code>
<p>
Calling <tt>call_f</tt>, passing in a <tt>base</tt> object:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>call_f</span><span class=special>(</span><span class=identifier>base</span><span class=special>)
</span><span class=number>0
</span></pre></code>
<p>
Calling <tt>call_f</tt>, passing in a <tt>derived</tt> object:</p>
<code><pre>
<span class=special>&gt;&gt;&gt; </span><span class=identifier>call_f</span><span class=special>(</span><span class=identifier>derived</span><span class=special>)
</span><span class=number>42
</span></pre></code>
<table border="0">
<tr>
<td width="30"><a href="../index.html"><img src="theme/u_arr.gif" border="0"></a></td>
<td width="30"><a href="deriving_a_python_class.html"><img src="theme/l_arr.gif" border="0"></a></td>
<td width="20"><a href="class_operators_special_functions.html"><img src="theme/r_arr.gif" border="0"></a></td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,156 +0,0 @@
<html>
<head>
<!-- Generated by the Spirit (http://spirit.sf.net) QuickDoc -->
<title>Boost Python Tutorial</title>
<link rel="stylesheet" href="doc/theme/style.css" type="text/css">
<link rel="next" href="quickstart.html">
</head>
<body>
<table width="100%" height="48" border="0" cellspacing="2">
<tr>
<td><img src="doc/theme/c%2B%2Bboost.gif">
</td>
<td width="85%">
<font size="6" face="Verdana, Arial, Helvetica, sans-serif"><b>Boost Python Tutorial</b></font>
</td>
</tr>
</table>
<br>
<table width="80%" border="0" align="center">
<tr>
<td class="toc_title">Table of contents</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/quickstart.html">QuickStart</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/building_hello_world.html">Building Hello World</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/exposing_classes.html">Exposing Classes</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/constructors.html">Constructors</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/class_data_members.html">Class Data Members</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/class_properties.html">Class Properties</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/inheritance.html">Inheritance</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/class_virtual_functions.html">Class Virtual Functions</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/deriving_a_python_class.html">Deriving a Python Class</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/virtual_functions_with_default_implementations.html">Virtual Functions with Default Implementations</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/class_operators_special_functions.html">Class Operators/Special Functions</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/functions.html">Functions</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/call_policies.html">Call Policies</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/overloading.html">Overloading</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/default_arguments.html">Default Arguments</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/auto_overloading.html">Auto-Overloading</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/object_interface.html">Object Interface</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/basic_interface.html">Basic Interface</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/derived_object_types.html">Derived Object types</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/extracting_c___objects.html">Extracting C++ objects</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/enums.html">Enums</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/embedding.html">Embedding</a>
</td>
</tr>
<tr>
<td class="toc_cells_L1">
<a href="doc/using_the_interpreter.html">Using the interpreter</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/iterators.html">Iterators</a>
</td>
</tr>
<tr>
<td class="toc_cells_L0">
<a href="doc/exception_translation.html">Exception Translation</a>
</td>
</tr>
</table>
<br>
<hr size="1"><p class="copyright">Copyright &copy; 2002-2003 David Abrahams<br>Copyright &copy; 2002-2003 Joel de Guzman<br><br>
<font size="2">Permission to copy, use, modify, sell and distribute this document
is granted provided this copyright notice appears in all copies. This document
is provided &quot;as is&quot; without express or implied warranty, and with
no claim as to its suitability for any purpose. </font> </p>
</body>
</html>

View File

@@ -1,163 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - April 2002 Progress Report</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">April 2002 Progress Report</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="index">
<dt><a href="#accomplishments">Accomplishments</a></dt>
<dl class="index">
<dt><a href="#arity">Arbitrary Arity Support</a></dt>
<dt><a href="#callbacks">New Callback Interface</a></dt>
<dt><a href="#policies">Call Policies for Construtors</a></dt>
<dt><a href="#bugs">Real Users, Real Bugs</a></dt>
<dt><a href="#insights">New Insights</a></dt>
<dt><a href="#v1">Boost.Python V1 Maintenance</a></dt>
</dl>
<dt><a href="#missing">What's Missing</a></dt>
</dl>
<h2><a name="accomplishments">Accomplishments</a></h2>
April was a short month as far as Boost.Python was concerned, since
the spring ISO C++ Committee Meeting (and associated vacation)
occupied me for the 2nd half of the month. However, a suprising amount
of work got done...
<h3><a name="arity">Arbitrary Arity Support</a></h3>
I began using the <a
href="../../../preprocessor/doc/index.htm">Boost.Preprocessor</a>
metaprogramming library to generate support for functions and member
functions of arbitrary arity, which was, to say the least, quite an
adventure. The feedback cycle resulting from my foray into
Boost.Preprocessor resulted in several improvements to the library,
most notably in its documentation.
<p>
Boost.Python now supports calls of up to 17 arguments on most
compilers. Because most EDG-based compilers have dismal preprocessor
performance, I had to &quot;manually&quot; expand the metaprograms for
arities from zero to fifteen arguments, and EDG-based compilers with
<code>__EDG_VERSION__&nbsp;&lt;=&nbsp;245</code> only support 15
arguments by default. If some crazy program finds a need for more than
the default arity support, users can increase the base support by
setting the <code>BOOST_PYTHON_MAX_ARITY</code> preprocessor symbol.
<h3><a name="callbacks">New Callback Interface</a></h3>
I mentioned in <a href="Mar2002.html">last month's report</a> that I
wasn't pleased with the interface for the interface for calling into
Python, so now it has been redesigned. The new interface is outlined
in <a
href="http://mail.python.org/pipermail/c++-sig/2002-April/000953.html">this
message</a> (though the GCC 2.95.3 bugs have been fixed).
<h3><a name="policies">Call Policies for Constructors</a></h3>
On April 2nd, I <a
href="http://mail.python.org/pipermail/c++-sig/2002-April/000916.html">announced</a>
support for the use of call policies with constructors.
<h3><a name="bugs">Real Users, Real Bugs</a></h3>
At least two people outside of Kull began actually using Boost.Python
v2 in earnest this month. Peter Bienstman and Pearu Pearson both
provided valuable real-world bug reports that helped me to improve the
library's robustness.
<h3><a name="insights">New Insights</a></h3>
<a
href="http://mail.python.org/pipermail/c++-sig/2002-May/001010.html"
>Answering some of Pearu's questions</a> about explicitly converting
objects between Python and C++ actually led me to a new understanding
of the role of the current conversion facilities. In Boost.Python v1,
all conversions between Python and C++ were handled by a single family
of functions, called <code>to_python()</code> and
<code>from_python()</code>. Since the primary role of Boost.Python is
to wrap C++ functions in Python, I used these names for the first kind
of converters I needed: those that extract C++ objects to be used as
function arguments and which C++ function return values to
Python. The better-considered approach in Boost.Python v2 uses a
completely different mechanism for conversions used when calling
Python from C++, as in wrapped virtual function implementations. I
usually think of this as a &quot;callback&quot;, as in &quot;calling
back into Python&quot;, and I named the converters used in callbacks
accordingly: <code>to_python_callback</code> and
<code>from_python_callback</code>. However, as it turns out, the
behavior of the &quot;callback&quot; converters is the appropriate one
for users who want to explicitly extract a C++ value from a Python
object, or create a Python object from a C++ value. The upshot is that
it probably makes sense to change the name of the existing <code>to_python</code> and
<code>from_python</code> so those names are available for the
user-friendly explicit converters.
<p>
<a
href="http://mail.python.org/pipermail/c++-sig/2002-May/001013.html">Another
of Pearu's questions</a> pushes momentum further in the direction of a
more-sophisticated overloading mechanism than the current
simple-minded &quot;first match&quot; approach, as I suggested <a
href="Mar2002.html#implicit_conversions">last month</a>.
<h3><a name="v1">Boost.Python V1 Maintenance</a></h3>
As much as I'm looking forward to retiring Boost.Python v1, a
significant amount of effort has been being spent dealing with support
problems; the saying that code rots when left alone is true, and
Boost.Python is no exception. Eventually it became obvious to me that
we were going to have to invest some effort in keeping V1 healthy
while working on V2. Ralf and I have expanded support for various
compilers and stabilized the V1 codebase considerably. We discarded
the obsolete Visual Studio projects which were causing so much
confusion. Still to do before the next Boost release:
<ol>
<li>Update the build/test documentation with detailed instructions for
configuring various toolsets.
<li>Provide some links to Boost.Python v2 to let people know what's
coming.
</ol>
<h2><a name="missing">What's Missing</a></h2>
Last month I announced that I would implement the following which are
not yet complete:
<ol>
<li>Document all implemented features
<li>Implement conversions for <code>char</code> types. This is
implemented but not tested, so we have to assume it doesn't work.
</ol>
These are my first priority for this month (especially the
documentation).
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a>
2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,153 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href=../../../../boost.css>
<title>Boost.Python - CallPolicies Concept</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">CallPolicies Concept</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#composition">CallPolicies Composition</a></dt>
<dt><a href="#concept-requirements">Concept Requirements</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#CallPolicies-concept">CallPolicies Concept</a></dt>
</dl>
</dd>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p>Models of the CallPolicies concept are used to specialize the behavior
of Python callable objects generated by Boost.Python to wrapped C++
objects like function and member function pointers, providing three
behaviors:</p>
<ol>
<li><code>precall</code> - Python argument tuple management before the
wrapped object is invoked</li>
<li><code>result_converter</code> - C++ return value handling</li>
<li><code>postcall</code> - Python argument tuple and result management
after the wrapped object is invoked</li>
</ol>
<h2><a name="composition"></a>CallPolicies Composition</h2>
In order to allow the use of multiple models of CallPolicies in the same
callable object, Boost.Python's CallPolicies class templates provide a
chaining interface which allows them to be recursively composed. This
interface takes the form of an optional template parameter,
<code>Base</code> which defaults to <a href=
"default_call_policies.html#default_call_policies-spec"><code>default_call_policies</code></a>.
By convention, the <code>precall</code> function of the <code>Base</code>
is invoked <i>after</i> the <code>precall</code> function supplied by the
outer template, and the <code>postcall</code> function of the
<code>Base</code> is invoked <i>before</i> the <code>postcall</code>
function of the outer template. If a <code>result_converter</code> is
supplied by the outer template, it <i>replaces</i> any
<code>result_converter</code> supplied by the <code>Base</code>. For an
example, see <a href=
"return_internal_reference.html#return_internal_reference-spec"><code>return_internal_reference</code></a>.
<h2><a name="concept-requirements"></a>Concept Requirements</h2>
<h3><a name="CallPolicies-concept"></a>CallPolicies Concept</h3>
<p>In the table below, <code><b>x</b></code> denotes an object whose type
<code><b>P</b></code> is a model of CallPolicies, <code><b>a</b></code>
denotes a <code>PyObject*</code> pointing to a Python argument tuple
object, and <code><b>r</b></code> denotes a <code>PyObject*</code>
referring to a "preliminary" result object.</p>
<table summary="CallPolicies expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Type</b></td>
<td><b>Result/Semantics</b></td>
</tr>
<tr>
<td valign="top"><code>x.precall(a)</code></td>
<td>convertible to <code>bool</code></td>
<td>returns <code>false</code> and <code><a href=
"http://www.python.org/doc/2.2/api/exceptionHandling.html#l2h-71">PyErr_Occurred</a>()&nbsp;!=&nbsp;0</code>
upon failure, <code>true</code> otherwise.</td>
</tr>
<tr>
<td valign="top"><code>P::result_converter</code></td>
<td>A model of <a href=
"ResultConverter.html#ResultConverterGenerator-concept">ResultConverterGenerator</a>.</td>
<td>An MPL unary <a href=
"../../../mpl/doc/paper/html/usage.html#metafunctions.classes">Metafunction
Class</a> used produce the "preliminary" result object.</td>
</tr>
<tr>
<td valign="top"><code>x.postcall(a, r)</code></td>
<td>convertible to <code>PyObject*</code></td>
<td>0 <code>0</code> and <code><a href=
"http://www.python.org/doc/2.2/api/exceptionHandling.html#l2h-71">PyErr_Occurred</a>()&nbsp;!=&nbsp;0</code>
upon failure. Must "conserve references" even in the event of an
exception. In other words, if <code>r</code> is not returned, its
reference count must be decremented; if another existing object is
returned, its reference count must be incremented.</td>
</tr>
</table>
Models of CallPolicies are required to be <a href=
"../../../utility/CopyConstructible.html">CopyConstructible</a>.
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
<p>Permission to copy, use, modify, sell and distribute this software is
granted provided this copyright notice appears in all copies. This
software is provided "as is" without express or implied warranty, and
with no claim as to its suitability for any purpose.</p>
</body>
</html>

View File

@@ -1,81 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href=../../../../boost.css>
<title>Boost.Python - Dereferenceable Concept</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Dereferenceable Concept</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#concept-requirements">Concept Requirements</a></dt>
<dl class="page-index">
<dt><a href="#Dereferenceable-concept">Dereferenceable Concept</a></dt>
</dl>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p>Instances of a dereferenceable type can be used like a pointer to access an lvalue.
<h2><a name="concept-requirements"></a>Concept Requirements</h2>
<h3><a name="Dereferenceable-concept"></a>Dereferenceable Concept</h3>
<p>In the table below, <code><b>x</b></code> denotes an object whose
type is a model of Dereferenceable.
<table summary="Dereferenceable expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Requirements</b></td>
</tr>
<tr>
<td valign="top"><code>*x</code></td>
<td>An lvalue
</tr>
</table>
If <code><b>x</b></code> is not a pointer type, it also must satsify the following expression:
<table summary="Dereferenceable expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Operational Semantics</b></td>
</tr>
<tr>
<td valign="top"><code>x.get()</code></td>
<td><code>&amp;*x</code>, or a null pointer
</tr>
</table>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
29 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i>
<p>Permission to copy, use, modify, sell
and distribute this software is granted provided this copyright notice appears
in all copies. This software is provided "as is" without express or implied
warranty, and with no claim as to its suitability for any purpose.
</body>
</html>

View File

@@ -1,93 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href=../../../../boost.css>
<title>Boost.Python - Extractor Concept</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Extractor Concept</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#concept-requirements">Concept Requirements</a></dt>
<dl class="page-index">
<dt><a href="#Extractor-concept">Extractor Concept</a></dt>
</dl>
<dt><a href="#notes">Notes</a></dt>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p>An Extractor is a class which Boost.Python can use to extract C++
objects from Python objects, and is typically used by facilities that
define <code>from_python</code> conversions for
&quot;traditional&quot; Python extension types.
<h2><a name="concept-requirements"></a>Concept Requirements</h2>
<h3><a name="Extractor-concept"></a>Extractor Concept</h3>
<p>In the table below, <code><b>X</b></code> denotes a model of
Extractor and <code><b>a</b></code> denotes an instance of a Python
object type.
<table summary="Extractor expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Type</b></td>
<td><b>Semantics</b></td>
</tr>
<tr>
<td valign="top"><code>X::execute(a)</code></td>
<td>non-void
<td>Returns the C++ object being extracted. The
<code>execute</code> function must not be overloaded.
</tr>
<tr>
<td valign="top"><code>&amp;a.ob_type</code>
<td><code><a
href="http://www.python.org/doc/2.2/ext/dnt-type-methods.html">PyTypeObject</a>**</code>
<td>Points to the <code>ob_type</code> field of an object which is
layout-compatible with <code>PyObject</code>
</tr>
</tr>
</table>
<h2><a name="notes"></a>Notes</h2>
Informally, an Extractor's <code>execute</code> member must be a
non-overloaded static function whose single argument is a Python
object type. Acceptable Python object types include those publicly (and
unambiguously) derived from <code>PyObject</code>, and POD types which
are layout-compatible with PyObject.
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i>
<p>Permission to copy, use, modify, sell
and distribute this software is granted provided this copyright notice appears
in all copies. This software is provided "as is" without express or implied
warranty, and with no claim as to its suitability for any purpose.
</body>
</html>

View File

@@ -1,71 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href=../../../../boost.css>
<title>Boost.Python - Holder Concept</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">HolderGenerator Concept</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#concept-requirements">Concept Requirements</a></dt>
<dl class="page-index">
<dt><a href="#HolderGenerator-concept">HolderGenerator Concept</a></dt>
</dl>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p>A HolderGenerator is a unary metafunction class which returns types
suitable for holding instances of its argument in a wrapped C++ class
instance.
<h2><a name="concept-requirements"></a>Concept Requirements</h2>
<h3><a name="HolderGenerator-concept"></a>HolderGenerator Concept</h3>
<p>In the table below, <code><b>G</b></code> denotes an type which
models HolderGenerator, and <code><b>X</b></code> denotes a class
type.
<table summary="Holder expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Requirements</b></td>
</tr>
<tr>
<td valign="top"><code>G::apply&lt;X&gt;::type</code></td>
<td>A concrete subclass of <a
href="instance_holder.html#instance_holder-spec">instance_holder</a>
which can hold objects of type <code>X</code>.
</tr>
</table>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i>
<p>Permission to copy, use, modify, sell
and distribute this software is granted provided this copyright notice appears
in all copies. This software is provided "as is" without express or implied
warranty, and with no claim as to its suitability for any purpose.
</body>
</html>

View File

@@ -1,226 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - June 2002 Progress Report</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">June 2002 Progress Report</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="index">
<dt><a href="#intro">Introduction</a></dt>
<dt><a href="#handle"><code>handle&lt;T&gt;</code></a></dt>
<dt><a href="#object"><code>object</code></a></dt>
<dl class="index">
<dt><a href="#operators"><code>object</code> operators</a></dt>
<dt><a href="#conversions"><code>object</code> conversions</a></dt>
</dl>
<dt><a href="#list"><code>list</code></a></dt>
<dt><a href="#numerics"><code>Numerics</code></a></dt>
<dt><a href="#community">Community</a></dt>
<dt><a href="#next">What's Next</a></dt>
</dl>
<h2><a name="intro">Introduction</a></h2>
July was mostly focused on allowing expressive manipulation of
individual Python objects, or what Ralf Grosse-Kunstleve calls
&quot;Writing Python in C++&quot;. The work began with this <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001311.html">posting</a>,
which outlines the issues and intention.
<h2><a name="handle"><code>handle&lt;T&gt;</code></a></h2>
The most basic element needed was a replacement for the
<code>reference&lt;&gt;</code> class template and the
<code>ref</code> typedef from Boost.Python v1, a simple smart
pointer to a Python object. The old v1 typedef
&quot;<code>ref</code>&quot; (for
<code>reference&lt;PyObject&gt;</code>) had to be retired because I
thought it would be too confusing given the importance of <code><a
href="../../../bind/ref.html">boost::ref</a>()</code> to this
library. I began a <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001311.html">discussion</a>of
possible names, and it was eventually <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001337.html">decided</a>
to rename <code>reference</code> to <code>handle</code> and supply a
default argument so that <code>ref</code> could be spelled
<code>handle&lt;&gt;</code> without an additional typedef. There
were also some interface changes to make it safer and more-efficient
to interface with the raw
<code>PyObject*</code>s forced on us by Python's 'C' API. A
discussion of those protocols can be found <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001401.html">here</a>.
<h2><a name="handle"><code>object</code></a></h2>
It is intended that users will seldom need or want to work with
<code>handle&lt;&gt;</code>; its major distinguishing features are
that it gives direct access to the underlying object representation
through <code>operator*</code> and <code>operator-&gt;</code>, and
that can be <code>NULL</code>, both sources of danger. Instead the
library provides a class called <code>object</code>, which
encapsulates a valid Python object and provides a similar interface to
Python's.
<h3><a name="operators"><code>object</code> operators</a></h3>
The first challenge was to provide support for object manipulations
using a Python-like syntax, mostly in the form of operator overloads:
<table border="1">
<tr><th>Python <th>C++
<tr>
<td><code>y = x.foo</code> <td><code>y = x.attr(&quot;foo&quot;);
<tr>
<td><code>x.foo = 1</code> <td><code>x.attr(&quot;foo&quot;) = 1;
<tr>
<td><code>y = x[z]</code> <td><code>y = x[z];
<tr>
<td><code>x[z] = 1</code> <td><code>x[z] = 1;
<tr>
<td><code>y = x[3:-1]</code> <td><code>y = x.slice(3,-1);
<tr>
<td><code>y = x[3:]</code> <td><code>y = x.slice(3,_);
<tr>
<td><code>y = x[:-2]</code> <td><code>y = x.slice(_,-2);
<tr>
<td><code>z = x(1, y)</code> <td><code>z = x(1, y);
<tr>
<td><code>z = x.f(1, y)</code> <td><code>z = x.attr(&quot;f&quot;)(1, y);
<tr>
<td><code>not x</code> <td><code>!x
<tr>
<td><code>x and y</code> <td><code>x and y
</table>
I'm still a unsatisfied with the interface for attribute access. There
original proposal used a syntax like this one:
<pre>
y = x._(&quot;foo&quot;);
x._(&quot;foo&quot;) = 1;
</pre>
which was only marginally better than what we've got. Niki Spahiev
then <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001447.html">pointed
out</a> a potential conflict with the macro which GNU Gettext <a
href="http://www.gnu.org/manual/gettext/html_mono/gettext.html#SEC6">suggests</a>
people define. This unfortunate state of affairs forced us into using
<code>attr</code> instead. I'd still like to find a better interface,
but the lack of overloadable C++ operators which aren't already used
in Python is an obstacle. The comma operator is still a possibility,
but it has the wrong precedence:
<pre>
y = x,&quot;foo&quot; // error
x,&quot;foo&quot; = 1; // error
y = (x,&quot;foo&quot;); // ok
(x,&quot;foo&quot;) = 1; // ok
</pre>
Well, I guess we could consider adding that to the interface without
removing <code>attr()</code>, to see how it plays out...
<h3><a name="operators"><code>object</code> conversions</a></h3>
The <code>object</code> class also provided an opportunity to replace
Boost.Python v1's <code>to_python()</code> as a user-level
interface. Instead, <code>object</code> has a templated constructor
which can be used to convert any C++ object to Python using the same
underlying mechanisms used for the arguments to <code><a
href="call.html">call</a>&lt;&gt;</code>.
<p>Incidentally, the implementation of operator and conversion support
for object uncovered an inordinate number of compiler bugs in our
targeted platforms. It was a lot more &quot;interesting&quot; than it
should have been.
<h2><a name="list"><code>list</code></a></h2>
With <code>object</code> implemented, it was time to begin replacing
the ad-hoc implementations of <code>list</code>, <code>string</code>,
and <code>dictionary</code> supplied by Boost.Python v1 with something
more robust. I started with <code>list</code> as an example. Because
<code>object</code> already provides all of the requisite operators,
publicly deriving <code>list</code> from object seemed like a good
choice. The remaining issues were what do do about the one-argument
list constructor (which in Python attempts to convert its argument to
a list), and how to deal converting with <code>list</code> arguments
to wrapped functions. Some of the issues are laid out in <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001551.html">this
thread</a>. Ultimately, it was decided that <code>list(x)</code>
should do the same thing in C++ as in Python (conversion), while
<code>list</code> arguments should only match Python
<code>list</code>s (and <code>list</code> subclasses). The
implementation worked well, and provided a <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001586.html">roadmap</a>
for the protocol to be used for implementation of the other built-in
types.
<h2><a name="numerics">Numerics</a></h2>
Support for C++ <code>long long</code> and <code>unsigned long
long</code>
(and <code>__int64</code> on MSVC) to/from python conversions was
added this month. We also improved handling of numeric overflows when
converting, e.g., a Python int to a type with a more limited range of
representation.
<h2><a name="community">Community</a></h2>
<ul>
<li>Ralf W. Grosse-Kunstleve and Nick Sauter have implemented
<a href="http://cci.lbl.gov/boost/">multiplatform nightly
build-and-test</a> runs for Boost.Python V2 at LBL.
<li>Dave Hawkes has made significant progress on generating the
Python <a
href="http://mail.python.org/pipermail/c++-sig/2002-June/001503.html">built-in
function and API wrappers</a>
<li>Achim Domma has agreed to take up the job of implementing the
<code>str</code>, <code>dict</code>, and <code>tuple</code> classes.
</ul>
Deep thanks to all the Boost.Python contributors! This project
wouldn't be possible without your participation.
<h2><a name="next">What's Next</a></h2>
As I write this we are already well into the month of July, so I
suggest you consult the <a
href="http://mail.python.org/pipermail/c++-sig/2002-July/">Mailing
List Archive</a> if you want to know what's been happening. Otherwise
you'll just have to wait till next month (hopefully the beginning).
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a>
2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,234 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - March 2002 Progress Report</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">March 2002 Progress Report</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="index">
<dt><a href="#accomplishments">Accomplishments</a></dt>
<dl class="index">
<dt><a href="#calling_python">Calling Python from C++</a></dt>
<dt><a href="#virtual_functions">Virtual Functions</a></dt>
<dt><a href="#abstract_classes">Abstract Classes</a></dt>
<dt><a href="#implicit_conversions">C++ Implicit Conversions</a></dt>
<dt><a href="#data_members">C++ Data Members</a></dt>
<dt><a href="#miscellaneous">Miscellaneous</a></dt>
</dl>
<dt><a href="#future">The Near future</a></dt>
<dt><a href="#notes">Notes</a></dt>
</dl>
<h2><a name="accomplishments">Accomplishments</a></h2>
March was mostly devoted to the reimplementation of features from
Boost.Python v1, and some new features. Re-examination of the features
from Boost.Python v1 allowed me to make significant improvements.
<h3><a name="calling_python">Calling Python from C++</a></h3>
The ability to call Python from C++ is crucial for virtual function
support. Implementing this feature well for V2 proved to be more
interesting than I expected. You can review most of the relevant
design decisions
<a href="callbacks.txt">here</a>.
<p>
One point which <i>isn't</i> emphasized in that document is that there
are subtle differences in the way <code>from_python</code> conversions
work when used for C++ function arguments and Python function return
values. In particular, while <code>T const&amp;</code> arguments may
invoke rvalue converters, a reference-to-const return value requires
an lvalue converter, since a temporary conversion result would leave
the returned reference dangling.
<p>I'm not particularly pleased with the current callback interface,
since it usually results in constructs like:
<pre>
<u>return returning</u>&lt;X&amp;&gt;::call(f, obj);
</pre>
However, I think the following may be possible and I plan to investigate:
<pre>
return apply&lt;X&amp;&gt;(f, obj);
</pre>
I'm open to suggestion for better names (and syntaxes)!
<h3><a name="virtual_functions">Virtual Functions</a></h3>
Once Python callbacks were implemented, it was just a short step to
implementing virtual functions. Python extension class exposing a C++
class whose virtual functions are overridable in Python must actually
hold a C++ instance of a class <i>derived</i> from the one exposed to
Python. Needing some way for users to specify that class, I added an
optional template argument to <code>value_holder_generator</code> and
<code>pointer_holder_generator&lt;&gt;</code> to specify the class
actually held. This move began to put pressure on the
<code>class_&lt;&gt;</code> interface, since the need for the user to
produce complicated instantations of
<code>class_&lt;&gt;</code> was increased:
<pre>
class&lt;Foo, bases&lt;&gt;, value_holder_generator&lt;Foo_callback&gt; &gt;(&quot;Foo&quot;)
.def(&quot;hello&quot;, &amp;Foo::hello)
...
</pre>
<h3><a name="abstract_classes">Abstract Classes</a></h3>
Normally when a C++ class is exposed to Python, the library registers
a conversion function which allows users to wrap functions returning
values of that type. Naturally, these return values are temporaries,
so the conversion function must make a copy in some
dynamically-allocated storage (a &quot;holder&quot;) which is managed
by the corresponding Python object.
<p>Unfortunately, in the case of abstract classes (and other types
without a publicly-accessible copy constructor), instantiating this
conversion function causes a compilation error. In order to support
non-copyable classes, there had to be some way to prevent the library
from trying to instantiate the conversion function. The only practical
approach I could think of was to add an additional template parameter
to the <code>class_&lt;&gt;</code> interface. When the number of
template parameters with useful defaults begins to grow, it is often
hard to choose an order which allows users to take advantage of the
defaults.
<p>
This was the straw that broke the
<code>class_&lt;&gt;</code> interface's back and caused the redesign
whose outcome is detailed <a
href="http://mail.python.org/pipermail/c++-sig/2002-March/000892.html">here</a>.
The approach allows the user to supply the optional parameters in an
arbitrary order. It was inspired by the use of <a
href="../../../utility/iterator_adaptors.htm#named_tempalte_parameters">named
template parameters</a> in the <a
href="../../../utility/iterator_adaptors.htm">Boost Iterator Adaptor
Library</a>, though in this case it is possible to deduce the meaning
of the template parameters entirely from their type properties,
resulting in a simpler interface. Although the move from a
policy-based design to what resembles a configuration DSL usually
implies a loss of flexibility, in this case I think any costs are far
outweighed by the advantages.
<p>Note: working around the limitations of the various compilers I'm
supporting was non-trivial, and resulted in a few messy implementation
details. It might be a good idea to switch to a more-straightforward
approach once Metrowerks CodeWarrior Pro8 is released.
<h3><a name="implicit_conversions">C++ Implicit Conversions</a></h3>
Support for C++ implicit conversion involves creating
<code>from_python</code> converters for a type <code>U</code> which in
turn use <code>from_python</code> converters registered for a type
<code>T</code> where there exists a implicit conversion from
<code>T</code> to <code>U</code>. The current implementation is
subject to two inefficiencies:
<ol>
<li>Because an rvalue <code>from_python</code> converter produces two
pieces of data (a function and a <code>void*</code>) from its
<code>convertible()</code> function, we end up calling the function
for <code>T</code> twice: once when the converter is looked up in the
registry, and again when the conversion is actually performed.
<li>A vector is used to mark the "visited" converters, preventing
infinite recursion as <code>T</code> to
<code>U</code> and <code>U</code> to <code>T</code> converters
continually search through one-another.
</ol>
I consider the former to be a minor issue. The second may or may not
prove to be computationally significant, but I believe that
architecturally, it points toward a need for more sophisticated
overload resolution. It may be that we want CLOS-style multimethod
dispatching along with C++ style rules that prevent more than one
implicit conversion per argument.
<h3><a name="data_members">C++ Data Members</a></h3>
To supply the ability to directly access data members, I was able to
hijack the new Python <a
href="http://www.python.org/2.2/descrintro.html#property">property</a>
type. I had hoped that I would also be able to re-use the work of <a
href="make_function.html">make_function</a> to create callable python
objects from C++ functions which access a data member of a given
class. C++ facilities for specifying data member pointer non-type
template arguments require the user to explicitly specify the type of
the data member and this under-utilized feature is also not
well-implemented on all compilers, so passing the member pointer as a
runtime value is the only practical approach. The upshot is that any
such entity would actually have to be a function <i>object</i>, and I
haven't implemented automatic wrapping of C++ callable function
objects yet, so there is less re-use in the implementation than I'd
like. I hope to implement callable object wrapping and refactor this
code one day. I also hope to implement static data member support,
for which Python's property will not be an appropriate descriptor.
<h3><a name="miscellaneous">Miscellaneous</a></h3>
<ul>
<li>Moved <code>args&lt;&gt;</code> and <code>bases&lt;&gt;</code> from unnamed namespace to <code>boost::python</code> in their own header files.
<li>Convert <code>NULL</code> pointers returned from wrapped C++ functions to <code>None</code>.
<li>Improved some compile-time error checks.
<li>Eliminated <code>boost/python/detail/eval.hpp</code> in favor of
more-general <code>boost/mpl/apply.hpp</code>.
<li>General code cleanup and refactoring.
<li>Works with Microsoft Visual C++ 7.0
<li>Warning suppression for many compilers
<li>Elegant interface design for exporting <code>enum</code> types.
</ul>
<hr>
<h2><a name="future">The Near Future</a></h2>
Before April 15th I plan to
<ol>
<li>Document all implemented features
<li>Implement a <code>CallPolicy</code> interface for constructors of wrapped
classes
<li>Implement conversions for <code>char</code> types.
<li>Implement automated code generation for all headers containing
families of overloaded functions to handle arbitrary arity.
</ol>
I also hope to implement a mechanism for generating conversions
between arbitrary Python sequences and C++ containers, if time permits
(and others haven't already done it)!
<h2><a name="notes">Notes</a></h2>
The older version of KCC used by Kull is generating lots of warnings
about a construct I use to instantiate static members of various class
templates. I'm thinking of moving to an idiom which uses a function
template to suppress it, but worry about bloating the size of debug
builds. Since KCC users may be moving to GCC, I'm not sure that it's
worth doing anything about it.
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a>
2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,309 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - May 2002 Progress Report</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">May 2002 Progress Report</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="index">
<dt><a href="#intro">Introduction</a></dt>
<dt><a href="#features">New Features</a></dt>
<dl>
<dt><a href="#aix_shared">Shared Library Support for AIX</a><dd>
<dt><a href="#class_enhancements">Class Enhancements</a><dd>
<dl>
<dt><a href="#operators">Operators</a><dd>
<dt><a href="#iterators">Iterators</a><dd>
<dt><a href="#properties">Properties</a><dd>
<dt><a href="#setattr">setattr</a><dd>
<dt><a href="#module">__module__ Attribute</a><dd>
</dl>
<dt><a href="#back_reference">back_reference</a><dd>
</dl>
<dt><a href="#documentation">Documentation</a></dt>
<dt><a href="#misc">Miscellaneous</a></dt>
<dl class="index">
<dt><a href="#converters">Converters</a></dt>
<dt><a href="#checkins">Checkins Mailing List</a></dt>
<dt><a href="#shared">Shared Libraries</a></dt>
</dl>
<dt><a href="#next">What's Next</a></dt>
</dl>
<h2><a name="intro">Introduction</a></h2>
Aside from library development, work on Boost.Python in May was
focused on reducing the support burden. In recent weeks, responding to
requests for support, espcially surrounding building the library, had
begun to impede progress on development. There was a major push to
release a stable 1.28.0 of Boost, including documentation of <a
href="../../../../tools/build/index.html">Boost.Build</a> and specific
<a href="../building.html">instructions</a> for building Boost.Python
v1. The documentation for Boost.Python v2 was also updated as
described <a href="#documentation">here</a>.
<h2><a name="features">New Features</a></h2>
<h3><a name="aix_shared">Shared Library Support for AIX</a></h3>
The Kull group required the ability to build and test Boost.Python
extensions on AIX, a platform with &quot;creatively designed&quot;
shared library semantics. Making this work was a multi-pronged
effort, involving changes to Boost.Build and some great research by
Martin Casado which uncovered the key mechanism required to allow
shared libraries to use functions from the Python executable. The
current solution used in Boost.Build relies on a <a
href="../../../../tools/build/gen_aix_import_file.py">Python
Script</a> as part of the build process. This is not a problem for
Boost.Python, as Python will be available. However, the commands
issued by the script are so simple that a 100%-pure-Boost.Jam
solution is surely possible. Linking on AIX is sufficiently
interesting to have skewed the Boost.Python development schedule a
bit.
<h3><a name="class_enhancements">Class Enhancements</a></h3>
<h4><a name="operators">Operators</a></h4>
Support for exposing C++ operators and functions as the corresponding
Python special methods was added. Thinking that the Boost.Python
<a href="../special.html#numeric">v1 interface</a> was a little too
esoteric (especially the use of
<code>left_operand&lt;...&gt;/right_operand&lt;...&gt;</code> for
asymmetric operands), I introduced a simple form of <a
href="http://osl.iu.edu/~tveldhui/papers/Expression-Templates/exprtmpl.html">expression
templates</a> which allow users to simply write the expressions that
should be wrapped, as in this <a href="operators.html#examples">example</a>.
<h4><a name="iterators">Iterators</a></h4>
Python iterator support as required by the Kull project resulted in a
highly flexible interface allowing:
<dl>
<dt>Direct exposure of a class' <code>begin()</code> and
<code>end()</code> functions:
<pre>
...
.def(&quot;__iter__&quot;, iterator&lt;list_int&gt;())
</pre>
<dd>
<dt>Creation of iterators from member functions...
<pre>
...
.def(&quot;__iter__&quot;
, range(&amp;my_class::x_begin, &amp;my_class::x_end))
)
</pre>
<dd>
<dt>...and member data:
<pre>
...
.def(&quot;__iter__&quot;
, range(&amp;std::pair&lt;char*,char*&gt;::first, &amp;std::pair&lt;char*,char*&gt;::second))
)
</pre>
<dd>
<dt>The ability to specify <a
href="CallPolicies.html">CallPolicies</a>, e.g. to prevent copying of
heavyweight values:
<pre>
...
.def(&quot;__iter__&quot;,
, range&lt;return_value_policy&lt;copy_non_const_reference&gt; &gt;(
&amp;my_sequence&lt;heavy&gt;::begin
, &amp;my_sequence&lt;heavy&gt;::end))
</pre>
<dd>
</dl>
<h4><a name="properties">Properties</a></h4>
The Kull iteration interfaces also required the ability to iterate
over a sequence specified by an instance's attribute:
<pre>
&gt;&gt;&gt; f = field()
&gt;&gt;&gt; for e in f.elements:
... print e,
</pre>
This forced the exposure of the <a
href="http://www.python.org/2.2/descrintro.html#property"><code>property</code></a>
interface used internally to implement the data member exposure
facility described in <a
href="Mar2002#data_members">March</a>. Properties are an
incredibly useful idiom, so it's good to be able to provide them
at little new development cost.
<h4><a name="setattr">setattr</a></h4>
<code>class_&lt;&gt;</code> acquired a <code>setattr</code> member
function which allows users to easily add new Python objects as class
attributes.
<h4><a name="module">__module__ Attribute</a></h4>
Ralf Grosse-Kunstleve has been working on pickling support for v2. To
make it work correctly, he had to make sure that a class'
<code>__module__</code> attribute was set correctly.
<h3><a name="back_reference"><code>back_reference</code></a></h3>
The new <code>back_reference&lt;T&gt;</code> template can be used as a
function parameter when the user needs access to both a <code>T</code>
argument and to the Python object which manages it. The function will
only match in the overload resolution process if it would match the
same function signature with <code>T</code> substituted for
<code>back_reference&lt;T&gt;</code>. This feature is not yet
documented.
<h2><a name="documentation">Documentation</a></h2>
In a major effort to prepare Boost.Python v2 to replace v1, many pages
of new reference documentation were added:
<blockquote>
<dl>
<dt><a href="CallPolicies.html">CallPolicies.html</a><dd>
<dt><a href="Dereferenceable.html">Dereferenceable.html</a><dd>
<dt><a href="Extractor.html">Extractor.html</a><dd>
<dt><a href="HolderGenerator.html">HolderGenerator.html</a><dd>
<dt><a href="ResultConverter.html">ResultConverter.html</a><dd>
<dt><a href="call_method.html">call_method.html</a><dd>
<dt><a href="callbacks.html">callbacks.html</a><dd>
<dt><a href="data_members.html">data_members.html</a><dd>
<dt><a href="has_back_reference.html">has_back_reference.html</a><dd>
<dt><a href="implicit.html">implicit.html</a><dd>
<dt><a href="instance_holder.html">instance_holder.html</a><dd>
<dt><a href="operators.html">operators.html</a><dd>
<dt><a href="ptr.html">ptr.html</a><dd>
<dt><a href="type_id.html">type_id.html</a><dd>
<dt><a href="with_custodian_and_ward.html">with_custodian_and_ward.html</a><dd>
</dl>
</blockquote>
Major updates were made to the following pages:
<blockquote>
<dl>
<dt><a href="call.html">call.html</a><dd> <dt><a href="updated">updated</a><dd>
<dt><a href="class.html">class.html</a><dd>
<dt><a href="reference.html">reference.html</a><dd>
</dl>
</blockquote>
As usual, careful documentation forces one to consider the
interface again, and there were many interface changes
associated with this effort, including the elevation of the
following components from implementation detail to
first-class library citizen:
<blockquote>
<dl>
<dt>type_id.hpp<dd>
<dt>pointee.hpp<dd>
<dt>lvalue_from_pytype.hpp<dd></dl>
</dl>
</blockquote>
<h2><a name="misc">Miscellaneous</a></h2>
<h3><a name="converters">Converters</a></h3>
It appears that the world of C++ &lt;==&gt; Python conversion rules is
an endlessly-rich area of exploration. Completing the conversions for
<code>char</code> and <code>char const*</code> types, as described at
the end of <a href="Apr2002.html#missing">April's report</a>,
uncovered some interesting new shades to the problem. It turns out to
be worth distinguishing mutable and immutable lvalue conversions,
because despite the fact that Python doesn't understand
<code>const</code>, it does understand immutability (c.f. Python
strings, which expose an immutable <code>char</code> pointer). It is
also worth recognizing types which represent lvalue <i>sequences</i>,
to prevent Python <code>&quot;foobar&quot;</code> from being silently
truncated to C++ <code>'f'</code>. More details on this insight can be
found in the mailing list <a
href="http://mail.python.org/pipermail/c++-sig/2002-May/001023.html">
archive</a>. I don't plan to do anything about this immediately, but I
do think it's the right direction to go in the long run.
<h3><a name="checkins">Checkins Mailing List</a></h3>
In order to better coordinate changes made by multiple developers, I
enabled <a
href="http://sourceforge.net/docman/display_doc.php?docid=772&group_id=1">syncmail</a>
for the Boost.Python CVS trees, and established an associated <a
href="http://lists.sourceforge.net/lists/listinfo/boost-python-cvs">mailing
list</a>. Subscribe to this list to receive notices of each new
checkin.
<h3><a name="shared">Shared Libraries</a></h3>
Beyond the vagaries of dynamic linking on AIX, I have been
participating in a more-general discussion of dynamic linking for
C++. Needless to say, C++ dynamic linking is of critical importance to
Boost.Python: all extension modules are normally built as shared
libraries, and Boost.Python extension modules share a common library
as well.
In fact, there are at least two separate conversations. One
in the C++ standard extensions mailing list concerns what can be
standardized for C++ and shared libraries; the other, mostly on the <a
href="http://gcc.gnu.org/ml/gcc/">gcc</a> mailing list, concerns the
behavior of GCC on Posix/ELF platforms.
Some of the GCC threads are here:
<blockquote>
<a
href="http://gcc.gnu.org/ml/gcc/2002-05/msg02002.html">http://gcc.gnu.org/ml/gcc/2002-05/msg02002.html</a><br>
<a
href="http://gcc.gnu.org/ml/gcc/2002-05/msg02945.html">http://gcc.gnu.org/ml/gcc/2002-05/msg02945.html</a><br>
<a href="http://gcc.gnu.org/ml/gcc/2002-05/msg01758.html">http://gcc.gnu.org/ml/gcc/2002-05/msg01758.html</a>
</blockquote>
<h2><a name="next">What's Next</a></h2>
Development is focused on what's needed to be able to retire
Boost.Python v1. At the moment, that means deciding the user-friendly
interfaces for to_/from_python conversion, and formally exposing the
Python object smart pointers and object wrapper classes. Quite a few
questions have also been showing up recently about how to embed Python
with Boost.Python, and how to link with it statically; the solutions
to these issues will probably have to be formalized before long.
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a>
2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,156 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - ObjectWrapper Concept</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">ObjectWrapper and TypeWrapper Concepts</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#concept-requirements">Concept Requirements</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#ObjectWrapper-concept">ObjectWrapper Concept</a></dt>
<dt><a href="#TypeWrapper-concept">TypeWrapper Concept</a></dt>
</dl>
</dd>
<dt><a href="#caveat">Caveat</a></dt>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p>This page defines two concepts used to describe classes which manage a
Python objects, and which are intended to support usage with a
Python-like syntax.</p>
<h2><a name="concept-requirements"></a>Concept Requirements</h2>
<h3><a name="ObjectWrapper-concept"></a>ObjectWrapper Concept</h3>
Models of the ObjectWrapper concept have <a href=
"object.html#object-spec">object</a> as a publicly-accessible base class,
and are used to supply special construction behavior and/or additional
convenient functionality through (often templated) member functions.
Except when the return type <code>R</code> is itself an <a href=
"#TypeWrapper-concept">TypeWrapper</a>, a member function invocation of
the form
<pre>
x.<i>some_function</i>(<i>a<small>1</small>, a<small>2</small>,...a<small>n</small></i>)
</pre>
always has semantics equivalent to:
<pre>
<a href=
"extract.html#extract-spec">extract</a>&lt;R&gt;(x.attr("<i>some_function</i>")(<a
href=
"object.html#object-spec-ctors">object</a>(<i>a<small>1</small></i>), <a
href=
"object.html#object-spec-ctors">object</a>(<i>a<small>2</small></i>),...<a
href="object.html#object-spec-ctors">object</a>(<i>a<small>n</small></i>)))()
</pre>
When the <code>R</code> is an <a href=
"#TypeWrapper-concept">TypeWrapper</a>, the result type may be
constructed by taking direct posession of:
<pre>
x.attr("<i>some_function</i>")(<a href=
"object.html#object-spec-ctors">object</a>(<i>a<small>1</small></i>), <a
href=
"object.html#object-spec-ctors">object</a>(<i>a<small>2</small></i>),...<a
href=
"object.html#object-spec-ctors">object</a>(<i>a<small>n</small></i>)).ptr()
</pre>
[see <a href="#caveat">caveat</a> below]
<h3><a name="TypeWrapper-concept"></a>TypeWrapper Concept</h3>
TypeWrapper is a refinement of ObjectWrapper which is associated with a
particular Python type <code>X</code>. For a given TypeWrapper
<code>T</code>, a valid constructor expression
<pre>
T(<i>a<small>1</small>, a<small>2</small>,...a<small>n</small></i>)
</pre>
builds a new <code>T</code> object managing the result of invoking
<code>X</code> with arguments corresponding to
<pre>
<a href=
"object.html#object-spec-ctors">object</a>(<i>a<small>1</small></i>), <a
href=
"object.html#object-spec-ctors">object</a>(<i>a<small>2</small></i>),...<a
href=
"object.html#object-spec-ctors">object</a>(<i>a<small>n</small></i>)
</pre>
When used as arguments to wrapped C++ functions, or as the template
parameter to <code><a
href="extract.html#extract-spec">extract</a>&lt;&gt;</code>, only
instances of the associated Python type will be considered a match.
<h3><a name="caveat">Caveat</a></h3>
The upshot of the special member function invocation rules when the
return type is a TypeWrapper is that it is possible for the returned
object to manage a Python object of an inappropriate type. This is not
usually a serious problem; the worst-case result is that errors will be
detected at runtime a little later than they might otherwise be. For an
example of how this can occur, note that the <code><a href=
"dict.html#dict-spec">dict</a></code> member function <code>items</code>
returns an object of type <code><a href=
"list.html#list-spec">list</a></code>. Now suppose the user defines this
<code>dict</code> subclass in Python:
<pre>
&gt;&gt;&gt; class mydict(dict):
... def items(self):
... return tuple(dict.items(self)) # return a tuple
</pre>
Since an instance of <code>mydict</code> is also an instance of
<code>dict</code>, when used as an argument to a wrapped C++ function,
<code><a href="dict.html#dict-spec">boost::python::dict</a></code> can
accept objects of Python type <code>mydict</code>. Invoking
<code>items()</code> on this object can result in an instance of <code><a
href="list.html#list-spec">boost::python::list</a></code> which actually
holds a Python tuple. Subsequent attempts to use list methods (e.g.
<code>append</code>, or any other mutating operation) on this object will
raise the same exception that would occur if you tried to do it from
Python.
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
<p>Permission to copy, use, modify, sell and distribute this software is
granted provided this copyright notice appears in all copies. This
software is provided "as is" without express or implied warranty, and
with no claim as to its suitability for any purpose.</p>
</body>
</html>

View File

@@ -1,110 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href=../../../../boost.css>
<title>Boost.Python - ResultConverter Concept</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">ResultConverter Concept</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#concept-requirements">Concept Requirements</a></dt>
<dl class="page-index">
<dt><a href="#ResultConverter-concept">ResultConverter Concept</a></dt>
<dt><a href="#ResultConverterGenerator-concept">ResultConverterGenerator Concept</a></dt>
</dl>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p>A ResultConverter for a type <code>T</code> is a type whose
instances can be used to convert C++ return values of type
<code>T</code> <code>to_python</code>. A ResultConverterGenerator is
an MPL unary metafunction class which, given the return type of a C++
function, returns a ResultConverter for that type. ResultConverters in
Boost.Python generally inspect library's registry of converters to
find a suitable converter, but converters which don't use the registry
are also possible.
<h2><a name="concept-requirements"></a>Concept Requirements</h2>
<h3><a name="ResultConverter-concept"></a>ResultConverter Concept</h3>
<p>In the table below, <code><b>C</b></code> denotes a ResultConverter
type for a type <b><code>R</code></b> , <code><b>c</b></code> denotes
an object of type <code><b>C</b></code> , and <code><b>r</b></code>
denotes an object of type <code><b>R</b></code>.
<table summary="ResultConverter expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Type</b></td>
<td><b>Semantics</b></td>
</tr>
<tr>
<td valign="top"><code>C c;</code></td>
<td>
<td>Constructs a <code>C</code> object.
</tr>
<tr>
<td valign="top"><code>c.convertible()</code></td>
<td>convertible to <code>bool</code></td>
<td><code>false</code> iff no conversion from any <code>R</code> value
to a Python object is possible.</td>
</tr>
<tr>
<td valign="top"><code>c(r)</code></td>
<td>convertible to <code>PyObject*</code></td>
<td>A pointer to a Python object corresponding to <code>r</code>,
or <code>0</code> iff <code>r</code> could not be converted
<code>to_python</code>, in which case <a
href="http://www.python.org/doc/current/api/exceptionHandling.html#l2h-71">PyErr_Occurred</a>
should return non-zero.</td>
</tr>
</table>
<h3><a name="ResultConverterGenerator-concept"></a>ResultConverterGenerator Concept</h3>
<p>In the table below, <code><b>G</b></code> denotes a
ResultConverterGenerator type and <code><b>R</b></code> denotes a possible
C++ function return type.
<table summary="ResultConverterGenerator expressions" border="1" cellpadding="5">
<tr>
<td><b>Expression</b></td>
<td><b>Requirements</b></td>
</tr>
<tr>
<td valign="top"><code>G::apply&lt;R&gt;::type</code></td>
<td>A ResultConverter type for <code>R</code>.</td>
</table>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
09 May, 2002 <!--Luann's birthday! -->
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i>
<p>Permission to copy, use, modify, sell
and distribute this software is granted provided this copyright notice appears
in all copies. This software is provided "as is" without express or implied
warranty, and with no claim as to its suitability for any purpose.
</body>
</html>

View File

@@ -1,133 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - Acknowledgments</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Acknowledgments</h2>
</td>
</tr>
</table>
<hr>
<p><a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a> is
the architect, designer, and implementor of <b>Boost.Python</b>.</p>
<p><a href="mailto:brett.calcott@paradise.net.nz">Brett Calcott</a>
contributed and maintains the Visual Studio project files and
documentation.</p>
<p><a href="mailto:Gottfried.Ganssauge-at-haufe.de">Gottfried
Ganßauge</a> supplied support for opaque pointer conversions,
complete with documentation and a regression test (and I didn't
even have to ask him for those)!
<p>Joel de Guzman implemented the <a href="overloads.html">default
argument support</a> and wrote the excellent <a href=
"../tutorial/index.html">tutorial documentation</a>.</p>
<p><a href="../../../../people/ralf_w_grosse_kunstleve.htm">Ralf W.
Grosse-Kunstleve</a> implemented the <a href="pickle.html">pickle
support</a>, and has enthusiastically supported the library since its
birth, contributing to design decisions and providing invaluable
real-world insight into user requirements. Ralf has written some <a href=
"faq.html#question2">extensions</a> for converting C++ containers that I
hope will be incorporated into the library soon. He also implemented the
cross-module support in the first version of Boost.Python. More
importantly, Ralf makes sure nobody forgets the near-perfect synergy of
C++ and Python for solving the problems of large-scale software
construction.</p>
<p><a href="../../../../people/aleksey_gurtovoy.htm">Aleksey Gurtovoy</a>
wrote an incredible C++ <a href="http://www.mywikinet.com/mpl">Template
Metaprogramming Library</a> which allows Boost.Python to perform much of
its compile-time magic. In addition, Aleksey very generously contributed
his time and deep knowledge of the quirks of various buggy compilers to
help us get around problems at crucial moments.</p>
<p><a href="../../../../people/paul_mensonides.htm">Paul Mensonides</a>,
building on the work <a href="../../../../people/vesa_karvonen.htm">Vesa
Karvonen</a>, wrote a similarly amazing <a href=
"../../../preprocessor/doc/index.html">Preprocessor Metaprogramming
Library</a>, and generously contributed the time and expertise to get it
working in the Boost.Python library, rewriting much of Boost.Python to
use the new preproccessor metaprogramming constructs and helping us to
work around buggy and slow C++ preprocessors.</p>
<p><a href="nicodemus-at-globalite.com.br">Bruno da Silva de
Oliveira</a> contributed the ingenious <a
href="../../pyste/index.html">Pyste</a> (&quot;Pie-Steh&quot;)
code generator.
<p><a href="mailto:nickm@sitius.com">Nikolay Mladenov</a> contributed
<code>staticmethod</code> support.</p>
<p>Martin Casado solved some sticky problems which allow us to build the
Boost.Python shared library for AIX's crazy dynamic linking model.</p>
<p><a href="mailto:achim@procoders.net">Achim Domma</a> contributed some
of the <a href="reference.html#object_wrappers">Object Wrappers</a> and
HTML templates for this documentation. Dave Hawkes contributed
inspiration for the use of the <code><a href=
"scope.html#scope-spec">scope</a></code> class to simplify module
definition syntax. Pearu Pearson wrote some of the test cases that are in
the current test suite.</p>
<p>The development of this version of Boost.Python was funded in part by
the <a href="http://www.llnl.gov/">Lawrence Livermore National
Laboratories</a> and by the <a href="http://cci.lbl.gov/">Computational
Crystallography Initiative</a> at Lawrence Berkeley National
Laboratories.</p>
<p><a href="http://kogs-www.informatik.uni-hamburg.de/~koethe/">Ullrich
Koethe</a> had independently developed a similar system. When he
discovered Boost.Python v1, he generously contributed countless hours of
coding and much insight into improving it. He is responsible for an early
version of the support for function overloading and wrote the support for
reflecting C++ inheritance relationships. He has helped to improve
error-reporting from both Python and C++ (we hope to do as well in v2
again soon), and has designed the original support for exposing numeric
operators, including a way to avoid explicit coercion by means of
overloading.</p>
<p>The members of the boost mailing list and the Python community
supplied invaluable early feedback. In particular, Ron Clarke, Mark
Evans, Anton Gluck, Chuck Ingold, Prabhu Ramachandran, and Barry Scott
took the brave step of trying to use Boost.Python while it was still in
early stages of development.</p>
<p>The first version of Boost.Python would not have been possible without
the support of Dragon Systems, which supported its development and
release as a Boost library.</p>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
26 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,103 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/args.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;boost/python/args.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#keyword-expression"><i>keyword-expressions</i></a></dt>
<dt><a href="#functions">Functions</a></dt>
<dd>
<dl class="page-index">
<dt><code><a href=
"#args-spec">args</a>(</code>...<code>)</code></dt>
</dl>
</dd>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p>Supplies a family of overloaded functions for specifying argument
keywords for wrapped C++ functions.</p>
<h2><a name="keyword-expression"></a><i>keyword-expressions</i></h2>
<p>A <b>keyword-expression</b> results in an object which holds a
sequence of <a href="definitions.html#ntbs">ntbs</a>es, and whose type
encodes the number of keywords specified.</p>
<h2><a name="functions"></a>Functions</h2>
<h3><a name="args-spec"></a><code>args(</code>...<code>)</code></h3>
<pre>
<i>unspecified1</i> args(char const*);
<i>unspecified2</i> args(char const*, char const*);
.
.
.
<i>unspecifiedN</i> args(char const*, char const*, ... char const*);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> Every argument must be a <a href=
"definitions.html#ntbs">ntbs</a>.</dt>
<dt><b>Returns:</b> an object representing a <a href=
"#keyword-expression"><i>keyword-expression</i></a> encapsulating the
arguments passed.</dt>
</dl>
<h2><a name="examples"></a>Example</h2>
<pre>
#include &lt;boost/python/def.hpp&gt;
using namespace boost::python;
int f(int x, int y, int z);
BOOST_PYTHON_MODULE(xxx)
{
def("f", f, args("x", "y", "z"));
}
</pre>
<p>Revised 05 November, 2001</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,32 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - Bibliography</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Bibliography</h2>
</td>
</tr>
</table>
<hr>
{{bibliographical information}}
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a>
2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,82 +0,0 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;call.hpp&gt;</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;call.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#functions">Functions</a></dt>
<dl class="page-index">
<dt><a href="#call-spec">call</a></dt>
</dl>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p>
<code>&lt;boost/python/call.hpp&gt;</code> defines the <a
href="#call-spec"><code>call</code></a> family of overloaded function
templates, used to invoke Python callable objects from C++.
<h2><a name="functions"></a>Functions</h2>
<pre>
<a name="call-spec">template &lt;class R, class A1, class A2, ... class A<i>n</i>&gt;</a>
R call(PyObject* callable, A1 const&amp;, A2 const&amp;, ... A<i>n</i> const&amp;)
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>R</code> is a pointer type, reference
type, or a complete type with an accessible copy constructor</dt>
<dt><b>Effects:</b> Invokes <code>callable(a1,&nbsp;a2,&nbsp;...a<i>n</i>)</code> in
Python, where <code>a1</code>...<code>a<i>n</i></code> are the arguments to
<code>call()</code>, converted to Python objects.
<dt><b>Returns:</b> The result of the Python call, converted to the C++ type <code>R</code>.</dt>
</dt>
<dt><b>Rationale:</b> For a complete semantic description and
rationale, see <a href="callbacks.html">this page</a>.
</dt>
</dl>
<h2><a name="examples"></a>Example(s)</h2>
The following C++ function applies a Python callable object to its two
arguments and returns the result. If a Python exception is raised or
the result can't be converted to a <code>double</code>, an exception
is thrown.
<pre>
double apply2(PyObject* func, double x, double y)
{
return boost::python::call&lt;double&gt;(func, x, y);
}
</pre>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
9 May, 2002 <!-- Luann's birthday! -->
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave Abrahams</a>
2002. All Rights Reserved.</i></p>
</body>
</html>

View File

@@ -1,159 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;call_method.hpp&gt;</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;call_method.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#functions">Functions</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#call_method-spec">call_method</a></dt>
</dl>
</dd>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p><code>&lt;boost/python/call_method.hpp&gt;</code> defines the <a href=
"#call_method-spec"><code>call_method</code></a> family of overloaded
function templates, used to invoke callable attributes of Python objects
from C++.</p>
<h2><a name="functions"></a>Functions</h2>
<pre>
<a name=
"call_method-spec">template &lt;class R, class A1, class A2, ... class A<i>n</i>&gt;</a>
R call_method(PyObject* self, char const* method, A1 const&amp;, A2 const&amp;, ... A<i>n</i> const&amp;)
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>R</code> is a pointer type, reference type,
or a complete type with an accessible copy constructor</dt>
<dt><b>Effects:</b> Invokes
<code>self.<i>method</i>(a1,&nbsp;a2,&nbsp;...a<i>n</i>)</code> in
Python, where <code>a1</code>...<code>a<i>n</i></code> are the
arguments to <code>call_method()</code>, converted to Python objects.
For a complete semantic description, see <a href="callbacks.html">this
page</a>.</dt>
<dt><b>Returns:</b> The result of the Python call, converted to the C++
type <code>R</code>.</dt>
<dt><b>Rationale:</b> <code>call_method</code> is critical to
implementing C++ virtual functions which are overridable in Python, as
shown by the example below.</dt>
</dl>
<h2><a name="examples"></a>Example(s)</h2>
The following C++ illustrates the use of <code>call_method</code> in
wrapping a class with a virtual function that can be overridden in
Python:
<h3>C++ Module Definition</h3>
<pre>
#include &lt;boost/python/module.hpp&gt;
#include &lt;boost/python/class.hpp&gt;
#include &lt;boost/utility.hpp&gt;
#include &lt;cstring&gt;
// class to be wrapped
class Base
{
public:
virtual char const* class_name() const { return "Base"; }
virtual ~Base();
};
bool is_base(Base* b)
{
return !std::strcmp(b-&gt;class_name(), "Base");
}
// Wrapper code begins here
using namespace boost::python;
// Callback class
class Base_callback : public Base
{
public:
Base_callback(PyObject* self) : m_self(self) {}
char const* class_name() const { return <b>call_method</b>&lt;char const*&gt;(m_self, "class_name"); }
char const* Base_name() const { return Base::class_name(); }
private:
PyObject* const m_self;
};
using namespace boost::python;
BOOST_PYTHON_MODULE(my_module)
{
def("is_base", is_base);
class_&lt;Base,Base_callback, noncopyable&gt;("Base")
.def("class_name", &amp;Base_callback::Base_name)
;
}
</pre>
<h3>Python Code</h3>
<pre>
&gt;&gt;&gt; from my_module import *
&gt;&gt;&gt; class Derived(Base):
... def __init__(self):
... Base.__init__(self)
... def class_name(self):
... return self.__class__.__name__
...
&gt;&gt;&gt; is_base(Base()) # calls the class_name() method from C++
1
&gt;&gt;&gt; is_base(Derived())
0
</pre>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,252 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - Calling Python Functions and Methods</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Calling Python Functions and Methods</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#argument_handling">Argument Handling</a></dt>
<dt><a href="#result_handling">Result Handling</a></dt>
<dt><a href="#result_handling">Rationale</a></dt>
</dl>
<hr>
<h2><a name="introduction">Introduction</a></h2>
The simplest way to call a Python function from C++, given an <code><a
href="object.html#object-spec">object</a></code> instance <code>f</code>
holding the function, is simply to invoke its function call operator.
<pre>
f("tea", 4, 2) // In Python: f('tea', 4, 2)
</pre>
And of course, a method of an <code><a href=
"object.html#object-spec">object</a></code> instance <code>x</code> can
be invoked by using the function-call operator of the corresponding
attribute:
<pre>
x.attr("tea")(4, 2); // In Python: x.tea(4, 2)
</pre>
<p>If you don't have an <code>object</code> instance, Boost.Python
provides two families of function templates, <code><a href=
"call.html#call-spec">call</a></code> and <code><a href=
"call_method.html#call_method-spec">call_method</a></code>, for invoking
Python functions and methods respectively on <code>PyObject*</code>s. The
interface for calling a Python function object (or any Python callable
object) looks like:</p>
<pre>
call&lt;ResultType&gt;(callable_object, a1, a2... a<i>N</i>);
</pre>
Calling a method of a Python object is similarly easy:
<pre>
call_method&lt;ResultType&gt;(self_object, "<i>method-name</i>", a1, a2... a<i>N</i>);
</pre>
This comparitively low-level interface is the one you'll use when
implementing C++ virtual functions that can be overridden in Python.
<h2><a name="argument_handling">Argument Handling</a></h2>
<p>Arguments are converted to Python according to their type. By default,
the arguments <code>a1</code>...<code>a<i>N</i></code> are copied into
new Python objects, but this behavior can be overridden by the use of
<code><a href="ptr.html#ptr-spec">ptr()</a></code> and <a href=
"../../../bind/ref.html#reference_wrapper">ref()</a>:</p>
<pre>
class X : boost::noncopyable
{
...
};
void apply(PyObject* callable, X&amp; x)
{
// Invoke callable, passing a Python object which holds a reference to x
boost::python::call&lt;void&gt;(callable, boost::ref(x));
}
</pre>
In the table below, <code><b>x</b></code> denotes the actual argument
object and <code><b>cv</b></code> denotes an optional
<i>cv-qualification</i>: "<code>const</code>", "<code>volatile</code>",
or "<code>const volatile</code>".
<table border="1" summary="class_ template parameters">
<tr>
<th>Argument Type</th>
<th>Behavior</th>
</tr>
<tr>
<td><code>T cv&amp;</code><br>
<code>T cv</code></td>
<td>The Python argument is created by the same means used for the
return value of a wrapped C++ function returning <code>T</code>. When
<code>T</code> is a class type, that normally means <code>*x</code>
is copy-constructed into the new Python object.</td>
</tr>
<tr>
<td><code>T*</code></td>
<td>If <code>x&nbsp;==&nbsp;0</code>, the Python argument will be
<code><a href=
"http://www.python.org/doc/current/lib/bltin-null-object.html">None</a></code>.
Otherwise, the Python argument is created by the same means used for
the return value of a wrapped C++ function returning <code>T</code>.
When <code>T</code> is a class type, that normally means
<code>*x</code> is copy-constructed into the new Python object.</td>
</tr>
<tr>
<td><code><a href=
"../../../bind/ref.html#reference_wrapper">boost::reference_wrapper</a>&lt;T&gt;</code></td>
<td>The Python argument contains a pointer to, rather than a copy of,
<code>x.get()</code>. Note: failure to ensure that no Python code
holds a reference to the resulting object beyond the lifetime of
<code>*x.get()</code> <b>may result in a crash!</b></td>
</tr>
<tr>
<td><code><a href=
"ptr.html#pointer_wrapper-spec">pointer_wrapper</a>&lt;T&gt;</code></td>
<td>If <code>x.get()&nbsp;==&nbsp;0</code>, the Python argument will
be <code><a href=
"http://www.python.org/doc/current/lib/bltin-null-object.html">None</a></code>.
Otherwise, the Python argument contains a pointer to, rather than a
copy of, <code>*x.get()</code>. Note: failure to ensure that no
Python code holds a reference to the resulting object beyond the
lifetime of <code>*x.get()</code> <b>may result in a crash!</b></td>
</tr>
</table>
<h2><a name="result_handling">Result Handling</a></h2>
In general, <code>call&lt;ResultType&gt;()</code> and
<code>call_method&lt;ResultType&gt;()</code> return
<code>ResultType</code> by exploiting all lvalue and rvalue
<code>from_python</code> converters registered for ResultType and
returning a copy of the result. However, when <code>ResultType</code> is
a pointer or reference type, Boost.Python searches only for lvalue
converters. To prevent dangling pointers and references, an exception
will be thrown if the Python result object has only a single reference
count.
<h2><a name="rationale">Rationale</a></h2>
In general, to get Python arguments corresponding to
<code>a1</code>...<code>a<i>N</i></code>, a new Python object must be
created for each one; should the C++ object be copied into that Python
object, or should the Python object simply hold a reference/pointer to
the C++ object? In general, the latter approach is unsafe, since the
called function may store a reference to the Python object somewhere. If
the Python object is used after the C++ object is destroyed, we'll crash
Python.
<p>In keeping with the philosophy that users on the Python side shouldn't
have to worry about crashing the interpreter, the default behavior is to
copy the C++ object, and to allow a non-copying behavior only if the user
writes <code><a href="../../../bind/ref.html">boost::ref</a>(a1)</code>
instead of a1 directly. At least this way, the user doesn't get dangerous
behavior "by accident". It's also worth noting that the non-copying
("by-reference") behavior is in general only available for class types,
and will fail at runtime with a Python exception if used otherwise[<a
href="#1">1</a>].</p>
<p>However, pointer types present a problem: one approach is to refuse to
compile if any aN has pointer type: after all, a user can always pass
<code>*aN</code> to pass "by-value" or <code>ref(*aN)</code> to indicate
a pass-by-reference behavior. However, this creates a problem for the
expected null pointer to <code>None</code> conversion: it's illegal to
dereference a null pointer value.</p>
<p>The compromise I've settled on is this:</p>
<ol>
<li>The default behavior is pass-by-value. If you pass a non-null
pointer, the pointee is copied into a new Python object; otherwise the
corresponding Python argument will be None.</li>
<li>if you want by-reference behavior, use <code>ptr(aN)</code> if
<code>aN</code> is a pointer and <code>ref(aN)</code> otherwise. If a
null pointer is passed to <code>ptr(aN)</code>, the corresponding
Python argument will be <code>None</code>.</li>
</ol>
<p>As for results, we have a similar problem: if <code>ResultType</code>
is allowed to be a pointer or reference type, the lifetime of the object
it refers to is probably being managed by a Python object. When that
Python object is destroyed, our pointer dangles. The problem is
particularly bad when the <code>ResultType</code> is char const* - the
corresponding Python String object is typically uniquely-referenced,
meaning that the pointer dangles as soon as <code>call&lt;char
const*&gt;(...)</code> returns.</p>
<p>The old Boost.Python v1 deals with this issue by refusing to compile
any uses of <code>call&lt;char const*&gt;()</code>, but this goes both
too far and not far enough. It goes too far because there are cases where
the owning Python string object survives beyond the call (just for
instance, when it's the name of a Python class), and it goes not far
enough because we might just as well have the same problem with a
returned pointer or reference of any other type.</p>
<p>In Boost.Python v2 this is dealt with by:</p>
<ol>
<li>lifting the compile-time restriction on const char* callback
returns</li>
<li>detecting the case when the reference count on the result Python
object is 1 and throwing an exception inside of
<code>call&lt;U&gt;(...)</code> when <code>U</code> is a pointer or
reference type.</li>
</ol>
This should be acceptably safe because users have to explicitly specify a
pointer/reference for <code>U</code> in <code>call&lt;U&gt;</code>, and
they will be protected against dangles at runtime, at least long enough
to get out of the <code>call&lt;U&gt;(...)</code> invocation.
<hr>
<a name="1">[1]</a> It would be possible to make it fail at compile-time
for non-class types such as int and char, but I'm not sure it's a good
idea to impose this restriction yet.
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,88 +0,0 @@
Here's the plan:
I aim to provide an interface similar to that of Boost.Python v1's
callback<>::call(...) for dealing with callbacks. The interface will
look like:
returning<ResultType>::call("method_name", self_object, a1, a2...);
or
returning<ResultType>::call(callable_object, a1, a2...);
ARGUMENT HANDLING
There is an issue concerning how to make Python objects from the
arguments a1...aN. A new Python object must be created; should the C++
object be copied into that Python object, or should the Python object
simply hold a reference/pointer to the C++ object? In general, the
latter approach is unsafe, since the called function may store a
reference to the Python object somewhere. If the Python object is used
after the C++ object is destroyed, we'll crash Python.
I plan to make the copying behavior the default, and to allow a
non-copying behavior if the user writes boost::ref(a1) instead of a1
directly. At least this way, the user doesn't get dangerous behavior "by
accident". It's also worth noting that the non-copying ("by-reference")
behavior is in general only available for class types, and will fail at
runtime with a Python exception if used otherwise**
However, pointer types present a problem: My first thought is to refuse
to compile if any aN has pointer type: after all, a user can always pass
*aN to pass "by-value" or ref(*aN) to indicate a pass-by-reference
behavior. However, this creates a problem for the expected NULL pointer
=> None conversion: it's illegal to dereference a null pointer value.
We could use another construct, say "ptr(aN)", to deal with null
pointers, but then what does it mean? We know what it does when aN is
NULL, but it might either have by-value or by-reference behavior when aN
is non-null.
The compromise I've settled on is this:
1. The default behavior is pass-by-value. If you pass a non-null
pointer, the pointee is copied into a new Python object; otherwise
the corresponding Python argument will be None.
2. if you want by-reference behavior, use ptr(aN) if aN is a pointer
and ref(aN) otherwise. If a null pointer is passed to ptr(aN), the
corresponding Python argument will be None.
RESULT HANDLING
As for results, we have a similar problem: if ResultType is allowed to
be a pointer or reference type, the lifetime of the object it refers to
is probably being managed by a Python object. When that Python object is
destroyed, our pointer dangles. The problem is particularly bad when the
ResultType is char const* - the corresponding Python String object is
typically uniquely-referenced, meaning that the pointer dangles as soon
as returning<char const*>::call() returns.
Boost.Python v1 deals with this issue by refusing to compile any uses of
callback<char const*>::call(), but IMO this goes both too far and not
far enough. It goes too far because there are cases where the owning
String object survives beyond the call (just for instance when it's the
name of a Python class), and it goes not far enough because we might
just as well have the same problem with any returned pointer or
reference.
I propose to address this in Boost.Python v2 by
1. lifting the compile-time restriction on const
char* callback returns
2. detecting the case when the reference count on the
result Python object is 1 and throwing an exception
inside of returning<U>::call() when U is a pointer or
reference type.
I think this is acceptably safe because users have to explicitly specify
a pointer/reference for U in returning<U>, and they will be protected
against dangles at runtime, at least long enough to get out of the
returning<U>::call() invocation.
-Dave
**It would be possible to make it fail at compile-time for non-class
types such as int and char, but I'm not sure it's a good idea to impose
this restriction yet.

View File

@@ -1,726 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/class.hpp&gt;,
&lt;boost/python/class_fwd.hpp&gt;</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Headers &lt;boost/python/class.hpp&gt;,
&lt;boost/python/class_fwd.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#class_-spec">Class template
<code>class_</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#class_-spec-synopsis">Class <code>class_</code>
synopsis</a></dt>
<dt><a href="#class_-spec-ctors">Class <code>class_</code>
constructors</a></dt>
<dt><a href="#class_-spec-modifiers">Class <code>class_</code>
modifier functions</a></dt>
</dl>
</dd>
<dt><a href="#bases-spec">Class template
<code>bases</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#bases-spec-synopsis">Class template
<code>bases</code> synopsis</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p><code>&lt;boost/python/class.hpp&gt;</code> defines the interface
through which users expose their C++ classes to Python. It declares the
<code>class_</code> class template, which is parameterized on the class
type being exposed. It also exposes the <code>init</code>,
<code>optional</code> and <code>bases</code> utility class templates,
which are used in conjunction with <code>class_</code>.</p>
<p><code>&lt;boost/python/class_fwd.hpp&gt;</code> contains a forward
declaration of the <code>class_</code> class template.</p>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="class_-spec"></a>Class template
<code>class_&lt;T,&nbsp;<font color="#007F00">Bases,&nbsp;HeldType,
NonCopyable</font>&gt;</code></h3>
<p>Creates a Python class associated with the C++ type passed as its
first parameter. Although it has four template parameters, only the first
one is required. The three optional arguments can actually be supplied
<font color="#007F00"><b>in any order</b></font>; Boost.Python determines
the role of the argument from its type.<br>
<br>
</p>
<table border="1" summary="class_ template parameters">
<tr>
<th>Template Parameter</th>
<th>Requirements</th>
<th>Semantics</th>
<th>Default</th>
</tr>
<tr>
<td><code>T</code></td>
<td>A class type.</td>
<td>The class being wrapped</td>
</tr>
<tr>
<td><code><font color="#007F00">Bases</font></code></td>
<td>A specialization of <a href=
"#bases-spec"><code>bases&lt;</code>...<code>&gt;</code></a> which
specifies previously-exposed C++ base classes of <code>T</code><a
href="#footnote_1">[1]</a>.</td>
<td>Registers <code>from_python</code> conversions from wrapped
<code>T</code> instances to each of its exposed direct and indirect
bases. For each polymorphic base <code>B</code>, registers
conversions from indirectly-held wrapped <code>B</code> instances to
<code>T</code>.</td>
<td><code><a href="#bases">bases&lt;&gt;</a></code></td>
</tr>
<tr>
<td><code><font color="#007F00">HeldType</font></code></td>
<td>Must be <code>T</code>, a class derived from <code>T</code>, or a
<a href="Dereferenceable.html">Dereferenceable</a> type for which
<code><a href=
"pointee.html#pointee-spec">pointee</a>&lt;HeldType&gt;::type</code>
is <code>T</code> or a class derived from <code>T</code>.</td>
<td>Specifies the type which is actually embedded in a Python object
wrapping a <code>T</code> instance. More details <a href=
"#HeldType">below</a>.</td>
<td><code>T</code></td>
</tr>
<tr>
<td><code><font color="#007F00">NonCopyable</font></code></td>
<td>If supplied, must be <a href=
"../../../utility/utility.htm#Class%20noncopyable">boost::noncopyable</a>.</td>
<td>Suppresses automatic registration of <code>to_python</code>
conversions which copy <code>T</code> instances. Required when
<code>T</code> has no publicly-accessible copy constructor.</td>
<td>An unspecified type other than
<code>boost::noncopyable</code>.</td>
</tr>
</table>
<h4><a name="#HeldType">HeldType Semantics</a></h4>
<ol>
<li>If <code>HeldType</code> is derived from T, its exposed
constructor(s) must accept an initial <code>PyObject*</code> argument
which refers back to the Python object that contains the
<code>HeldType</code> instance, as shown in <a href=
"call_method.html#example">this example</a>. This argument is not
included in the <em><a href=
"init.html#init-expressions">init-expression</a></em> passed to <a
href="#class-spec-modifiers"><code>def(init_expr)</code></a>, below,
nor is it passed explicitly by users when Python instances of
<code>T</code> are created. This idiom allows C++ virtual functions
which will be overridden in Python to access the Python object so the
Python method can be invoked. Boost.Python automatically registers
additional converters which allow wrapped instances of <code>T</code>
to be passed to wrapped C++ functions expecting <code>HeldType</code>
arguments.</li>
<li>Because Boost.Python will always allow wrapped instances of
<code>T</code> to be passed in place of <code>HeldType</code>
arguments, specifying a smart pointer for <code>HeldType</code> allows
users to pass Python <code>T</code> instances where a smart
pointer-to-<code>T</code> is expected. Smart pointers such as
<code>std::auto_ptr&lt;&gt;</code> or <code><a href=
"../../../smart_ptr/shared_ptr.htm">boost::shared_ptr&lt;&gt;</a></code>
which contain a nested type <code>element_type</code> designating the
referent type are automatically supported; additional smart pointer
types can be supported by specializing <a href=
"pointee.html#pointee-spec">pointee&lt;HeldType&gt;</a>.</li>
<li>As in case 1 above, when <code>HeldType</code> is a smart pointer
to a class derived from <code>T</code>, the initial
<code>PyObject*</code> argument must be supplied by all of
<code>HeldType</code>'s exposed constructors.</li>
<li>Except in cases 1 and 3, users may optionally specify that T itself
gets initialized with a similar initial <code>PyObject*</code> argument
by specializing <a href=
"has_back_reference.html#has_back_reference-spec">has_back_reference&lt;T&gt;</a>.</li>
</ol>
<h4><a name="class_-spec-synopsis"></a>Class template <code>class_</code>
synopsis</h4>
<pre>
namespace boost { namespace python
{
template &lt;class T
<font color="#007F00"> , class Bases = bases&lt;&gt;
, class HeldType = T
, class NonCopyable = <i>unspecified</i>
&gt;
</font> class class_ : public <a href="object.html#object-spec">object</a>
{
// Constructors with default __init__
class_(char const* name);
class_(char const* name, char const* docstring);
// Constructors, specifying non-default __init__
template &lt;class Init&gt;
class_(char const* name, Init);
template &lt;class Init&gt;
class_(char const* name, char const* docstring, Init);
// Exposing additional __init__ functions
template &lt;class Init&gt;
class_&amp; def(Init);
// defining methods
template &lt;class F&gt;
class_&amp; def(char const* name, F f);
template &lt;class Fn, class A1&gt;
class_&amp; def(char const* name, Fn fn, A1 const&amp;);
template &lt;class Fn, class A1, class A2&gt;
class_&amp; def(char const* name, Fn fn, A1 const&amp;, A2 const&amp;);
template &lt;class Fn, class A1, class A2, class A3&gt;
class_&amp; def(char const* name, Fn fn, A1 const&amp;, A2 const&amp;, A3 const&amp;);
// declaring method as static
class_&amp; staticmethod(char const* name);
// exposing operators
template &lt;<i>unspecified</i>&gt;
class_&amp; def(<a href=
"operators.html#operator_-spec">detail::operator_</a>&lt;unspecified&gt;);
// Raw attribute modification
template &lt;class U&gt;
class_&amp; setattr(char const* name, U const&amp;);
// exposing data members
template &lt;class D&gt;
class_&amp; def_readonly(char const* name, D T::*pm);
template &lt;class D&gt;
class_&amp; def_readwrite(char const* name, D T::*pm);
// property creation
template &lt;class Get&gt;
void add_property(char const* name, Get const&amp; fget);
template &lt;class Get, class Set&gt;
void add_property(char const* name, Get const&amp; fget, Set const&amp; fset);
// pickle support
template &lt;typename PickleSuite&gt;
self&amp; def_pickle(PickleSuite const&amp;);
};
}}
</pre>
<h4><a name="class_-spec-ctors"></a>Class template <code>class_</code>
constructors</h4>
<pre>
class_(char const* name);
class_(char const* name, char const* docstring);
template &lt;class Init&gt;
class_(char const* name, Init init_spec);
template &lt;class Init&gt;
class_(char const* name, char const* docstring, Init init_spec);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>. If <code>docstring</code> is supplied, it must be an
<a href="definitions.html#ntbs">ntbs</a>. If <code>init_spec</code> is
supplied, it must be either the special enumeration constant
<code>no_init</code> or an <a href=
"init.html#init-expression">init-expression</a> compatible with
<code>T</code>.</dt>
<dt><b>Effects:</b> Constructs a <code>class_</code> object holding a
Boost.Python extension class named <code>name</code>. The
<code>name</code>d attribute of the <a href=
"scope.html#introduction">current scope</a> is bound to the new
extension class.</dt>
<dd>
<ul>
<li>If supplied, the value of <code>docstring</code> is bound to
the <code>__doc__</code> attribute of the extension class.</li>
<li>If <code>init_spec</code> is <code>no_init</code>, a special
<code>__init__</code> function is generated which always raises a
Python exception. Otherwise, <code>this-&gt;def(init_spec)</code>
is called.</li>
<li>If <code>init_spec</code> is not supplied,
<code>this-&gt;def(init&lt;&gt;())</code> is called.</li>
</ul>
</dd>
<dt><b>Rationale:</b>Allowing the user to specify constructor arguments
in the <code>class_&lt;&gt;</code> constructor helps her to avoid the
common run-time errors which result from invoking wrapped member
functions without having exposed an <code>__init__</code> function
which creates the requisite <code>T</code> instance. Types which are
not default-constructible will cause a compile-time error unless
<code>Init</code> is supplied. The user must always supply
<code>name</code> as there is currently no portable method to derive
the text of the class name from its type.</dt>
</dl>
<h4><a name="class_-spec-modifiers"></a>Class template
<code>class_</code> modifier functions</h4>
<pre>
template &lt;class Init&gt;
class_&amp; def(Init init_expr);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>init_expr</code> is the result of an <a
href="init.html#init-expression">init-expression</a> compatible with
<code>T</code>.</dt>
<dt><b>Effects:</b> For each <a href="init.html#init-expressions">valid
prefix</a> <em>P</em> of <code>Init</code>, adds an
<code>__init__(</code>...<code>)</code> function overload to the
extension class accepting <em>P</em> as arguments. Each overload
generated constructs an object of <code>HeldType</code> according to
the semantics described <a href="#HeldType">above</a>, using a copy of
<code>init_expr</code>'s <a href="CallPolicies.html">call policies</a>.
If the longest <a href="init.html#init-expressions">valid prefix</a> of
<code>Init</code> contains <em>N</em> types and <code>init_expr</code>
holds <em>M</em> keywords, an initial sequence of the keywords are used
for all but the first <em>N</em>&nbsp;-&nbsp;<em>M</em> arguments of
each overload.</dt>
<dt><b>Returns:</b> <code>*this</code></dt>
<dt><b>Rationale:</b> Allows users to easily expose a class'
constructor to Python.</dt>
</dl>
<br>
<pre>
template &lt;class F&gt;
class_&amp; def(char const* name, Fn fn);
template &lt;class Fn, class A1&gt;
class_&amp; def(char const* name, Fn fn, A1 const&amp; a1);
template &lt;class Fn, class A1, class A2&gt;
class_&amp; def(char const* name, Fn fn, A1 const&amp; a1, A2 const&amp; a2);
template &lt;class Fn, class A1, class A2, class A3&gt;
class_&amp; def(char const* name, Fn fn, A1 const&amp; a1, A2 const&amp; a2, A3 const&amp; a3);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.</dt>
<dd>
<ul>
<li>
If <code>a1</code> is the result of an <a href=
"overloads.html#overload-dispatch-expression"><em>overload-dispatch-expression</em></a>,
only the second form is allowed and fn must be a pointer to
function or pointer to member function whose <a href=
"definitions.html#arity">arity</a> is the same as A1's <a href=
"overloads.html#overload-dispatch-expression"><em>maximum
arity</em></a>.
<dl>
<dt><b>Effects:</b> For each prefix <em>P</em> of
<code>Fn</code>'s sequence of argument types, beginning with
the one whose length is <code>A1</code>'s <a href=
"overloads.html#overload-dispatch-expression"><em>minimum
arity</em></a>, adds a
<code><em>name</em>(</code>...<code>)</code> method overload to
the extension class. Each overload generated invokes
<code>a1</code>'s call-expression with <em>P</em>, using a copy
of <code>a1</code>'s <a href="CallPolicies.html">call
policies</a>. If the longest valid prefix of <code>A1</code>
contains <em>N</em> types and <code>a1</code> holds <em>M</em>
keywords, an initial sequence of the keywords are used for all
but the first <em>N</em>&nbsp;-&nbsp;<em>M</em> arguments of
each overload.<br>
</dt>
</dl>
</li>
<li>
Otherwise, a single method overload is built around fn, which
must not be null:
<ul>
<li>If fn is a function pointer, its first argument must be of
the form <code>U</code>, <code>U&nbsp;<em>cv</em>&amp;</code>,
<code>U&nbsp;<em>cv</em>*</code>, or
<code>U&nbsp;<em>cv</em>*&nbsp;const&amp;</code>, where
<code>T*</code> is convertible to <code>U*</code>, and
<code>a1</code>-<code>a3</code>, if supplied, may be selected
in any order from the table below.</li>
<li>Otherwise, if fn is a member function pointer, its target
must be <code>T</code> or one of its public base classes, and
<code>a1</code>-<code>a3</code>, if supplied, may be selected
in any order from the table below.</li>
<li>Otherwise, <code>Fn</code> must be [derived from] <code><a
href="object.html#object-spec">object</a></code>, and
<code>a1-a2</code>, if supplied, may be selcted in any order
from the first two rows of the table below. To be useful,
<code>fn</code> should be <a href=
"http://www.python.org/doc/current/lib/built-in-funcs.html#l2h-6">
callable</a>.</li>
</ul>
<table border="1" summary="def() optional arguments">
<tr>
<th>Memnonic Name</th>
<th>Requirements/Type properties</th>
<th>Effects</th>
</tr>
<tr>
<td>docstring</td>
<td>Any <a href="definitions.html#ntbs">ntbs</a>.</td>
<td>Value will be bound to the <code>__doc__</code> attribute
of the resulting method overload.</td>
</tr>
<tr>
<td>policies</td>
<td>A model of <a href=
"CallPolicies.html">CallPolicies</a></td>
<td>A copy will be used as the call policies of the resulting
method overload.</td>
</tr>
<tr>
<td>keywords</td>
<td>The result of a <a href=
"args.html#keyword-expression"><em>keyword-expression</em></a>
specifying no more arguments than the <a href=
"definitions.html#arity">arity</a> of <code>fn</code>.</td>
<td>A copy will be used as the call policies of the resulting
method overload.</td>
</tr>
</table>
</li>
</ul>
</dd>
<dt><b>Returns:</b> <code>*this</code></dt>
</dl>
<pre>
class_&amp; staticmethod(char const* name);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>, and corresponds to a method whose overloads have all
been defined.</dt>
<dt><b>Effects:</b> Replaces the existing named attribute <i>x</i> with
the result of invoking <code>staticmethod(</code><i>x</i><code>)</code>
in Python. Specifies that the corresponding method is static and
therefore no object instance will be passed to it. This is equivalent
to the Python statement:</dt>
<dd>
<pre>
setattr(self, name, staticmethod(getattr(self, name)))
</pre>
</dd>
<dt><b>Note:</b> Attempting to invoke <code>def(name,...)</code> after
invoking <code>staticmethod(name)</code> will <a href=
"definitions.html#raise">raise</a> a RuntimeError.</dt>
<dt><b>Returns:</b> <code>*this</code></dt>
</dl>
<br>
<pre>
template &lt;<i>unspecified</i>&gt;
class_&amp; def(<a href=
"operators.html#operator_-spec">detail::operator_</a>&lt;unspecified&gt;);
</pre>
<dl class="function-semantics">
<dt><b>Effects:</b> Adds a Python <a href=
"http://www.python.org/doc/ref/specialnames.html">special method</a> as
described <a href="operators.html">here</a>.</dt>
<dt><b>Returns:</b> <code>*this</code></dt>
</dl>
<pre>
template &lt;class U&gt;
class_&amp; setattr(char const* name, U const&amp; u);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.</dt>
<dt><b>Effects:</b> Converts u to Python and adds it to the attribute
dictionary of the extension class:</dt>
<dd>
<blockquote>
<code><a href=
"http://www.python.org/doc/current/api/object.html#l2h-166">PyObject_SetAttrString</a>(this-&gt;ptr(),&nbsp;name,&nbsp;<a
href="object.html#object-spec-ctors">object</a>(u).ptr());</code>
</blockquote>
</dd>
<dt><b>Returns:</b> <code>*this</code></dt>
</dl>
<br>
<pre>
template &lt;class Get&gt;
void add_property(char const* name, Get const&amp; fget);
template &lt;class Get, class Set&gt;
void add_property(char const* name, Get const&amp; fget, Set const&amp; fset);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.</dt>
<dt><b>Effects:</b> Creates a new Python <a href=
"http://www.python.org/2.2.2/descrintro.html#property"><code>property</code></a>
class instance, passing <code><a href=
"object.html#object-spec-ctors">object</a>(fget)</code> (and <code><a
href="object.html#object-spec-ctors">object</a>(fset)</code> in the
second form) to its constructor, then adds that property to the Python
class object under construction with the given attribute
<code>name</code>.</dt>
<dt><b>Returns:</b> <code>*this</code></dt>
<dt><b>Rationale:</b> Allows users to easily expose functions that can
be invoked from Python with attribute access syntax.</dt>
</dl>
<br>
<pre>
template &lt;class D&gt;
class_&amp; def_readonly(char const* name, D T::*pm);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.</dt>
<dt><b>Effects:</b></dt>
<dd>
<pre>
this-&gt;add_property(name, <a href=
"data_members.html#make_getter-spec">make_getter</a>(pm));
</pre>
</dd>
<dt><b>Returns:</b> <code>*this</code></dt>
<dt><b>Rationale:</b> Allows users to easily expose a class' data
member such that it can be inspected from Python with a natural
syntax.</dt>
</dl>
<pre>
template &lt;class D&gt;
class_&amp; def_readwrite(char const* name, D T::*pm);
</pre>
<dl class="function-semantics">
<dt><b>Effects:</b></dt>
<dd>
<pre>
this-&gt;add_property(name, <a href=
"data_members.html#make_getter-spec">make_getter</a>(pm), <a href=
"data_members.html#make_setter-spec">make_setter</a>(pm));
</pre>
</dd>
<dt><b>Returns:</b> <code>*this</code></dt>
<dt><b>Rationale:</b> Allows users to easily expose a class' data
member such that it can be inspected and set from Python with a natural
syntax.</dt>
</dl>
<pre>
template &lt;typename PickleSuite&gt;
class_&amp; def_pickle(PickleSuite const&amp;);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> PickleSuite must be publically derived from <a
href="pickle.html"><code>pickle_suite</code></a>.</dt>
<dt><b>Effects:</b> Defines a legal combination of the special
attributes and methods: <code>__getinitargs__</code>,
<code>__getstate__</code>, <code>__setstate__</code>,
<code>__getstate_manages_dict__</code>,
<code>__safe_for_unpickling__</code>, <code>__reduce__</code></dt>
<dt><b>Returns:</b> <code>*this</code></dt>
<dt><b>Rationale:</b> Provides an <a href="pickle.html">easy to use
high-level interface</a> for establishing complete pickle support for
the wrapped class. The user is protected by compile-time consistency
checks.</dt>
</dl>
<br>
<h3><a name="bases-spec"></a>Class template
<code>bases&lt;T1,&nbsp;T2,</code>...<code>TN&gt;</code></h3>
<p>An <a href="../../../mpl/doc/ref/Sequences.html">MPL sequence</a>
which can be used in <code>class_&lt;</code>...<code>&gt;</code>
instantiations indicate a list of base classes.</p>
<h4><a name="bases-spec-synopsis"></a>Class template <code>bases</code>
synopsis</h4>
<pre>
namespace boost { namespace python
{
template &lt;T1 = <i>unspecified</i>,...T<i>n</i> = <i>unspecified</i>&gt;
struct bases
{};
}}
</pre>
<h2><a name="examples"></a>Example(s)</h2>
<p>Given a C++ class declaration:</p>
<pre>
class Foo : public Bar, public Baz
{
public:
Foo(int x, char const* y);
Foo(double);
std::string const&amp; name() { return m_name; }
void name(char const*);
double value; // public data
private:
...
};
</pre>
A corresponding Boost.Python extension class can be created with:
<pre>
using namespace boost::python;
class_&lt;Foo,bases&lt;Bar,Baz&gt; &gt;("Foo",
"This is Foo's docstring."
"It describes our Foo extension class",
init&lt;int,char const*&gt;(args("x","y"), "__init__ docstring")
)
.def(init&lt;double&gt;())
.def("get_name", &amp;Foo::get_name, return_internal_reference&lt;&gt;())
.def("set_name", &amp;Foo::set_name)
.def_readwrite("value", &amp;Foo::value)
;
</pre>
<hr>
<a name="footnote_1">[1]</a> By "previously-exposed" we mean that the for
each <code>B</code> in <code>bases</code>, an instance of
<code>class_&lt;B<font color="#007F00">, ...</font>&gt;</code> must have
already been constructed.
<pre>
class_&lt;Base&gt;("Base");
class_&lt;Derived, bases&lt;Base&gt; &gt;("Derived");
</pre>
Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,153 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - Configuration</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Configuration</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#app-defined">Application Defined Macros</a></dt>
<dt><a href="#lib-defined-impl">Library Defined Implementation
Macros</a></dt>
</dl>
<h2><a name="introduction"></a>Introduction</h2>
<p><b>Boost.Python</b> uses several configuration macros in <a href=
"http://www.boost.org/libs/config/config.htm">&lt;boost/config.hpp&gt;</a>,
as well as configuration macros meant to be supplied by the application.
These macros are documented here.</p>
<h2><a name="app-defined"></a>Application Defined Macros</h2>
<p>These are the macros that may be defined by an application using
<b>Boost.Python</b>. Note that if you extend a strict interpretation of
the C++ standard to cover dynamic libraries, using different values of
these macros when compiling different libraries (including extension
modules and the <b>Boost.Python</b> library itself) is a violation of the
<a href="definitions.html#ODR">ODR</a>. However, we know of no C++
implementations on which this particular violation is detectable or
causes any problems.</p>
<table summary="application defined macros" width="100%" cellpadding=
"10">
<tr>
<th align="left"><b>Macro</b></th>
<th><b>Default</b></th>
<th align="left"><b>Meaning</b></th>
</tr>
<tr>
<td valign="top"><code>BOOST_PYTHON_MAX_ARITY</code></td>
<td valign="top" align="center">15</td>
<td valign="top">The maximum <a href=
"definitions.html#arity">arity</a> of any function, member function,
or constructor to be wrapped, invocation of a <b>Boost.Python</b>
function wich is specified as taking arguments
<code>x1,&nbsp;x2,</code>...<code>X</code><i>n</i>. This includes, in
particular, callback mechanisms such as <code><a href=
"object.html#object-spec">object</a>::operator()(</code>...<code>)</code>
or <code><a href=
"call_method.html#call_method-spec">call_method</a>&lt;R&gt;(</code>...
<code>)</code>.</td>
</tr>
<tr>
<td valign="top"><code>BOOST_PYTHON_MAX_BASES</code></td>
<td valign="top" align="center">10</td>
<td valign="top">The maximum number of template arguments to the
<code><a href=
"class.html#bases-spec">bases</a>&lt;</code>...<code>&gt;</code>
class template, which is used to specify the bases of a wrapped C++
class..</td>
</tr>
<tr>
<td valign="top"><code>BOOST_PYTHON_STATIC_MODULE</code></td>
<td valign="top" align="center"><i>not&nbsp;defined</i></td>
<td valign="top">If defined, prevents your module initialization
function from being treated as an exported symbol on platforms which
support that distinction in-code</td>
</tr>
</table>
<h2><a name="lib-defined-impl"></a>Library Defined Implementation
Macros</h2>
<p>These macros are defined by <b>Boost.Python</b> and are implementation
details of interest only to implementors and those porting to new
platforms.</p>
<table summary="library defined implementation macros" width="100%"
cellpadding="10">
<tr>
<th align="left"><b>Macro</b></th>
<th><b>Default</b></th>
<th align="left"><b>Meaning</b></th>
</tr>
<tr>
<td valign="top"><code>BOOST_PYTHON_TYPE_ID_NAME</code></td>
<td valign="top" align="center"><i>not&nbsp;defined</i></td>
<td valign="top">If defined, this indicates that the type_info
comparison across shared library boundaries does not work on this
platform. In other words, if shared-lib-1 passes
<code>typeid(T)</code> to a function in shared-lib-2 which compares
it to <code>typeid(T)</code>, that comparison may return
<code>false</code>. If this macro is #defined, Boost.Python uses and
compares <code>typeid(T).name()</code> instead of using and comparing
the <code>std::type_info</code> objects directly.</td>
</tr>
</table>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
7 January, 2003
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,147 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python -
&lt;boost/python/copy_const_reference.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header
&lt;boost/python/copy_const_reference.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#copy_const_reference-spec">Class
<code>copy_const_reference</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#copy_const_reference-spec-synopsis">Class
<code>copy_const_reference</code> synopsis</a></dt>
<dt><a href="#copy_const_reference-spec-metafunctions">Class
<code>copy_const_reference</code> metafunctions</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#examples">Example</a></dt>
</dl>
<hr>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="copy_const_reference-spec"></a>Class
<code>copy_const_reference</code></h3>
<p><code>copy_const_reference</code> is a model of <a href=
"ResultConverter.html#ResultConverterGenerator-concept">ResultConverterGenerator</a>
which can be used to wrap C++ functions returning a reference-to-const
type such that the referenced value is copied into a new Python
object.</p>
<h4><a name="copy_const_reference-spec-synopsis"></a>Class
<code>copy_const_reference</code> synopsis</h4>
<pre>
namespace boost { namespace python
{
struct copy_const_reference
{
template &lt;class T&gt; struct apply;
};
}}
</pre>
<h4><a name="copy_const_reference-spec-metafunctions"></a>Class
<code>copy_const_reference</code> metafunctions</h4>
<pre>
template &lt;class T&gt; struct apply
</pre>
<dl class="metafunction-semantics">
<dt><b>Requires:</b> <code>T</code> is <code>U const&amp;</code> for
some <code>U</code>.</dt>
<dt><b>Returns:</b> <code>typedef <a href=
"to_python_value.html#to_python_value-spec">to_python_value</a>&lt;T&gt;
type;</code></dt>
</dl>
<h2><a name="examples"></a>Example</h2>
<h3>C++ Module Definition</h3>
<pre>
#include &lt;boost/python/module.hpp&gt;
#include &lt;boost/python/class.hpp&gt;
#include &lt;boost/python/copy_const_reference.hpp&gt;
#include &lt;boost/python/return_value_policy.hpp&gt;
// classes to wrap
struct Bar { int x; }
struct Foo {
Foo(int x) : { b.x = x; }
Bar const&amp; get_bar() const { return b; }
private:
Bar b;
};
// Wrapper code
using namespace boost::python;
BOOST_PYTHON_MODULE(my_module)
{
class_&lt;Bar&gt;("Bar");
class_&lt;Foo&gt;("Foo", init&lt;int&gt;())
.def("get_bar", &amp;Foo::get_bar
, return_value_policy&lt;copy_const_reference&gt;())
;
}
</pre>
<h3>Python Code</h3>
<pre>
&gt;&gt;&gt; from my_module import *
&gt;&gt;&gt; f = Foo(3) # create a Foo object
&gt;&gt;&gt; b = f.get_bar() # make a copy of the internal Bar object
</pre>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,147 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python -
&lt;boost/python/copy_non_const_reference.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header
&lt;boost/python/copy_non_const_reference.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#copy_non_const_reference-spec">Class
<code>copy_non_const_reference</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#copy_non_const_reference-spec-synopsis">Class
<code>copy_non_const_reference</code> synopsis</a></dt>
<dt><a href=
"#copy_non_const_reference-spec-metafunctions">Class
<code>copy_non_const_reference</code> metafunctions</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#examples">Example</a></dt>
</dl>
<hr>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="copy_non_const_reference-spec"></a>Class
<code>copy_non_const_reference</code></h3>
<p><code>copy_non_const_reference</code> is a model of <a href=
"ResultConverter.html#ResultConverterGenerator-concept">ResultConverterGenerator</a>
which can be used to wrap C++ functions returning a
reference-to-non-const type such that the referenced value is copied into
a new Python object.</p>
<h4><a name="copy_non_const_reference-spec-synopsis"></a>Class
<code>copy_non_const_reference</code> synopsis</h4>
<pre>
namespace boost { namespace python
{
struct copy_non_const_reference
{
template &lt;class T&gt; struct apply;
};
}}
</pre>
<h4><a name="copy_non_const_reference-spec-metafunctions"></a>Class
<code>copy_non_const_reference</code> metafunctions</h4>
<pre>
template &lt;class T&gt; struct apply
</pre>
<dl class="metafunction-semantics">
<dt><b>Requires:</b> <code>T</code> is <code>U&amp;</code> for some
non-const <code>U</code>.</dt>
<dt><b>Returns:</b> <code>typedef <a href=
"to_python_value.html#to_python_value-spec">to_python_value</a>&lt;T&gt;
type;</code></dt>
</dl>
<h2><a name="examples"></a>Example</h2>
<p>C++ code:</p>
<pre>
#include &lt;boost/python/module.hpp&gt;
#include &lt;boost/python/class.hpp&gt;
#include &lt;boost/python/copy_non_const_reference.hpp&gt;
#include &lt;boost/python/return_value_policy.hpp&gt;
// classes to wrap
struct Bar { int x; }
struct Foo {
Foo(int x) : { b.x = x; }
Bar&amp; get_bar() { return b; }
private:
Bar b;
};
// Wrapper code
using namespace boost::python;
BOOST_PYTHON_MODULE(my_module)
{
class_&lt;Bar&gt;("Bar");
class_&lt;Foo&gt;("Foo", init&lt;int&gt;())
.def("get_bar", &amp;Foo::get_bar
, return_value_policy&lt;copy_non_const_reference&gt;())
;
}
</pre>
Python Code:
<pre>
&gt;&gt;&gt; from my_module import *
&gt;&gt;&gt; f = Foo(3) # create a Foo object
&gt;&gt;&gt; b = f.get_bar() # make a copy of the internal Bar object
</pre>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,167 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/data_members.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header
&lt;boost/python/data_members.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#functions">Functions</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#make_getter-spec">make_getter</a></dt>
<dt><a href="#make_setter-spec">make_setter</a></dt>
</dl>
</dd>
<dt><a href="#examples">Example</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p><code><a href="#make_getter-spec">make_getter</a>()</code> and
<code><a href="#make_setter-spec">make_setter</a>()</code> are the
functions used internally by <code>class_&lt;&gt;::<a href=
"class.html#class_-spec-modifiers">def_readonly</a></code> and
<code>class_&lt;&gt;::<a href=
"class.html#class_-spec-modifiers">def_readwrite</a></code> to produce
Python callable objects which wrap C++ data members.</p>
<h2><a name="functions"></a>Functions</h2>
<pre>
<a name="make_getter-spec">template &lt;class C, class D&gt;</a>
<a href="object.html#object-spec">object</a> make_getter(D C::*pm);
template &lt;class C, class D, class Policies&gt;
<a href=
"object.html#object-spec">object</a> make_getter(D C::*pm, Policies const&amp; policies);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>Policies</code> is a model of <a href=
"CallPolicies.html">CallPolicies</a>.</dt>
<dt><b>Effects:</b> Creates a Python callable object which accepts a
single argument that can be converted <code>from_python</code> to
<code>C*</code>, and returns the corresponding member <code>D</code>
member of the <code>C</code> object, converted <code>to_python</code>.
If <code>policies</code> is supplied, it will be applied to the
function as described <a href="CallPolicies.html">here</a>. Otherwise,
the library attempts to determine whether <code>D</code> is a
user-defined class type, and if so uses <code><a href=
"return_internal_reference.html#return_internal_reference-spec">return_internal_reference</a>&lt;&gt;</code></dt>
<dt>for <code>Policies</code>. Note that this test may inappropriately
choose <code>return_internal_reference&lt;&gt;</code> in some cases
when <code>D</code> is a smart pointer type. This is a known
defect.</dt>
<dt><b>Returns:</b> An instance of <a href=
"object.html#object-spec">object</a> which holds the new Python
callable object.</dt>
</dl>
<pre>
<a name="make_setter-spec">template &lt;class C, class D&gt;</a>
<a href="object.html#object-spec">object</a> make_setter(D C::*pm);
template &lt;class C, class D, class Policies&gt;
<a href=
"object.html#object-spec">object</a> make_setter(D C::*pm, Policies const&amp; policies);
</pre>
<dl class="function*-semantics">
<dt><b>Requires:</b> <code>Policies</code> is a model of <a href=
"CallPolicies.html">CallPolicies</a>.</dt>
<dt><b>Effects:</b> Creates a Python callable object which, when called
from Python, expects two arguments which can be converted
<code>from_python</code> to <code>C*</code> and
<code>D&nbsp;const&amp;</code>, respectively, and sets the
corresponding <code>D</code> member of the <code>C</code> object. If
<code>policies</code> is supplied, it will be applied to the function
as described <a href="CallPolicies.html">here</a>.</dt>
<dt><b>Returns:</b> An instance of <a href=
"object.html#object-spec">object</a> which holds the new Python
callable object.</dt>
</dl>
<h2><a name="examples"></a>Example</h2>
<p>The code below uses make_getter and make_setter to expose a data
member as functions:</p>
<pre>
#include &lt;boost/python/data_members.hpp&gt;
#include &lt;boost/python/module.hpp&gt;
#include &lt;boost/python/class.hpp&gt;
struct X
{
X(int x) : y(x) {}
int y;
};
using namespace boost::python;
BOOST_PYTHON_MODULE_INIT(data_members_example)
{
class_&lt;X&gt;("X", init&lt;int&gt;())
.def("get", make_getter(&amp;X::y))
.def("set", make_setter(&amp;X::y))
;
}
</pre>
It can be used this way in Python:
<pre>
&gt;&gt;&gt; from data_members_example import *
&gt;&gt;&gt; x = X(1)
&gt;&gt;&gt; x.get()
1
&gt;&gt;&gt; x.set(2)
&gt;&gt;&gt; x.get()
2
</pre>
<p>
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,189 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Cygwin (vers 1st April 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/def.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;boost/python/def.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#functions">Functions</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#def-spec">def</a></dt>
</dl>
</dd>
<dt><a href="#examples">Example</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p><code><a href="#def-spec">def</a>()</code> is the function which can
be used to expose C++ functions and callable objects as Python functions
in the current <code><a href="scope.html">scope</a></code>.</p>
<h2><a name="functions"></a>Functions</h2>
<a name="def-spec"></a>def
<pre>
template &lt;class F&gt;
void def(char const* name, F f);
template &lt;class Fn, class A1&gt;
void def(char const* name, Fn fn, A1 const&amp;);
template &lt;class Fn, class A1, class A2&gt;
void def(char const* name, Fn fn, A1 const&amp;, A2 const&amp;);
template &lt;class Fn, class A1, class A2, class A3&gt;
void def(char const* name, Fn fn, A1 const&amp;, A2 const&amp;, A3 const&amp;);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.</dt>
<dd>
<ul>
<li>If <code>Fn</code> is [derived from] <code><a href=
"object.html#object-spec">object</a></code>, it will be added to
the current scope as a single overload. To be useful,
<code>fn</code> should be <a href=
"http://www.python.org/doc/current/lib/built-in-funcs.html#l2h-6">callable</a>.</li>
<li>
If <code>a1</code> is the result of an <a href=
"overloads.html#overload-dispatch-expression"><em>overload-dispatch-expression</em></a>,
only the second form is allowed and fn must be a pointer to
function or pointer to member function whose <a href=
"definitions.html#arity">arity</a> is the same as A1's <a href=
"overloads.html#overload-dispatch-expression"><em>maximum
arity</em></a>.
<dl>
<dt><b>Effects:</b> For each prefix <em>P</em> of
<code>Fn</code>'s sequence of argument types, beginning with
the one whose length is <code>A1</code>'s <a href=
"overloads.html#overload-dispatch-expression"><em>minimum
arity</em></a>, adds a
<code><em>name</em>(</code>...<code>)</code> function overload
to the <a href="scope.html">current scope</a>. Each overload
generated invokes <code>a1</code>'s call-expression with
<em>P</em>, using a copy of <code>a1</code>'s <a href=
"CallPolicies.html">call policies</a>. If the longest valid
prefix of <code>A1</code> contains <em>N</em> types and
<code>a1</code> holds <em>M</em> keywords, an initial sequence
of the keywords are used for all but the first
<em>N</em>&nbsp;-&nbsp;<em>M</em> arguments of each
overload.<br>
</dt>
</dl>
</li>
<li>Otherwise, fn must be a non-null function or member function
pointer, and a single function overload built around fn is added to
the <a href="scope.html">current scope</a>. If any of
<code>a1</code>-<code>a3</code> are supplied, they may be selected
in any order from the table below.</li>
</ul>
<table border="1" summary="def() optional arguments">
<tr>
<th>Memnonic Name</th>
<th>Requirements/Type properties</th>
<th>Effects</th>
</tr>
<tr>
<td>docstring</td>
<td>Any <a href="definitions.html#ntbs">ntbs</a>.</td>
<td>Value will be bound to the <code>__doc__</code> attribute of
the resulting method overload.</td>
</tr>
<tr>
<td>policies</td>
<td>A model of <a href="CallPolicies.html">CallPolicies</a></td>
<td>A copy will be used as the call policies of the resulting
method overload.</td>
</tr>
<tr>
<td>keywords</td>
<td>The result of a <a href=
"args.html#keyword-expression"><em>keyword-expression</em></a>
specifying no more arguments than the <a href=
"definitions.html#arity">arity</a> of <code>fn</code>.</td>
<td>A copy will be used as the call policies of the resulting
method overload.</td>
</tr>
</table>
</dd>
</dl>
<h2><a name="examples"></a>Example</h2>
<pre>
#include &lt;boost/python/def.hpp&gt;
#include &lt;boost/python/module.hpp&gt;
#include &lt;boost/python/args.hpp&gt;
using namespace boost::python;
char const* foo(int x, int y) { return "foo"; }
BOOST_PYTHON_MODULE(def_test)
{
def("foo", foo, args("x", "y"), "foo's docstring");
}
</pre>
<p>
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
7 March, 2003
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,170 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<meta name="generator" content="HTML Tidy, see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python -
&lt;boost/python/default_call_policies.hpp&gt;</title>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277" alt=
"C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header
&lt;boost/python/default_call_policies.hpp&gt;</h2>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#classes">Classes</a>
<dd>
<dl class="page-index">
<dt><a href="#default_call_policies-spec">Class
<code>default_call_policies</code></a>
<dd>
<dl class="page-index">
<dt><a href="#default_call_policies-spec-synopsis">Class
<code>default_call_policies</code> synopsis</a>
<dt><a href="#default_call_policies-spec-statics">Class
<code>default_call_policies</code> static functions</a>
</dl>
<dt><a href="#default_result_converter-spec">Class
<code>default_result_converter</code></a>
<dd>
<dl class="page-index">
<dt><a href="#default_result_converter-spec-synopsis">Class
<code>default_result_converter</code> synopsis</a>
<dt><a href="#default_result_converter-spec-metafunctions">Class
<code>default_result_converter</code> metafunctions</a>
</dl>
</dl>
<dt><a href="#examples">Example</a>
</dl>
<hr>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="default_call_policies-spec"></a>Class
<code>default_call_policies</code></h3>
<p><code>default_call_policies</code> is a model of <a href=
"CallPolicies.html">CallPolicies</a> with no <code>precall</code> or
<code>postcall</code> behavior and a <code>result_converter</code> which
handles by-value returns. Wrapped C++ functions and member functions use
<code>default_call_policies</code> unless otherwise specified. You may find
it convenient to derive new models of <a href=
"CallPolicies.html">CallPolicies</a> from
<code>default_call_policies</code>.
<h4><a name="default_call_policies-spec-synopsis"></a>Class
<code>default_call_policies</code> synopsis</h4>
<pre>
namespace boost { namespace python
{
struct default_call_policies
{
static bool precall(PyObject*);
static PyObject* postcall(PyObject*, PyObject* result);
typedef <a href=
"#default_result_converter-spec">default_result_converter</a> result_converter;
};
}}
</pre>
<h4><a name="default_call_policies-spec-statics"></a>Class
<code>default_call_policies</code> static functions</h4>
<pre>
bool precall(PyObject*);
</pre>
<dl class="function-semantics">
<dt><b>Returns:</b> <code>true</code>
<dt><b>Throws:</b> nothing
</dl>
<pre>
PyObject* postcall(PyObject*, PyObject* result);
</pre>
<dl class="function-semantics">
<dt><b>Returns:</b> <code>result</code>
<dt><b>Throws:</b> nothing
</dl>
<h3><a name="default_result_converter-spec"></a>Class
<code>default_result_converter</code></h3>
<p><code>default_result_converter</code> is a model of <a href=
"ResultConverter.html#ResultConverterGenerator-concept">ResultConverterGenerator</a> which can be
used to wrap C++ functions returning non-pointer types, <code>char
const*</code>, and <code>PyObject*</code>, by-value.
<h4><a name="default_result_converter-spec-synopsis"></a>Class
<code>default_result_converter</code> synopsis</h4>
<pre>
namespace boost { namespace python
{
struct default_result_converter
{
template &lt;class T&gt; struct apply;
};
}}
</pre>
<h4><a name="default_result_converter-spec-metafunctions"></a>Class
<code>default_result_converter</code> metafunctions</h4>
<pre>
template &lt;class T&gt; struct apply
</pre>
<dl class="metafunction-semantics">
<dt><b>Requires:</b> <code>T</code> is not a reference type. If
<code>T</code> is a pointer type, <code>T</code> is <code>const
char*</code> or <code>PyObject*</code>.
<dt><b>Returns:</b> <code>typedef <a href=
"to_python_value.html#to_python_value-spec">to_python_value</a>&lt;T
const&amp;&gt; type;</code>
</dl>
<h2><a name="examples"></a>Example</h2>
<p>This example comes from the Boost.Python implementation itself. Because
the <a href=
"return_value_policy.html#return_value_policy-spec">return_value_policy</a>
class template does not implement <code>precall</code> or
<code>postcall</code> behavior, its default base class is
<code>default_call_policies</code>:
<pre>
template &lt;class Handler, class Base = default_call_policies&gt;
struct return_value_policy : Base
{
typedef Handler result_converter;
};
</pre>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
<p><i>&copy; Copyright <a href="../../../../people/dave_abrahams.htm">Dave
Abrahams</a> 2002. All Rights Reserved.</i>

View File

@@ -1,100 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - Definitions</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Definitions</h2>
</td>
</tr>
</table>
<hr>
<dl class="definitions">
<dt><b><a name="arity">arity</a>:</b> The number of arguments accepted
by a function or member function. Unless otherwise specified, the
hidden "<code>this</code>" argument to member functions is not counted
when specifying arity</dt>
<dd><br>
</dd>
<dt><b><a name="ntbs">ntbs</a>:</b> Null-Terminated Byte String, or
`C'-string. C++ string literals are <strong>ntbs</strong>es. An
<strong>ntbs</strong> must never be null.</dt>
<dd><br>
</dd>
<dt><b><a name="raise">raise</a>:</b> Exceptions in Python are
"raised", not "thrown", as they are in C++. When this documentation
says that some Python exception is "raised" in the context of C++ code,
it means that the corresponding Python exception is set via the <a
href=
"http://www.python.org/doc/current/api/exceptionHandling.html">Python/'C'
API</a>, and <code><a href=
"errors.html#throw_error_already_set-spec">throw_error_already_set</a>()</code>
is called.</dt>
<dd><br>
</dd>
<dt><b><a name="POD">POD</a>:</b> A technical term from the C++
standard. Short for "Plain Ol'Data": A POD-struct is an aggregate class
that has no non-static data members of type pointer to member,
non-POD-struct, non-POD-union (or array of such types) or reference,
and has no user-defined copy assign- ment operator and no user-defined
destructor. Similarly, a POD-union is an aggregate union that has no
non-static data members of type pointer to member, non-POD-struct,
non-POD-union (or array of such types) or reference, and has no
user-defined copy assignment operator and no user-defined destructor. A
POD class is a class that is either a POD-struct or a POD-union. An
aggregate is an array or a class (clause 9) with no user-declared
constructors (12.1), no private or protected non-static data members
(clause 11), no base classes (clause 10), and no virtual functions
(10.3).</dt>
<dd><br>
</dd>
<dt><b><a name="ODR">ODR</a>:</b> The &quot;One Definition
Rule&quot;, which says that any entity in a C++ program must have the same definition in all translation units (object files) which make up a program.
</dt>
<dd><br>
</dd>
</dl>
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,150 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/dict.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;boost/python/dict.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#dict-spec">Class <code>dict</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#dict-spec-synopsis">Class <code>dict</code>
synopsis</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p>Exposes a <a href=
"ObjectWrapper.html#TypeWrapper-concept">TypeWrapper</a> for the Python
<a href=
"http://www.python.org/dev/doc/devel/lib/typesmapping.html">dict</a>
type.</p>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="dict-spec"></a>Class <code>dict</code></h3>
<p>Exposes the <a href=
"http://www.python.org/dev/doc/devel/lib/typesmapping.html">mapping
protocol</a> of Python's built-in <code>dict</code> type. The semantics
of the constructors and member functions defined below can be fully
understood by reading the <a href=
"ObjectWrapper.html#TypeWrapper-concept">TypeWrapper</a> concept
definition. Since <code>dict</code> is publicly derived from <code><a
href="object.html#object-spec">object</a></code>, the public object
interface applies to <code>dict</code> instances as well.</p>
<h4><a name="dict-spec-synopsis"></a>Class <code>dict</code>
synopsis</h4>
<pre>
namespace boost { namespace python
{
class dict : public object
{
dict();
template&lt; class T &gt;
dict(T const &amp; data);
// modifiers
void clear();
dict copy();
template &lt;class T1, class T2&gt;
tuple popitem();
template &lt;class T&gt;
object setdefault(T const &amp;k);
template &lt;class T1, class T2&gt;
object setdefault(T1 const &amp; k, T2 const &amp; d);
void update(object_cref E);
template&lt; class T &gt;
void update(T const &amp; E);
// observers
list values() const;
object get(object_cref k) const;
template&lt;class T&gt;
object get(T const &amp; k) const;
object get(object_cref k, object_cref d) const;
object get(T1 const &amp; k, T2 const &amp; d) const;
bool has_key(object_cref k) const;
template&lt; class T &gt;
bool has_key(T const &amp; k) const;
list items() const;
object iteritems() const;
object iterkeys() const;
object itervalues() const;
list keys() const;
};
}}
</pre>
<h2><a name="examples"></a>Example</h2>
<pre>
using namespace boost::python;
dict swap_object_dict(object target, dict d)
{
dict result = extract&lt;dict&gt;(target.attr("__dict__"));
target.attr("__dict__") = d;
return result;
}
</pre>
<p>Revised 30 September, 2002</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,232 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/enum.hpp&gt;</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;boost/python/enum.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#enum_-spec">Class template
<code>enum_</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#enum_-spec-synopsis">Class template <code>enum_</code>
synopsis</a></dt>
<dt><a href="#enum_-spec-ctors">Class template <code>enum_</code>
constructors</a></dt>
<dt><a href="#enum_-spec-modifiers">Class template <code>enum_</code>
modifier functions</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p><code>&lt;boost/python/enum.hpp&gt;</code> defines the
interface through which users expose their C++ enumeration types
to Python. It declares the
<code>enum_</code> class template, which is parameterized on the
enumeration type being exposed. </p>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="enum_-spec"></a>Class template
<code>enum_&lt;T&gt;</code></h3>
<p>Creates a Python class derived from Python's <code>int</code>
type which is associated with the C++ type passed as its first
parameter.
<h4><a name="enum_-spec-synopsis"></a>Class template <code>enum_</code>
synopsis</h4>
<pre>
namespace boost { namespace python
{
template &lt;class T&gt;
class enum_ : public <a href="object.html#object-spec">object</a>
{
enum_(char const* name);
enum_&lt;T&gt;&amp; value(char const* name, T);
enum_&lt;T&gt;&amp; export_values();
};
}}
</pre>
<h4><a name="enum_-spec-ctors"></a>Class template <code>enum_</code>
constructors</h4>
<pre>
enum_(char const* name);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.
<dt><b>Effects:</b> Constructs an <code>enum_</code> object
holding a Python extension type derived from <code>int</code>
which is named <code>name</code>. The
<code>name</code>d attribute of the <a href=
"scope.html#introduction">current scope</a> is bound to the new
extension type.</dt>
</dl>
<h4><a name="enum_-spec-modifiers"></a>Class template
<code>enum_</code> modifier functions</h4>
<pre>
inline enum_&lt;T&gt;&amp; value(char const* name, T x);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> <code>name</code> is an <a href=
"definitions.html#ntbs">ntbs</a> which conforms to Python's <a
href=
"http://www.python.org/doc/current/ref/identifiers.html">identifier
naming rules</a>.
<dt><b>Effects:</b> adds an instance of the wrapped enumeration
type with value <code>x</code> to the type's dictionary as the
<code>name</code>d attribute</dt>.
<dt><b>Returns:</b> <code>*this</code></dt>
</dl>
<pre>
inline enum_&lt;T&gt;&amp; export_values();
</pre>
<dl class="function-semantics">
<dt><b>Effects:</b> sets attributes in the current <a
href="scope.html#scope-spec"><code>scope</code></a> with the
same names and values as all enumeration values exposed so far
by calling <code>value()</code></dt>.
<dt><b>Returns:</b> <code>*this</code></dt>
</dl>
<h2><a name="examples"></a>Example(s)</h2>
<p>C++ module definition
<pre>
#include &lt;boost/python/enum.hpp&gt;
#include &lt;boost/python/def.hpp&gt;
#include &lt;boost/python/module.hpp&gt;
using namespace boost::python;
enum color { red = 1, green = 2, blue = 4 };
color identity_(color x) { return x; }
BOOST_PYTHON_MODULE(enums)
{
enum_&lt;color&gt;(&quot;color&quot;)
.value(&quot;red&quot;, red)
.value(&quot;green&quot;, green)
.export_values()
.value(&quot;blue&quot;, blue)
;
def(&quot;identity&quot;, identity_);
}
</pre>
<p>Interactive Python:
<pre>
&gt;&gt;&gt; from enums import *
&gt;&gt;&gt; identity(red)
enums.color.red
&gt;&gt;&gt; identity(color.red)
enums.color.red
&gt;&gt;&gt; identity(green)
enums.color.green
&gt;&gt;&gt; identity(color.green)
enums.color.green
&gt;&gt;&gt; identity(blue)
Traceback (most recent call last):
File &quot;&lt;stdin&gt;&quot;, line 1, in ?
NameError: name blue' is not defined
&gt;&gt;&gt; identity(color.blue)
enums.color.blue
&gt;&gt;&gt; identity(color(1))
enums.color.red
&gt;&gt;&gt; identity(color(2))
enums.color.green
&gt;&gt;&gt; identity(color(3))
enums.color(3)
&gt;&gt;&gt; identity(color(4))
enums.color.blue
&gt;&gt;&gt; identity(1)
Traceback (most recent call last):
File &quot;&lt;stdin&gt;&quot;, line 1, in ?
TypeError: bad argument type for built-in operation
</pre>
<hr>
Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 December, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,287 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/errors.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;boost/python/errors.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#error_already_set-spec">Class
<code>error_already_set</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#error_already_set-spec-synopsis">Class
<code>error_already_set</code> synopsis</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#functions">Functions</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#handle_exception-spec">handle_exception</a></dt>
<dt><a href="#expect_non_null-spec">expect_non_null</a></dt>
<dt><a href=
"#throw_error_already_set-spec">throw_error_already_set</a></dt>
</dl>
</dd>
<dt><a href="#examples">Examples</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p><code>&lt;boost/python/errors.hpp&gt;</code> provides types and
functions for managing and translating between Python and C++ exceptions.
This is relatively low-level functionality that is mostly used internally
by Boost.Python. Users should seldom need it.</p>
<h2><a name="classes"></a>Classes</h2>
<h3><a name="error_already_set-spec"></a>Class
<code>error_already_set</code></h3>
<p><code>error_already_set</code> is an exception type which can be
thrown to indicate that a Python error has occurred. If thrown, the
precondition is that <a href=
"http://www.python.org/doc/2.2/api/exceptionHandling.html#l2h-71">PyErr_Occurred()</a>
returns a value convertible to <code>true</code>. Portable code shouldn't
throw this exception type directly, but should instead use <code><a href=
"#throw_error_already_set-spec">throw_error_already_set</a>()</code>,
below.</p>
<h4><a name="error_already_set-spec-synopsis"></a>Class error_already_set
synopsis</h4>
<pre>
namespace boost { namespace python
{
class error_already_set {};
}}
</pre>
<h2><a name="functions"></a>Functions</h2>
<pre>
<a name=
"handle_exception-spec">template &lt;class T&gt; bool handle_exception</a>(T f) throw();
void handle_exception() throw();
</pre>
<dl class="handle_exception-semantics">
<dt><b>Requires:</b> The first form requires that the expression
<code><a href=
"../../../function/doc/reference.html#functionN">function0</a>&lt;void&gt;(f)</code>
is valid. The second form requires that a C++ exception is currently
being handled (see section 15.1 in the C++ standard).</dt>
<dt><b>Effects:</b> The first form calls <code>f()</code> inside a
<code>try</code> block which first attempts to use all registered <a
href="exception_translator.html">exception translators</a>. If none of
those translates the exception, the <code>catch</code> clauses then set
an appropriate Python exception for the C++ exception caught, returning
<code>true</code> if an exception was thrown, <code>false</code>
otherwise. The second form passes a function which rethrows the
exception currently being handled to the first form.</dt>
<dt><b>Postconditions:</b> No exception is being handled</dt>
<dt><b>Throws:</b> nothing</dt>
<dt><b>Rationale:</b> At inter-language boundaries it is important to
ensure that no C++ exceptions escape, since the calling language
usually doesn't have the equipment neccessary to properly unwind the
stack. Use <code>handle_exception</code> to manage exception
translation whenever your C++ code is called directly from the Python
API. This is done for you automatically by the usual function wrapping
facilities: <code><a href=
"make_function.html#make_function-spec">make_function</a>()</code>,
<code><a href=
"make_function.html#make_constructor-spec">make_constructor</a>()</code>,
<code><a href="def.html#def-spec">def</a>()</code> and <code><a href=
"class.html#def-spec">class_::def</a>()</code>. The second form can be
more convenient to use (see the <a href="#examples">example</a> below),
but various compilers have problems when exceptions are rethrown from
within an enclosing <code>try</code> block.</dt>
</dl>
<pre>
<a name=
"expect_non_null-spec">template &lt;class T&gt; T* expect_non_null(T* x);</a>
</pre>
<dl class="expect_non_null-semantics">
<dt><b>Returns:</b> <code>x</code></dt>
<dt><b>Throws:</b> <code><a href=
"#error_already_set-spec">error_already_set</a>()</code> iff <code>x ==
0</code>.</dt>
<dt><b>Rationale:</b> Simplifies error-handling when calling functions
in the <a href="http://www.python.org/doc/2.2/api/api.html">Python/C
API</a> which return 0 on error.</dt>
</dl>
<pre>
<a name="throw_error_already_set-spec">void throw_error_already_set();</a>
</pre>
<dl class="throw_error_already_set-semantics">
<dt><b>Effects:</b> <code>throw&nbsp;<a href=
"#error_already_set-spec">error_already_set</a>();</code></dt>
</dl>
<dl>
<dt><b>Rationale:</b> Many platforms and compilers are not able to
consistently catch exceptions thrown across shared library boundaries.
Using this function from the Boost.Python library ensures that the
appropriate <code>catch</code> block in <code><a href=
"#handle_exception-spec">handle_exception</a>()</code> can catch the
exception.</dt>
</dl>
<h2><a name="examples"></a>Examples</h2>
<pre>
#include &lt;string&gt;
#include &lt;boost/python/errors.hpp&gt;
#include &lt;boost/python/object.hpp&gt;
#include &lt;boost/python/handle.hpp&gt;
// Returns a std::string which has the same value as obj's "__name__"
// attribute.
std::string get_name(boost::python::object obj)
{
// throws if there's no __name__ attribute
PyObject* p = boost::python::expect_non_null(
PyObject_GetAttrString(obj.ptr(), "__name__"));
char const* s = PyString_AsString(p);
if (s != 0)
Py_DECREF(p);
// throws if it's not a Python string
std::string result(
boost::python::expect_non_null(
PyString_AsString(p)));
Py_DECREF(p); // Done with p
return result;
}
//
// Demonstrate form 1 of handle_exception
//
// Place into result a Python Int object whose value is 1 if a and b have
// identical "__name__" attributes, 0 otherwise.
void same_name_impl(PyObject*&amp; result, boost::python::object a, boost::python::object b)
{
result = PyInt_FromLong(
get_name(a) == get_name(a2));
}
object borrowed_object(PyObject* p)
{
return boost::python::object(
boost::python::handle&lt;&gt;(
boost::python::borrowed(a1)));
}
// This is an example Python 'C' API interface function
extern "C" PyObject*
same_name(PyObject* args, PyObject* keywords)
{
PyObject* a1;
PyObject* a2;
PyObject* result = 0;
if (!PyArg_ParseTuple(args, const_cast&lt;char*&gt;("OO"), &amp;a1, &amp;a2))
return 0;
// Use boost::bind to make an object compatible with
// boost::Function0&lt;void&gt;
if (boost::python::handle_exception(
boost::bind&lt;void&gt;(same_name_impl, boost::ref(result), borrowed_object(a1), borrowed_object(a2))))
{
// an exception was thrown; the Python error was set by
// handle_exception()
return 0;
}
return result;
}
//
// Demonstrate form 2 of handle_exception. Not well-supported by all
// compilers.
//
extern "C" PyObject*
same_name2(PyObject* args, PyObject* keywords)
{
PyObject* a1;
PyObject* a2;
PyObject* result = 0;
if (!PyArg_ParseTuple(args, const_cast&lt;char*&gt;("OO"), &amp;a1, &amp;a2))
return 0;
try {
return PyInt_FromLong(
get_name(borrowed_object(a1)) == get_name(borrowed_object(a2)));
}
catch(...)
{
// If an exception was thrown, translate it to Python
boost::python::handle_exception();
return 0;
}
}
</pre>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
13 November, 2002
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,148 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python -
&lt;boost/python/exception_translator.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header
&lt;boost/python/exception_translator.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#functions">Functions</a></dt>
<dd>
<dl class="page-index">
<dt><a href=
"#register_exception_translator-spec">register_exception_translator</a></dt>
</dl>
</dd>
<dt><a href="#examples">Example(s)</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p>As described <a href="errors.html#handle_exception-spec">here</a>, it
is important to make sure that exceptions thrown by C++ code do not pass
into the Python interpreter core. By default, Boost.Python translates all
C++ exceptions thrown by wrapped functions and module init functions into
Python, but the default translators are extremely limited: most C++
exceptions will appear in Python as a <a href=
"http://www.python.org/doc/current/lib/module-exceptions.html">RuntimeError</a>
exception whose representation is
<code>'Unidentifiable&nbsp;C++&nbsp;Exception'</code>. To produce better
error messages, users can register additional exception translators as
described below.</p>
<h2><a name="functions"></a>Functions</h2>
<h3><code><a name="register_exception_translator-spec">register_exception_translator</a></code></h3>
<pre>
<a name="register_exception_translator-spec">template&lt;class ExceptionType, class Translate&gt;</a>
void register_exception_translator(Translate const&amp; translate);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b></dt>
<dd>
<code>Translate</code> is <a href=
"../../../utility/CopyConstructible.html">Copyconstructible</a>, and
the following code must be well-formed:
<pre>
void f(ExceptionType x) { translate(x); }
</pre>
The expression <code>translate(x)</code> must either throw a C++
exception, or a subsequent call to <code><a href=
"http://www.python.org/doc/current/api/exceptionHandling.html">PyErr_Occurred</a>()</code>
must return 1.
</dd>
<p>
<dt><b>Effects:</b> Adds a copy of <code>translate</code> to the sequence of
exception translators tried when Boost.Python catches an exception that
is about to pass into Python's core interpreter. The new translator
will get "first shot" at translating all exceptions matching the catch
clause shown above. Any subsequently-registered translators will be
allowed to translate the exception earlier. A translator which cannot
translate a given C++ exception can re-throw it, and it will be handled
by a translator which was registered earlier (or by the default
translator).</dt>
</dl>
<h2><a name="examples"></a>Example</h2>
<pre>
#include &lt;boost/python/module.hpp&gt;
#include &lt;boost/python/def.hpp&gt;
#include &lt;boost/python/exception_translator.hpp&gt;
#include &lt;exception&gt;
struct my_exception : std::exception
{
char const* what() throw() { return "One of my exceptions"; }
};
void translate(my_exception const&amp; e)
{
// Use the Python 'C' API to set up an exception object
PyErr_SetString(PyExc_RuntimeError, e.what());
}
void something_which_throws()
{
...
throw my_exception();
...
}
BOOST_PYTHON_MODULE(exception_translator_ext)
{
using namespace boost::python;
register_exception_translator&lt;my_exception&gt;(&amp;translate);
def("something_which_throws", something_which_throws);
}
</pre>
<br>
<br>
<hr>
<p>Revised 03 October, 2002</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,230 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Windows (vers 1st August 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - &lt;boost/python/extract.hpp&gt;</title>
</head>
<body>
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Header &lt;boost/python/extract.hpp&gt;</h2>
</td>
</tr>
</table>
<hr>
<h2>Contents</h2>
<dl class="page-index">
<dt><a href="#introduction">Introduction</a></dt>
<dt><a href="#classes">Classes</a></dt>
<dd>
<dl class="page-index">
<dt><a href="#extract-spec">Class <code>extract</code></a></dt>
<dd>
<dl class="page-index">
<dt><a href="#extract-spec-synopsis">Class <code>extract</code>
synopsis</a></dt>
<dt><a href="#extract-spec-ctors">Class <code>extract</code>
constructors and destructor</a></dt>
<dt><a href="#extract-spec-observers">Class
<code>extract</code> observer functions</a></dt>
</dl>
</dd>
</dl>
</dd>
<dt><a href="#examples">Example</a></dt>
</dl>
<hr>
<h2><a name="introduction"></a>Introduction</h2>
<p>Exposes a mechanism for extracting C++ object values from
generalized Python objects. Note that
<code>extract&lt;</code>...<code>&gt;</code> can also be used to
&quot;downcast&quot; an <a
href="object.html#object-spec">object</a> to some specific <a
href="ObjectWrapper.html#ObjectWrapper-concept">ObjectWrapper</a>. Because
invoking a mutable python type with an argument of the same type
(e.g. <code>list([1,2])</code> typically makes a <em>copy</em> of
the argument object, this may be the only way to access the <a
href="ObjectWrapper.html#ObjectWrapper-concept">ObjectWrapper</a>'s
interface on the original object.
<h2><a name="classes"></a>Classes</h2>
<h3><a name="extract-spec"></a>Class template <code>extract</code></h3>
<p><code>extract&lt;T&gt;</code> can be used to extract a value of
an arbitrary C++ type from an instance of <code><a
href="object.html#object-spec">object</a></code>. Two usages are supported:
<ol>
<li><b><code>extract&lt;T&gt;(o)</code></b> is a temporary object
which is implicitly convertible to <code>T</code> (explicit conversion
is also available through the object's function-call
operator). However, if no conversion is available which can convert
<code>o</code> to an object of type <code>T</code>, a Python
<code>TypeError</code> exception will be <a
href="definitions.html#raise">raised</a>.
<li><b><code>extract&lt;T&gt; x(o);</code></b> constructs an extractor
whose <code>check()</code> member function can be used to ask whether
a conversion is available without causing an exception to be thrown.
</ol>
<h4><a name="extract-spec-synopsis"></a>Class template <code>extract</code>
synopsis</h4>
<pre>
namespace boost { namespace python
{
template &lt;class T&gt;
struct extract
{
typedef <i>unspecified</i> result_type;
extract(PyObject*);
extract(object const&amp;);
result_type operator()() const;
operator result_type() const;
bool check() const;
};
}}
</pre>
<h4><a name="extract-spec-ctors"></a>Class <code>extract</code>
constructors and destructor</h4>
<pre>
extract(PyObject* p);
extract(object const&amp;);
</pre>
<dl class="function-semantics">
<dt><b>Requires:</b> The first form requires that <code>p</code> is non-null.</dt>
<dt><b>Effects:</b>Stores a pointer to the Python object managed
by its constructor argument. In particular, the reference
count of the object is not incremented. The onus is on the user
to be sure it is not destroyed before the extractor's conversion
function is called.</dt>
</dl>
<h4><a name="extract-spec-observers"></a>Class <code>extract</code>
observer functions</h4>
<pre>
result_type operator()() const;
operator result_type() const;
</pre>
<dl class="function-semantics">
<dt><b>Effects:</b> Converts the stored pointer to
<code>result_type</code>, which is either <code>T</code> or
<code>T const&amp;</code>.
</dt>
<dt><b>Returns:</b> An object of <code>result_type</code>
corresponding to the one referenced by the stored pointer.</dt>
<dt><b>Throws:</b> <code><a
href="errors.html#error_already_set-spec">error_already_set</a></code>
and sets a <code>TypeError</code> if no such conversion is
available. May also emit other unspecified exceptions thrown by
the converter which is actually used.</dt>
</dl>
<pre>
bool check() const;
</pre>
<dl class="function-semantics">
<dt><b>Postconditions:</b> None. In particular, note that a
return value of <code>true</code> does not preclude an exception
being thrown from <code>operator result_type()</code> or
<code>operator()()</code>.</dt>
<dt><b>Returns:</b> <code>false</code> <i>only</i> if no conversion from the
stored pointer to <code>T</code> is available.</dt>
</dl>
<h2><a name="examples"></a>Examples</h2>
<pre>
#include &lt;cstdio&gt;
using namespace boost::python;
int Print(str s)
{
// extract a C string from the Python string object
char const* c_str = extract&lt;char const*&gt;(s);
// Print it using printf
std::printf(&quot;%s\n&quot;, c_str);
// Get the Python string's length and convert it to an int
return extract&lt;int&gt;(s.attr(&quot;__len__&quot;)())
}
</pre>
The following example shows how extract can be used along with
<code><a
href="class.html#class_-spec">class_</a>&lt;</code>...<code>&gt;</code>
to create and access an instance of a wrapped C++ class.
<pre>
struct X
{
X(int x) : v(x) {}
int value() { return v; }
private:
int v;
};
BOOST_PYTHON_MODULE(extract_ext)
{
object x_class(
class_&lt;X&gt;(&quot;X&quot;, init&lt;int&gt;())
.def(&quot;value&quot;, &amp;X::value))
;
// Instantiate an X object through the Python interface.
// Its lifetime is now managed by x_obj.
object x_obj = x_class(3);
// Get a reference to the C++ object out of the Python object
X&amp; x = extract&lt;X&amp;&gt;(x_obj);
assert(x.value() == 3);
}
</pre>
<p>Revised 15 November, 2002</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002. All Rights
Reserved.</i></p>
</body>
</html>

View File

@@ -1,504 +0,0 @@
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta name="generator" content=
"HTML Tidy for Cygwin (vers 1st April 2002), see www.w3.org">
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<link rel="stylesheet" type="text/css" href="../boost.css">
<title>Boost.Python - FAQ</title>
</head>
<body link="#0000ff" vlink="#800080">
<table border="0" cellpadding="7" cellspacing="0" width="100%" summary=
"header">
<tr>
<td valign="top" width="300">
<h3><a href="../../../../index.htm"><img height="86" width="277"
alt="C++ Boost" src="../../../../c++boost.gif" border="0"></a></h3>
</td>
<td valign="top">
<h1 align="center"><a href="../index.html">Boost.Python</a></h1>
<h2 align="center">Frequently Asked Questions (FAQs)</h2>
</td>
</tr>
</table>
<hr>
<dl class="page-index">
<dt><a href="#dangling">I'm getting the "attempt to return dangling
reference" error. What am I doing wrong?</a></dt>
<dt><a href="#question1">Is return_internal_reference
efficient?</a></dt>
<dt><a href="#question2">How can I wrap functions which take C++
containers as arguments?</a></dt>
<dt><a href="#c1204">fatal error C1204:Compiler limit:internal
structure overflow</a></dt>
<dt><a href="#debugging">How do I debug my Python extensions?</a></dt>
<dt><a href="#imul">Why doesn't my <code>*=</code> operator
work?</a></dt>
<dt><a href="#macosx">Does Boost.Python work with Mac OS X?</a></dt>
<dt><a href="#xref">How can I find the existing PyObject that holds a
C++ object?</a></dt>
<dt><a href="#ownership">How can I wrap a function which needs to take
ownership of a raw pointer?</a></dt>
</dl>
<hr>
<h2><a name="dangling">I'm getting the "attempt to return dangling
reference" error. What am I doing wrong?</a></h2>
That exception is protecting you from causing a nasty crash. It usually
happens in response to some code like this:
<pre>
period const&amp; get_floating_frequency() const
{
return boost::python::call_method&lt;period const&amp;&gt;(
m_self,"get_floating_frequency");
}
</pre>
And you get:
<pre>
ReferenceError: Attempt to return dangling reference to object of type:
class period
</pre>
<p>In this case, the Python method invoked by <code>call_method</code>
constructs a new Python object. You're trying to return a reference to a
C++ object (an instance of <code>class period</code>) contained within
and owned by that Python object. Because the called method handed back a
brand new object, the only reference to it is held for the duration of
<code>get_floating_frequency()</code> above. When the function returns,
the Python object will be destroyed, destroying the instance of
<code>class period</code>, and leaving the returned reference dangling.
That's already undefined behavior, and if you try to do anything with
that reference you're likely to cause a crash. Boost.Python detects this
situation at runtime and helpfully throws an exception instead of letting
you do that.<br>
&nbsp;</p>
<hr>
<h2><a name="question1"></a>Is return_internal_reference efficient?</h2>
<blockquote>
<b>Q:</b> <i>I have an object composed of 12 doubles. A const&amp; to
this object is returned by a member function of another class. From the
viewpoint of using the returned object in Python I do not care if I get
a copy or a reference to the returned object. In Boost.Python Version 2
I have the choice of using copy_const_reference or
return_internal_reference. Are there considerations that would lead me
to prefer one over the other, such as size of generated code or memory
overhead?</i>
<p><b>A:</b> copy_const_reference will make an instance with storage
for one of your objects, size = base_size + 12 * sizeof(double).
return_internal_reference will make an instance with storage for a
pointer to one of your objects, size = base_size + sizeof(void*).
However, it will also create a weak reference object which goes in the
source object's weakreflist and a special callback object to manage the
lifetime of the internally-referenced object. My guess?
copy_const_reference is your friend here, resulting in less overall
memory use and less fragmentation, also probably fewer total
cycles.</p>
</blockquote>
<hr>
<h2><a name="question2"></a>How can I wrap functions which take C++
containers as arguments?</h2>
<p>Ralf W. Grosse-Kunstleve provides these notes:</p>
<ol>
<li>
Using the regular <code>class_&lt;&gt;</code> wrapper:
<pre>
class_&lt;std::vector&lt;double&gt; &gt;("std_vector_double")
.def(...)
...
;
</pre>
This can be moved to a template so that several types (double, int,
long, etc.) can be wrapped with the same code. This technique is used
in the file
<blockquote>
scitbx/include/scitbx/array_family/boost_python/flex_wrapper.h
</blockquote>
in the "scitbx" package. The file could easily be modified for
wrapping std::vector&lt;&gt; instantiations.
<p>This type of C++/Python binding is most suitable for containers
that may contain a large number of elements (&gt;10000).</p>
</li>
<li>
Using custom rvalue converters. Boost.Python "rvalue converters"
match function signatures such as:
<pre>
void foo(std::vector&lt;double&gt; const&amp; array); // pass by const-reference
void foo(std::vector&lt;double&gt; array); // pass by value
</pre>
Some custom rvalue converters are implemented in the file
<blockquote>
scitbx/include/scitbx/boost_python/container_conversions.h
</blockquote>
This code can be used to convert from C++ container types such as
std::vector&lt;&gt; or std::list&lt;&gt; to Python tuples and vice
versa. A few simple examples can be found in the file
<blockquote>
scitbx/array_family/boost_python/regression_test_module.cpp
</blockquote>
Automatic C++ container &lt;-&gt; Python tuple conversions are most
suitable for containers of moderate size. These converters generate
significantly less object code compared to alternative 1 above.
</li>
</ol>
A disadvantage of using alternative 2 is that operators such as
arithmetic +,-,*,/,% are not available. It would be useful to have custom
rvalue converters that convert to a "math_array" type instead of tuples.
This is currently not implemented but is possible within the framework of
Boost.Python V2 as it will be released in the next couple of weeks. [ed.:
this was posted on 2002/03/10]
<p>It would also be useful to also have "custom lvalue converters" such
as std::vector&lt;&gt; &lt;-&gt; Python list. These converters would
support the modification of the Python list from C++. For example:</p>
<p>C++:</p>
<pre>
void foo(std::vector&lt;double&gt;&amp; array)
{
for(std::size_t i=0;i&lt;array.size();i++) {
array[i] *= 2;
}
}
</pre>
Python:
<pre>
&gt;&gt;&gt; l = [1, 2, 3]
&gt;&gt;&gt; foo(l)
&gt;&gt;&gt; print l
[2, 4, 6]
</pre>
Custom lvalue converters require changes to the Boost.Python core library
and are currently not available.
<p>P.S.:</p>
<p>The "scitbx" files referenced above are available via anonymous
CVS:</p>
<pre>
cvs -d:pserver:anonymous@cvs.cctbx.sourceforge.net:/cvsroot/cctbx login
cvs -d:pserver:anonymous@cvs.cctbx.sourceforge.net:/cvsroot/cctbx co scitbx
</pre>
<hr>
<h2><a name="c1204"></a>fatal error C1204:Compiler limit:internal
structure overflow</h2>
<blockquote>
<b>Q:</b> <i>I get this error message when compiling a large source
file. What can I do?</i>
<p><b>A:</b> You have two choices:</p>
<ol>
<li>Upgrade your compiler (preferred)</li>
<li>
Break your source file up into multiple translation units.
<p><code><b>my_module.cpp</b></code>:</p>
<pre>
...
void more_of_my_module();
BOOST_PYTHON_MODULE(my_module)
{
def("foo", foo);
def("bar", bar);
...
more_of_my_module();
}
</pre>
<code><b>more_of_my_module.cpp</b></code>:
<pre>
void more_of_my_module()
{
def("baz", baz);
...
}
</pre>
If you find that a <code><a href=
"class.html#class_-spec">class_</a>&lt;...&gt;</code> declaration
can't fit in a single source file without triggering the error, you
can always pass a reference to the <code>class_</code> object to a
function in another source file, and call some of its member
functions (e.g. <code>.def(...)</code>) in the auxilliary source
file:
<p><code><b>more_of_my_class.cpp</b></code>:</p>
<pre>
void more_of_my_class(class&lt;my_class&gt;&amp; x)
{
x
.def("baz", baz)
.add_property("xx", &amp;my_class::get_xx, &amp;my_class::set_xx)
;
...
}
</pre>
</li>
</ol>
</blockquote>
<hr>
<h2><a name="debugging"></a>How do I debug my Python extensions?</h2>
<p>Greg Burley gives the following answer for Unix GCC users:</p>
<blockquote>
Once you have created a boost python extension for your c++ library or
class, you may need to debug the code. Afterall this is one of the
reasons for wrapping the library in python. An expected side-effect or
benefit of using BPL is that debugging should be isolated to the c++
library that is under test, given that python code is minimal and
boost::python either works or it doesn't. (ie. While errors can occur
when the wrapping method is invalid, most errors are caught by the
compiler ;-).
<p>The basic steps required to initiate a gdb session to debug a c++
library via python are shown here. Note, however that you should start
the gdb session in the directory that contains your BPL my_ext.so
module.</p>
<pre>
(gdb) target exec python
(gdb) run
&gt;&gt;&gt; from my_ext import *
&gt;&gt;&gt; [C-c]
(gdb) break MyClass::MyBuggyFunction
(gdb) cont
&gt;&gt;&gt; pyobj = MyClass()
&gt;&gt;&gt; pyobj.MyBuggyFunction()
Breakpoint 1, MyClass::MyBuggyFunction ...
Current language: auto; currently c++
(gdb) do debugging stuff
</pre>
</blockquote>
<p>Greg's approach works even better using Emacs' "<code>gdb</code>"
command, since it will show you each line of source as you step through
it.</p>
<p>On <b>Windows</b>, my favorite debugging solution is the debugger that
comes with Microsoft Visual C++ 7. This debugger seems to work with code
generated by all versions of Microsoft and Metrowerks toolsets; it's rock
solid and "just works" without requiring any special tricks from the
user.</p>
<p>Unfortunately for Cygwin and MinGW users, as of this writing gdb on
Windows has a very hard time dealing with shared libraries, which could
make Greg's approach next to useless for you. My best advice for you is
to use Metrowerks C++ for compiler conformance and Microsoft Visual
Studio as a debugger when you need one.</p>
<h3>Debugging extensions through Boost.Build</h3>
If you are launching your extension module tests with <a href=
"../../../tools/build">Boost.Build</a> using the
<code>boost-python-runtest</code> rule, you can ask it to launch your
debugger for you by adding "-sPYTHON_LAUNCH=<i>debugger</i>" to your bjam
command-line:
<pre>
bjam -sTOOLS=metrowerks "-sPYTHON_LAUNCH=devenv /debugexe" test
bjam -sTOOLS=gcc -sPYTHON_LAUNCH=gdb test
</pre>
It can also be extremely useful to add the <code>-d+2</code> option when
you run your test, because Boost.Build will then show you the exact
commands it uses to invoke it. This will invariably involve setting up
PYTHONPATH and other important environment variables such as
LD_LIBRARY_PATH which may be needed by your debugger in order to get
things to work right.
<hr>
<h2><a name="imul"></a>Why doesn't my <code>*=</code> operator work?</h2>
<blockquote>
<b>Q:</b> <i>I have exported my class to python, with many overloaded
operators. it works fine for me except the</i> <code>*=</code>
<i>operator. It always tells me "can't multiply sequence with non int
type". If I use</i> <code>p1.__imul__(p2)</code> <i>instead of</i>
<code>p1 *= p2</code><i>, it successfully executes my code. What's
wrong with me?</i>
<p><b>A:</b> There's nothing wrong with you. This is a bug in Python
2.2. You can see the same effect in Pure Python (you can learn a lot
about what's happening in Boost.Python by playing with new-style
classes in Pure Python).</p>
<pre>
&gt;&gt;&gt; class X(object):
... def __imul__(self, x):
... print 'imul'
...
&gt;&gt;&gt; x = X()
&gt;&gt;&gt; x *= 1
</pre>
To cure this problem, all you need to do is upgrade your Python to
version 2.2.1 or later.
</blockquote>
<hr>
<h2><a name="macosx"></a>Does Boost.Python work with Mac OS X?</h2>
<blockquote>
<p>The short answer: as of January 2003, unfortunately not.</p>
<p>The longer answer: using Mac OS 10.2.3 with the December Developer's
Kit, Python 2.3a1, and bjam's darwin-tools.jam, Boost.Python compiles
fine, including the examples. However, there are problems at runtime
(see <a href=
"http://mail.python.org/pipermail/c++-sig/2003-January/003267.html">http://mail.python.org/pipermail/c++-sig/2003-January/003267.html</a>).
Solutions are currently unknown.</p>
<p>It is known that under certain circumstances objects are
double-destructed. See <a href=
"http://mail.python.org/pipermail/c++-sig/2003-January/003278.html">http://mail.python.org/pipermail/c++-sig/2003-January/003278.html</a>
for details. It is not clear however if this problem is related to the
Boost.Python runtime issues.</p>
</blockquote>
<hr>
<h2><a name="xref">How can I find the existing PyObject that holds a C++
object?</a></h2>
<blockquote>
"I am wrapping a function that always returns a pointer to an
already-held C++ object."
</blockquote>
One way to do that is to hijack the mechanisms used for wrapping a class
with virtual functions. If you make a wrapper class with an initial
PyObject* constructor argument and store that PyObject* as "self", you
can get back to it by casting down to that wrapper type in a thin wrapper
function. For example:
<pre>
class X { X(int); virtual ~X(); ... };
X* f(); // known to return Xs that are managed by Python objects
// wrapping code
struct X_wrap : X
{
X_wrap(PyObject* self, int v) : self(self), X(v) {}
PyObject* self;
};
handle&lt;&gt; f_wrap()
{
X_wrap* xw = dynamic_cast&lt;X_wrap*&gt;(f());
assert(xw != 0);
return handle&lt;&gt;(borrowed(xw-&gt;self));
}
...
def("f", f_wrap());
class_&lt;X,X_wrap&gt;("X", init&lt;int&gt;())
...
;
</pre>
Of course, if X has no virtual functions you'll have to use
<code>static_cast</code> instead of <code>dynamic_cast</code> with no
runtime check that it's valid. This approach also only works if the
<code>X</code> object was constructed from Python, because
<code>X</code>s constructed from C++ are of course never
<code>X_wrap</code> objects.
<p>Another approach to this requires you to change your C++ code a bit;
if that's an option for you it might be a better way to go. work we've
been meaning to get to anyway. When a <code>shared_ptr&lt;X&gt;</code> is
converted from Python, the shared_ptr actually manages a reference to the
containing Python object. When a shared_ptr&lt;X&gt; is converted back to
Python, the library checks to see if it's one of those "Python object
managers" and if so just returns the original Python object. So you could
just write <code>object(p)</code> to get the Python object back. To
exploit this you'd have to be able to change the C++ code you're wrapping
so that it deals with shared_ptr instead of raw pointers.</p>
<p>There are other approaches too. The functions that receive the Python
object that you eventually want to return could be wrapped with a thin
wrapper that records the correspondence between the object address and
its containing Python object, and you could have your f_wrap function
look in that mapping to get the Python object out.</p>
<h2><a name="ownership">How can I wrap a function which needs to take
ownership of a raw pointer?</a></h2>
<blockquote>
<i>Part of an API that I'm wrapping goes something like this:</i>
<pre>
struct A {}; struct B { void add( A* ); }
where B::add() takes ownership of the pointer passed to it.
</pre>
<p><i>However:</i></p>
<pre>
a = mod.A()
b = mod.B()
b.add( a )
del a
del b
# python interpreter crashes
# later due to memory corruption.
</pre>
<p><i>Even binding the lifetime of a</i> to b via
with_custodian_and_ward doesn't prevent the python object a from
ultimately trying to delete the object it's pointing to. Is there a way
to accomplish a 'transfer-of-ownership' of a wrapped C++ object?</p>
<p><i>--Bruce Lowery</i></p>
</blockquote>
Yes: Make sure the C++ object is held by auto_ptr:
<pre>
class_&lt;A, std::auto_ptr&lt;A&gt; &gt;("A")
...
;
</pre>
Then make a thin wrapper function which takes an auto_ptr parameter:
<pre>
void b_insert(B&amp; b, std::auto_ptr&lt;A&gt; a)
{
b.insert(a.get());
a.release();
}
</pre>
Wrap that as B.add. Note that pointers returned via <code><a href=
"manage_new_object.html#manage_new_object-spec">manage_new_object</a></code>
will also be held by <code>auto_ptr</code>, so this transfer-of-ownership
will also work correctly.
<hr>
<p>Revised
<!--webbot bot="Timestamp" S-Type="EDITED" S-Format="%d %B, %Y" startspan -->
23 January, 2003
<!--webbot bot="Timestamp" endspan i-checksum="39359" -->
</p>
<p><i>&copy; Copyright <a href=
"../../../../people/dave_abrahams.htm">Dave Abrahams</a> 2002-2003. All
Rights Reserved.</i></p>
</body>
</html>

Some files were not shown because too many files have changed in this diff Show More