2
0
mirror of https://github.com/boostorg/build.git synced 2026-02-22 03:12:16 +00:00
Commit Graph

10944 Commits

Author SHA1 Message Date
Vladimir Prus
a2322a224d Unbreak darwin, after darwin solib suffix was moved to types/lib.jam.
Patch from Craig Rodrigues.


[SVN r29242]
2005-05-27 14:53:08 +00:00
Rene Rivera
30692a4049 Fix building of bjam itself after floating options change. Needed to make sure that we eat up the multi arg options if they exist.
[SVN r29239]
2005-05-27 03:48:04 +00:00
Rene Rivera
a97e4369a4 Fix building of bjam itself after floating options change. Needed to make sure that we eat up the multi arg options if they exist.
[SVN r29239]
2005-05-27 03:48:04 +00:00
Dave Abrahams
0b205d30d3 Eliminate undefined behavior and fix a spelling error.
[SVN r29238]
2005-05-27 03:18:08 +00:00
Dave Abrahams
f97111f4cb Eliminate undefined behavior and fix a spelling error.
[SVN r29238]
2005-05-27 03:18:08 +00:00
Dave Abrahams
815f689d3b Desperate attempt to avoid confusion between the Boost distribution and the Boost.Build distribution.
[SVN r29237]
2005-05-26 21:44:31 +00:00
Dave Abrahams
7810e36929 Desperate attempt to avoid confusion between the Boost distribution and the Boost.Build distribution.
[SVN r29237]
2005-05-26 21:44:31 +00:00
Dave Abrahams
0603fa3563 No config checking or warnings if --without-python is passed.
[SVN r29236]
2005-05-26 20:48:34 +00:00
Rene Rivera
b31b3bd5d4 features.jam; Add "aix" ABI object model for building on AIX64, etc.
gcc-tools.jam; Support some of the new RS/PPC processors, and fix the building of 64 bit in those architectures.


[SVN r29235]
2005-05-26 20:17:02 +00:00
Rene Rivera
7e66e867b3 Add vc-x_x toolsets to normalize the names of the toolsets use by users.
[SVN r29234]
2005-05-26 20:15:08 +00:00
Dave Abrahams
0125e21817 Tell non-python users they can ignore the Boost.Python message
[SVN r29233]
2005-05-26 19:17:26 +00:00
Rene Rivera
a9e2ffa6a1 Remove restriction that option arguments need to be before target arguments. Options can now be intermixed with targets.
[SVN r29232]
2005-05-26 17:20:32 +00:00
Rene Rivera
40a630e54b Remove restriction that option arguments need to be before target arguments. Options can now be intermixed with targets.
[SVN r29232]
2005-05-26 17:20:32 +00:00
Dave Abrahams
7cb7380cc3 Fix manifest generation yet again.
[SVN r29231]
2005-05-26 17:12:59 +00:00
Vladimir Prus
3da085194d Unroll incomplete changes to response files support.
I think that Dave committed then unintentionally in revision 1.36 and they
cause C++ files to be always recompiled. A better version is in the works
but cannot be committed yet, so I'll just restore the previous one for now.


[SVN r29198]
2005-05-26 05:54:16 +00:00
Vladimir Prus
05d64fbddb Unroll incomplete changes to response files support.
I think that Dave committed then unintentionally in revision 1.36 and they
cause C++ files to be always recompiled. A better version is in the works
but cannot be committed yet, so I'll just restore the previous one for now.


[SVN r29198]
2005-05-26 05:54:16 +00:00
Dave Abrahams
b02c3921b0 Only do manifest generation for python modules
[SVN r29194]
2005-05-25 19:18:13 +00:00
Dave Abrahams
348f90824b This change was required in order to get it to work on my installations.
[SVN r29193]
2005-05-25 19:04:12 +00:00
Dave Abrahams
6b1d811f50 This change was required in order to get it to work on my installations.
[SVN r29193]
2005-05-25 19:04:12 +00:00
Dave Abrahams
0d74108229 manifest generation
[SVN r29192]
2005-05-25 19:03:32 +00:00
Dave Abrahams
d9ce5913a0 manifest generation
[SVN r29192]
2005-05-25 19:03:32 +00:00
Vladimir Prus
411d1931dc Remove floatsam.
[SVN r29190]
2005-05-25 14:32:37 +00:00
Vladimir Prus
860e5915b4 Remove floatsam.
[SVN r29190]
2005-05-25 14:32:37 +00:00
Vladimir Prus
648a7b2970 Do no call sequence.unique on result of GLOB-RECURSIVELY.
That was an old hack which was only needed in one context -- when we
passed both Jamroot and jamroot as patterns, got back Jamroot and jamroot
(because windows is case-insensitive and reports that both exist), and get
confused.

Calling 'unique' does not harm in most cases, but in Boost Jamfiles we glob
for all headers, and O(n^2) algorithm used by unique considerably slows us.


[SVN r29187]
2005-05-25 13:51:33 +00:00
Vladimir Prus
b54d80c3e2 Do no call sequence.unique on result of GLOB-RECURSIVELY.
That was an old hack which was only needed in one context -- when we
passed both Jamroot and jamroot as patterns, got back Jamroot and jamroot
(because windows is case-insensitive and reports that both exist), and get
confused.

Calling 'unique' does not harm in most cases, but in Boost Jamfiles we glob
for all headers, and O(n^2) algorithm used by unique considerably slows us.


[SVN r29187]
2005-05-25 13:51:33 +00:00
Vladimir Prus
bb2104ccb6 Fix error when installing with <install-source-root> and source file names
are absolute.


[SVN r29180]
2005-05-25 06:10:23 +00:00
Vladimir Prus
7ef55e8276 Fix error when installing with <install-source-root> and source file names
are absolute.


[SVN r29180]
2005-05-25 06:10:23 +00:00
Vladimir Prus
bc159afedb Clarify comments.
[SVN r29179]
2005-05-25 05:55:10 +00:00
Vladimir Prus
708c904345 Clarify comments.
[SVN r29179]
2005-05-25 05:55:10 +00:00
Dave Abrahams
a02a2e7018 Made it work with an unmodified CWPro8 installation. Pro9 may still need its setup script tweaked.
[SVN r29174]
2005-05-24 17:13:02 +00:00
Dave Abrahams
95ffc9b080 Made it work with an unmodified CWPro8 installation. Pro9 may still need its setup script tweaked.
[SVN r29174]
2005-05-24 17:13:02 +00:00
Dave Abrahams
3219877dd7 diagnose binding non-const reference to rvalue as a hard error.
[SVN r29173]
2005-05-24 17:11:30 +00:00
Dave Abrahams
4d30e08d12 diagnose binding non-const reference to rvalue as a hard error.
[SVN r29173]
2005-05-24 17:11:30 +00:00
Toon Knapen
fdbb4ae754 added capability to handle <fflags>
[SVN r29171]
2005-05-24 15:27:30 +00:00
Toon Knapen
ec44facdd2 added capability to handle <fflags>
[SVN r29171]
2005-05-24 15:27:30 +00:00
Dave Abrahams
ed4953f961 Added version-8.0-path to the list of possible-paths; its omission can't possibly have been intentional.
[SVN r29168]
2005-05-24 13:27:02 +00:00
Dave Abrahams
635a997daa Added version-8.0-path to the list of possible-paths; its omission can't possibly have been intentional.
[SVN r29168]
2005-05-24 13:27:02 +00:00
Vladimir Prus
cdf5cbe153 Allow the 'source-location' attribute to list several directories.
Patch from Alexander Kabaev and Craig Rodrigues.


[SVN r29164]
2005-05-24 08:57:52 +00:00
Vladimir Prus
f5ba4db1f7 Allow the 'source-location' attribute to list several directories.
Patch from Alexander Kabaev and Craig Rodrigues.


[SVN r29164]
2005-05-24 08:57:52 +00:00
Vladimir Prus
4e9af774ce Revamp path.glob.
The problem with the previous implementation is that it would iterate
over all elements in a dir even if 'pattern' had no metacharacters. First,
that was slow -- if you handed

  /cygdrive/w/My Documents/boost/test/a.cpp

to it, it would crawl all the way to the top, listing each directory and
checking each file. Second, it would fail, because Cygwin is broken and
does not show 'cygdrive' is the list of directory elements of '/'.

Now we check if a pattern has metacharacters, and if not, just do
a simple 'timestamp' call.

The new glob is implemented as new 'GLOB-RECURSIVELY' builtin. I've decided
to use builtin since otherwise, we'd need 'does this name exist' builtin,
and if we need new builtin, why don't implement all globbing in core.


[SVN r29163]
2005-05-24 07:57:56 +00:00
Vladimir Prus
a9144a5ec3 Revamp path.glob.
The problem with the previous implementation is that it would iterate
over all elements in a dir even if 'pattern' had no metacharacters. First,
that was slow -- if you handed

  /cygdrive/w/My Documents/boost/test/a.cpp

to it, it would crawl all the way to the top, listing each directory and
checking each file. Second, it would fail, because Cygwin is broken and
does not show 'cygdrive' is the list of directory elements of '/'.

Now we check if a pattern has metacharacters, and if not, just do
a simple 'timestamp' call.

The new glob is implemented as new 'GLOB-RECURSIVELY' builtin. I've decided
to use builtin since otherwise, we'd need 'does this name exist' builtin,
and if we need new builtin, why don't implement all globbing in core.


[SVN r29163]
2005-05-24 07:57:56 +00:00
Vladimir Prus
c7bce34d14 Don't use 'jam' in tests -- it's not built by default now.
[SVN r29157]
2005-05-23 11:09:31 +00:00
Vladimir Prus
c6b899c0ea Don't use 'jam' in tests -- it's not built by default now.
[SVN r29157]
2005-05-23 11:09:31 +00:00
Vladimir Prus
63a403d05c Some commentary.
[SVN r29154]
2005-05-23 06:35:28 +00:00
Vladimir Prus
5eb52082ab Some commentary.
[SVN r29154]
2005-05-23 06:35:28 +00:00
Rene Rivera
506b018fb7 Adjust toolset tag for CodeWarrior to match the names generated by the autolink code.
[SVN r29153]
2005-05-23 05:00:20 +00:00
Dave Abrahams
f9c8f3e34a Support the required manifest generation step for vc8.0
[SVN r29143]
2005-05-22 10:45:44 +00:00
Toon Knapen
d8a85aca97 xlf compiler has no preprocessing and the -D option is used for soth. else
[SVN r29056]
2005-05-19 06:51:42 +00:00
Toon Knapen
88369144fe xlf compiler has no preprocessing and the -D option is used for soth. else
[SVN r29056]
2005-05-19 06:51:42 +00:00
Douglas Gregor
00d6aaee47 Fix Python detection for darwin-derived toolsets
[SVN r29021]
2005-05-18 03:40:06 +00:00