Gennaro Prota
9bba7af0b5
Avoid a redundant first != last check for append< BlockInputIterator >() with input iterators
...
Note that this also avoids a BOOST_ASSERT().
2025-10-15 11:41:43 +02:00
Gennaro Prota
4cf19a2b0a
Fix append< BlockInputIterator >() for input iterators
...
We were using `AllocatorOrContainer` as if it were still guaranteed to
be an allocator type. Unfortunately, the append() tests didn't exercise
input iterators, so this went unnoticed. I've now added tests.
2025-10-15 11:41:43 +02:00
Gennaro Prota
a740393527
Remove extra parentheses in the implementation of operator<()
...
Reason: Consistency.
2025-10-15 11:41:43 +02:00
Gennaro Prota
7ae245eabf
Add missing spaces in a comment
...
Reason: Consistency.
2025-10-15 11:41:43 +02:00
Gennaro Prota
9d604eba5a
Simplify the implementation of all()
2025-10-15 11:41:43 +02:00
Gennaro Prota
fae55af67d
Remove a reference to a removed comment
...
The referenced comment was removed when porting the documentation to
MrDocs. It advised against using the constructor from basic_string, but,
perhaps, that advice was a bit too strong (constructing from strings is
handy, and probably faster than inputting from a stringstream).
2025-10-15 11:41:43 +02:00
Gennaro Prota
81011d8161
Add trailing periods to the \pre paragraphs that missed them
...
Reason: Consistency.
2025-10-15 11:41:43 +02:00
Gennaro Prota
51db99cf32
Fix a few documentation glitches
2025-10-15 11:41:43 +02:00
Gennaro Prota
ac6b01140d
Make the three "from string" constructors explicit
...
Making them non-explicit was an oversight.
2025-10-15 11:41:43 +02:00
Gennaro Prota
26dac17a4a
Settle the documentation of the iterators
...
We choose to hide the iterator type definitions and just document their
category, which is enough.
2025-10-15 11:41:43 +02:00
Gennaro Prota
d3c4e2751a
Improve the encapsulation of bit_iterator_base
2025-10-15 11:41:43 +02:00
Gennaro Prota
c58487c26a
Fix an error in the documentation of operator==()
2025-10-15 11:41:43 +02:00
Gennaro Prota
6f3634aa4b
Remove the dependency on core/allocator_access.hpp
2025-10-15 11:41:43 +02:00
Gennaro Prota
d8365ee015
Make the max_size() tests a bit more meaningful
2025-10-15 11:41:43 +02:00
Gennaro Prota
4ca26518ec
Add a missing comment
2025-10-15 11:41:43 +02:00
Gennaro Prota
8f07dd356a
Add push_front() and pop_front()
2025-10-15 11:41:43 +02:00
Gennaro Prota
efd4d41f60
No longer use BOOST_NO_CXX11_RVALUE_REFERENCES
...
Reason: We require decent C++11 support anyway.
2025-10-15 11:41:43 +02:00
Gennaro Prota
30ee0f68b5
Add two blank lines
...
Reason: Readability, consistency.
2025-10-15 11:41:43 +02:00
Gennaro Prota
eba192e3a3
Make the descriptions of the preconditions of the three "from string" constructors more precise
2025-10-15 11:41:43 +02:00
Gennaro Prota
bb489c186a
Add a constructor from basic_string_view in C++17 or later
...
Note that we have MrDocs compile as C++17 again (a change that we had
reverted).
2025-10-15 11:41:43 +02:00
Gennaro Prota
0177f4adfa
Add a constructor from const CharT * (pointer to C-style string)
...
This is in preparation of the next commit, which will add a constructor
from std::basic_string_view (otherwise, construction from a C-style
string would be ambiguous between the basic_string_view and the
basic_string constructor).
Note that, rather than duplicating code to test the new member, we
changed many constructions in dyn_bitset_unit_tests1.cpp to use const
char *, instead of std::string. (The constructor from std::basic_string
is tested separately anyway.)
2025-10-15 11:41:43 +02:00
Gennaro Prota
079cff781d
Don't test for Tr::eq( c, one ) twice in init_from_string()
2025-10-15 11:41:43 +02:00
Gennaro Prota
e00d9b30ae
No longer use boost::move()
...
Reason: We require C++11 now, so we can use std::move().
2025-10-15 11:41:43 +02:00
Gennaro Prota
ae9703b119
Actually test the move constructor in the move constructor tests
...
We were testing the *copy* constructor, instead.
2025-10-15 11:41:43 +02:00
Gennaro Prota
3ccf097dc1
Replace a #include of a deprecated include file
2025-10-15 11:41:43 +02:00
Gennaro Prota
88995c08b0
Remove a superfluous #include directive
2025-10-15 11:41:43 +02:00
Gennaro Prota
3531f6703b
Add missing const qualifiers
2025-10-15 11:41:43 +02:00
Gennaro Prota
0ff8509214
Make lowest_bit() more efficient
...
As a nice bonus, this also removes the dependency on Boost.Integer.
2025-10-15 11:41:43 +02:00
Gennaro Prota
08191fc025
Remove a spurious space
2025-10-15 11:41:43 +02:00
Gennaro Prota
759c4f4e4a
Fix a misleading variable name
2025-10-15 11:41:43 +02:00
Gennaro Prota
12dfcea22f
Get rid of BOOST_STATIC_ASSERT() and BOOST_STATIC_CONSTANT()
...
Reason: They are no longer necessary, given that we require C++11 now.
2025-10-15 11:41:43 +02:00
Gennaro Prota
526104dbe9
Assert that Block is the same type as AllocatorOrContainer::value_type
...
Reason: That's required both when AllocatorOrContainer is an allocator
type and when it's a container type. Now, when it is an allocator type,
the assert is usually in the implementation of std::vector, which we
use. But, when it's a container type, we need to assert that on our own.
2025-10-15 11:41:43 +02:00
Gennaro Prota
b6a875976d
Remove the parameter-less overloads of find_first() and find_first_off()
...
Reason: We can use a default argument, instead, which is what
dynamic_bitset usually does.
2025-10-15 11:41:43 +02:00
Gennaro Prota
394854535a
Make pop_back() more efficient
2025-10-15 11:41:43 +02:00
Gennaro Prota
1bef343d12
Back out the support for operator<=> in our iterators
...
Reason: It required operator<=> for the underlying container, and
boost::container::small_vector doesn't support it.
2025-10-15 11:41:42 +02:00
Gennaro Prota
0ca460971e
Make buffer_type public
...
Reason: Failures from GCC and Clang. See the new code comment.
2025-10-15 11:41:42 +02:00
Gennaro Prota
6747f81566
Remove some outdated macros (workarounds for MSVC 6)
2025-10-15 11:41:42 +02:00
Gennaro Prota
b428ca6365
Remove another workaround for CodeWarrior
...
Reason: Obsolete compiler.
2025-10-15 11:41:14 +02:00
Gennaro Prota
35b0094529
Add C++20 iterators
...
Reason: Requested on the mailing list.
2025-10-14 19:24:47 +02:00
Gennaro Prota
1b8ef09564
Allow choosing the underlying container type
...
The user can now choose the underlying container type used by
dynamic_bitset. This allows e.g. having small buffer optimization by
using boost::container::small_vector.
Since the underlying container is no longer guaranteed to be
std::vector, we revert "Reflect some noexcept specifications of
std::vector in dynamic_bitset" and the related "Let MrDocs compile the
code as C++17".
Note that we didn't add serialization tests (dyn_bitset_unit_tests5.cpp)
because boost::container::small_vector has no serialization support.
This closes issue #76 .
2025-10-14 19:24:47 +02:00
Gennaro Prota
e194055264
Add find_first_off(), find_first_off( size_type ) and find_next_off( size_type )
2025-10-14 19:24:47 +02:00
Gennaro Prota
54b32707c4
Don't repeat the preprocessing condition to specialize std::hash
2025-10-14 19:24:47 +02:00
Alan de Freitas
07e110d522
Make the MrDocs documentation multipage
2025-10-14 19:24:47 +02:00
Alan de Freitas
17e2192154
Support CMake root mode
2025-10-14 19:24:47 +02:00
Gennaro Prota
4a928c8787
Ensure all #includes are at the top of the source file
2025-10-14 19:24:47 +02:00
Gennaro Prota
f429a0a50d
Document our support for std::hash
2025-10-14 19:24:47 +02:00
Gennaro Prota
7f4078b1a9
Remove some symbol filters from mrdocs.yml
...
Reason: They no longer seem necessary.
2025-10-14 19:24:46 +02:00
Gennaro Prota
043b5e1c86
Reword a comment using singular they
2025-10-14 19:24:46 +02:00
Gennaro Prota
0052e61ccc
Document that max_size() doesn't emit exceptions
2025-10-14 19:24:46 +02:00
Gennaro Prota
f262bea730
Use noexcept instead of BOOST_NOEXCEPT
...
Reason: We require C++11 support anyway.
2025-10-14 19:24:46 +02:00