- Boost - lists.boost.org

[old-shmem] Updated boost-sandbox cvs with bug-fixes
by Ion Gaztañaga 18 Mar '06

18 Mar '06

Hi to all, Several people have reported some bugs against Shmem 0.9 that I've tried to correct: -> deque not including allocator_utilities -> named_semaphore bug in windows in timed_wait -> broken named_mutex in unix -> broken try_receive in shared_message_queue Since shmem -> interprocess transformation will need some time I will try to fix Shmem 0.9 bugs in sandbox-cvs. Regards, Ion

1 0

[serialisation] reading and writing reals
by John Maddock 18 Mar '06

18 Mar '06

Attached is a sketch of how reals can be serialised in hex notation, without data loss when round-tripping from binary-to-text-to-binary. Note: 1) The iostream code could be improved, particularly formatting errors on reading, if any iostream experts would like to give it the once over they're very welcome. 2) There is no check on reading that the target type is large enough to hold all the bits that were previously written out (neither is there for decimal io come to that). 3) I've only tested this code with VC8, so portability may vary. 4) I've made no attempt to beautify the output format, it's frankly rather ugly to read at present. 5) It's not clear whether on output the digit before the decimal point should always be a 1: at present it's a full hex digit, which means the value "1" comes out as 0x8p-3 rather than 0x1p0 6) Zero still needs to handled as a special case, at present it prints as 0x0p-3 rather than 0x0p0: it makes no difference on input, but it would be nice to print it prettily :-) 7) I've been unable to reproduce the original problem with native << and >> operators with either random value tests, or with sequential nextafter tests, unless I start with the specific value that caused all the fuss in the first place :-( Any ideas for better tests much appreciated. Enjoy, John.

2 3

[ANN] Wave V1.2.4 in CVS
by Hartmut Kaiser 18 Mar '06

18 Mar '06

Hi all, I've just committed the version of Wave (V1.2.4) which will be part of the upcomming Boost release. It has a bunch of new features in it (with kind approval by Thomas Witt) and some bugfixes applied making it worth to have a look at: - automatic include guard detection - full interactive mode of the Wave tool, now you can use the Wave tool just like Python or Perl for instance to interactively try out your BOOST_PP macros - load/save macro tables to store the state of interactive sessions (that's implemented using the Boost.Serialization library) - performance improvements by upto 30-40% - new pragmas allowing to control certain library features from inside the preprocessed sources (partial output redirection, control of generated whitespace and #line directives) - and bunch of fixes For a full list of changes please look at: http://boost-consulting.com/boost/libs/wave/ChangeLog Feedback is as always welcome! Regards Hartmut

1 0

missing __volatile__ tags in sp_counted_base_gcc_xxx.hpp
by Howard Hinnant 18 Mar '06

18 Mar '06

Hello, I've been investigating some failures on Mac OS X, ppc concerning shared_ptr. One of the things that has struck me is that in boost/ detail/sp_counted_base_gcc_ppc.hpp, the __asm__ block in atomic_decrement is decorated with __volatile__, but the __asm__ blocks in atomic_increment and atomic_conditional_increment are not. The failure I'm seeing disappears when the __volatile__ attribute is added in the two missing places. Would this be a proper fix to boost/ detail/sp_counted_base_gcc_ppc.hpp? Should boost/detail/ sp_counted_base_gcc_x86.hpp be treated similarly? Thanks, Howard

2 8

[Boost-bugs] [ boost-Patches-1452743 ] [regex] fixes warning on mingw
by SourceForge.net 17 Mar '06

17 Mar '06

Patches item #1452743, was opened at 2006-03-17 22:00 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=307586&aid=1452743&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Baraclese (baraclese) Assigned to: Nobody/Anonymous (nobody) Summary: [regex] fixes warning on mingw Initial Comment: [regex] [current cvs] mingw automatically defines NOMINMAX in bits/ os_defines.h to protect some standard headers. Compiler thus emits a warning about redefinition of NOMINMAX in static_mutex.cpp. [solution] Add #undef NOMINMAX before #define NOMINMAX to file libs\regex\src\static_mutex.cpp(27) Other files which define this are: libs\regex\src\regex.cpp(31) libs\regex\src\w32_regex_traits.cpp(27) ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=307586&aid=1452743&group_… ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642 _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

[filesystem] Some comments
by Martin Adrian 17 Mar '06

17 Mar '06

1. The exception specifications are not clear - All predicates (exists, is_*) will throw on errors but you need to look into the status functions to see it. - What happens if an error occurs while iteratoring a directory or when the recursive iterator can't descend into a sub-folder. 2. Why not include the system_error code in the file_status object? If you want to avoid exceptions with the currenct code you need to write code like: system_error_code ec; file_status f(status(p, ec)); if (exists(f) && is_directory(f)) .. else if (status_known(f)) cout << "path is not a directory"; else cout << system_error_message(ec); // can't find this function in the TR2 and // I don't care what the problem is. system_error_code ec; if (is_regular(status(p, ec))) ... Which is kind of messy since the ec and f can't be scoped. It also seem strange that you need to define a system_error_code variable just to avoid exceptions. With a file_status object defined as struct file_status { explicit file_status(file_type v = status_unknown, system_error_code e = 0); file_type type() const; system_error_code error_code() const; std::string system_error_message() const; operator safe_bool() const { return ec == 0; } }; the code can be written as: if (file_status f(p, std::nothrow_t)) if (exists(f) && is_directory(f)) ... else cout << "Path is not a directory"; else cout << f.system_error_message(); and if (is_regular(status(p, std::nothrow_t))) ... 3. Why not differentiate storage type and interface type in the basic_path class?

2 2

[Boost-bugs] [ boost-Patches-1452698 ] iostreams // file_descriptor::seek BUG on files > 4 GB
by SourceForge.net 17 Mar '06

17 Mar '06

Patches item #1452698, was opened at 2006-03-17 12:08 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=307586&aid=1452698&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Nobody/Anonymous (nobody) Summary: iostreams // file_descriptor::seek BUG on files > 4 GB Initial Comment: Boost 1.33.1 iostreams library file_descriptor::seek() method file_descriptor.cpp, line 198 Win32, Visual C++ 2005 The method seek() can return a wrong value of a newly set offset when the offset is further than 4 GB. //--------------- return offset_to_position((lDistanceToMoveHigh << 32) + dwResultLow); //--------------- The code above has a bug. On 32-bit systems the expression (lDistanceToMoveHigh << 32) equals to lDistanceToMoveHigh. The correct code should first cast the lDistanceToMoveHigh variable to 64-bit and then shift it. //--------------- return (static_cast<boost::intmax_t> lDistanceToMoveHigh) << 32) + dwResultLow; //--------------- Sergey Kolodkin ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=307586&aid=1452698&group_… ------------------------------------------------------- This SF.Net email is sponsored by xPML, a groundbreaking scripting language that extends applications into web and mobile media. Attend the live webcast and join the prime developer group breaking into this new coding territory! http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642 _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

what do you get when you cross MPL metadata and boost::variant?
by Korcan Hussein 17 Mar '06

17 Mar '06

Enumerated Types (& simulated pattern matching): #include <iostream> #include <boost/mpl/size_t.hpp> #include <boost/variant.hpp> namespace mpl = boost::mpl; struct APPLES_ : mpl::size_t<0u> {} const APPLES; struct ORANGES_ : mpl::size_t<1u> {} const ORANGES; struct PEACHES_ : mpl::size_t<2u> {} const PEACHES; typedef boost::variant<APPLES_, ORANGES_, PEACHES_> fruit_type; struct fruit_printer : boost::static_visitor<void> { void operator()(APPLES_ a) const { std::cout << "apples: " << a << '\n'; } template < typename Tp > void operator()(Tp v) const { // catch all std::cout << typeid(Tp).name() << ": " << v << '\n'; } } const print_fruit; int main() { fruit_type my_fruit = APPLES; boost::apply_visitor(print_fruit, my_fruit); my_fruit = ORANGES; boost::apply_visitor(print_fruit, my_fruit); my_fruit = PEACHES; boost::apply_visitor(print_fruit, my_fruit); // my_fruit = mpl::size_t<4u>(); // compile-time error // my_fruit = 6u; // compile-time error } There might be some merit here such that generalization is applicable with some macro metaprogramming.

1 0

[numeric cast] Borland 2006 patch
by AlisdairM 17 Mar '06

17 Mar '06

Workarounds for old compiler are still necessary, simple update of version check should suffice: cvs diff -u -wb -- boost\numeric\conversion\cast.hpp (in directory E:\sourceforge\devel\boost\) Index: boost/numeric/conversion/cast.hpp =================================================================== RCS file: /cvsroot/boost/boost/boost/numeric/conversion/cast.hpp,v retrieving revision 1.8 diff -u -w -b -r1.8 cast.hpp --- boost/numeric/conversion/cast.hpp 27 Jun 2005 15:05:12 -0000 1.8 +++ boost/numeric/conversion/cast.hpp 19 Feb 2006 12:19:22 -0000 @@ -22,7 +22,7 @@ #include <boost/detail/workaround.hpp> -#if BOOST_WORKAROUND(BOOST_MSVC, < 1300) || BOOST_WORKAROUND(__BORLANDC__, == 0x564) +#if BOOST_WORKAROUND(BOOST_MSVC, < 1300) || BOOST_WORKAROUND(__BORLANDC__, BOOST_TESTED_AT(0x581)) # include<boost/numeric/conversion/detail/old_numeric_cast.hpp> OK to commit? -- AlisdairM

2 2

[thread] monitor
by Vaclav Vesely 17 Mar '06

17 Mar '06

Hi, I all the time use mutexs and conditions in in following manner: // declaration Type m_value; mutex m_value_mutex; condition m_value_condition; // set new value { mutex::scoped_lock lock(m_mutex); m_value = new_value; } // wait for some condition { mutex::scoped_lock lock(m_mutex); while(<condition>) m_condition.wait(lock); } I'm wondering about something like this: // declaration monitor<Type> m_value; // set new value - locking in done in a monitor assignment m_value = new_value // wait for some condition m_value.wait(<lambda expression>) Do you think this concept is generally useful and should be added to the library? Regards, Vaclav

2 1