- Boost - lists.preview.boost.org

2nd try, New library proposal: PolicyLoader
by Rob Meijer 18 Nov '09

18 Nov '09

I posted this to this list ones before (as the site sugests), and didn't get any response, possibly becource it is simply to useless for you guys to reply to, but I'm not sure. If it is indeed useless please let me know, so I know not to put any effort in it. The Open Computer Forensics Architecture Library (OcfaLib) hides a litle piece of template code for loading a class from a dynamically loaded library. Although this code is currently bound to OcfaLib and usage on *NIX systems, I wondered if the concept used would not be extendable to a generically usable library for this purpose. I've tried to strip the OCFA specific code to show the simple templating concept of this library. I would like to know if it is a good idea to work on this code in order to make it a powerfull cross-platform boost library that addresses this issue in an even more generic way. If it is useless, please also let me know. First lets show an example of usage: typedef SinglePointerConstructor<AbstractRepository,const char> CRepositoryType; PolicyLoader<CRepositoryType> pl(myLib,"createRepository"); CRepositoryType *constructor=pl.constructor(); _instance = (*constructor)(repRoot.c_str()); Here the 'AbstractRepository *' type _instance gets assigned a new object of a class that implements AbstractRepository to it, that is created with the "createRepository" function in the so file designated by myLib. Here is the stripped down template code: ---------------------------------- template <class T,class U> class SinglePointerConstructor { public: typedef T * (*fp)(U *); SinglePointerConstructor (fp func) : mFunction(func) {} T *operator() (U *val) { return (mFunction)(val); } private: fp mFunction; }; template <class T> class VoidToFunction { public: static T *cast(void *p) { #ifdef VOID_FP_CAST_WORKAROUND #ifdef POINTERS_64BIT return new T(reinterpret_cast<typename T::fp>(reinterpret_cast<long long>(p))); #else return new T(reinterpret_cast<typename T::fp>(reinterpret_cast<int>(p))); #endif #else return new T(reinterpret_cast<typename T::fp>(p)); #endif }; }; class PolicyLoader: { public: PolicyLoader(std::string myLib,std::string loadername):mLibHandle(0),mConstructor(0) { mLibHandle=dlopen(myLib.c_str(),RTLD_NOW|RTLD_GLOBAL); if (mLibHandle == 0) { throw std::runtime_error( std::string("Problem loading library '") + myLib + std::string("' (") + std::string(dlerror()) + std::string(")") ); } mConstructor = VoidToFunction<TConstructor>::cast(dlsym(mLibHandle,loadername.c_str())); if (mConstructor == 0) { std::string errString=std::string("Problem looking up symbol '") + loadername + std::string("' in library '") + myLib + std::string("' (") + std::string(dlerror()) + std::string(")"); dlclose(mLibHandle); throw std::runtime_error(errString); } } ~PolicyLoader() { delete mConstructor; } void closeLib() { if (mLibHandle) { dlclose(mLibHandle); } mLibHandle=0; } TConstructor *constructor() { return mConstructor; } private: void *mLibHandle; TConstructor *mConstructor; }; -----------------

3 3

[foreach] Request for auto type deduction
by Andrzej Krzemienski 18 Nov '09

18 Nov '09

Hi, Boost already provides a tool for emulating C++0x auto syntax: auto iter = map_of_strings.begin(); // C++0x BOOST_AUTO( iter, map_of_strings.begin() ); // boost It even works in more complex contexts like: if( BOOST_AUTO(ptr, get_pointer()) ) { ptr->run(); } or: BOOST_AUTO( const& val, vect.front() ); It already provides a tool for emulating C++0x range-based for syntax: for( int & x : vect ) { ...} // C++0x BOOST_FOREACH( int & x, vect ) { ... }; // boost However currently it is impossible to emulate the following C++0x syntax: for( auto & x : collection ) { ... } // C++0x Therefore, I would like to request a new tool in Boost.Foreach that would emulate the above construct, and would look something like: BOOST_FOREACH_AUTO( const& x, collection ) {...} I do not anticipate any technical difficulties with implementing this, as it can already be achiever with a verbose: BOOST_FOREACH( BOOST_TYPEOF( boost::begin(collection) ) const& x, collection ) {..} Regards, &rzej

3 2

[datetime] feature request
by Sean Farrow 17 Nov '09

17 Nov '09

Hi; I'm currently working with a company who have locked there systems down to such an extent that users can only write to/read from certain areas. I am using the date/time library. What i need to be able to do is to load the date_time_zonespec.csv file from an rc file under windows. I can write the rc loading stuff, but how do I then go about submitting this, and more importantly only compiling this .cpp file under windows within the build framework. Chances are the code will be apart of the timeZoneDatabase class. Any help appreciated. Sean.

1 0

[1.41.0] Release candidate 2
by Beman Dawes 17 Nov '09

17 Nov '09

1.41.0 release candidate 2 packages are available. The only change from RC1 is the fix for the serialization snafu. http://boost.cowic.de/rc/boost_1_41_0.tar.bz2 - Unix line endings http://boost.cowic.de/rc/boost_1_41_0.tar.gz<http://boost.cowic.de/rc/boost_1_41_1_beta1.tar.gz>- Unix line endings http://boost.cowic.de/rc/boost_1_41_0.7z<http://boost.cowic.de/rc/boost_1_41_1_beta1.7z>- Windows line endings http://boost.cowic.de/rc/boost_1_41_0.zip<http://boost.cowic.de/rc/boost_1_41_1_beta1.zip>- Windows line endings Before pushing the release out to SourceForge, I'd appreciate it if several people would try the release candidate and report success, failure, or any problems encountered. Thanks, --Beman

3 7

[program_options] Feature request for untyped variables
by Dave Johnson 17 Nov '09

17 Nov '09

I usually store option values in a global structure so that I can access them by field name rather than by doing a map lookup with a string. This has the benefit that typos in option names are caught at compile time rather than run time. For example, I can use Options.input instead of vm["input"]. For untyped options, typically used as boolean flags (i.e. --help, --verbose), there is no way to specify where to store a value (or a count). This means I need to access them by map lookup or manually copy them to my global options structure. I'm not happy with either one. Ideally I'd like some way to specify that this option does not have a value, but I'd like to give the parser a pointer to where I want it to put a count of its instances in the command line or maybe a pointer to a bool instead. Example: // Global struct for command line options struct { bool help; bool verbose; string inputFile; ... } Options; // Parse command line options_description desc("Allowed options"); desc.add_options() ("help" , "produce help message") ("verbose" , "create more detailed output") ("input", po::value<string>(&Options.inputFile), "Name of the input file") ... ; variables_map vm; store(po::parse_command_line(argc, argv, desc), vm); notify(vm); // Hack! Need to manually assign bool options! // Options.input is automatically assigned Options.help = vm.count("help"); Options.verbose = vm.count("verbose");

2 2

[iterator] Interest check in specialized reproduction_iterator
by Christoph Heindl 17 Nov '09

17 Nov '09

Good afternoon, Would there be interest in adding a specialized iterator, called reproduction_iterator, that allows specifying a single element which is iterated n-times. Example: const std::vector<int> values = boost::assign::list_of(1)(2)(3)(5); const int c = 3; std::vector<int> result; std::transform( extboost::make_reproduction_iterator(c), extboost::make_reproduction_iterator(c, values.size()), values.begin(), std::back_inserter(result), std::multiplies<int>() ); // result contains 3, 6, 9, 15 The rationale for implementing the iterator is to provide a mechanism for the cheap iteration over a 'sequence' of equal objects. This is used (in my case) for adaption of functions that otherwise take two ranges and where I can bind one range to a single object. In the example above one can achieve the same effect by using a binder and std::transform that takes a single range. This is IMHO not always possible. 'reproduction_iterator' is implemented as random access iterator using boost::iterator_facade. Best regards, Christoph

1 0

[BGL] [variant] problem (and perhaps solution?) with strong_components.hpp and variant
by Nil Geisweiller 17 Nov '09

17 Nov '09

Hi, it seems the problem is indeed a conflict between the get function of variant.hpp and the get function of two_bit_color_map. After renaming get by getTEST in 1) two_bit_color_map.hpp line 61 2) breadth_first_search.hpp line 78 3) property_map.hpp line 164 the compiling error has disappeared. Of course it is certainly not the right way to fix that bug, maybe wrapping two_bit_color_map get (and alike) function name under some namespace... but I'm not confident enough to take such decision. Maybe there is a better way. Please if someone is more knowledgeable could help me fix that bug. Thanks for your help Nil On Tue, Oct 27, 2009 at 7:31 AM, Nil Geisweiller <ngeiswei(a)googlemail.com> wrote: > Hi, > > I resend that email with the right title header. > > ---------- Forwarded message ---------- > From: Nil Geisweiller <ngeiswei(a)googlemail.com> > Date: Sun, Oct 25, 2009 at 6:11 PM > Subject: [Graph] problem with strong_components.hpp and variant > To: boost(a)lists.boost.org > > > Hi, > > it seems there is a conflict between strong_components.hpp and > variant, maybe the method boost::get is wrongly overloaded. > > The problem occurs with 1.40 but not with 1.37, and it is still occurs > with the last svn. > > I attach a minimal code to reproduce the bug + compile error > > ###### > test.cpp > ###### > > //uncomment the line below to get the bug > #include <boost/graph/strong_components.hpp> > #include <boost/variant.hpp> > > typedef boost::variant < int, float > intorfloat; > > int main() { > int i(2); > intorfloat iof(i); > return boost::get<int>(iof); //problem is here > } > > ########### > compile error > ########### > > a-lin@a-lin-desktop:~/OpenCog/test/bin$ make > [100%] Building CXX object CMakeFiles/test.dir/test.cpp.o > In file included from /usr/include/boost/graph/graph_concepts.hpp:17, > from /usr/include/boost/graph/depth_first_search.hpp:18, > from /usr/include/boost/graph/strong_components.hpp:17, > from /home/a-lin/OpenCog/test/test.cpp:2: > /usr/include/boost/property_map/property_map.hpp: In instantiation of > ‘boost::property_traits<int>’: > /usr/include/boost/graph/two_bit_color_map.hpp:46: instantiated from > ‘boost::two_bit_color_map<int>’ > /home/a-lin/OpenCog/test/test.cpp:10: instantiated from here > /usr/include/boost/property_map/property_map.hpp:31: error: ‘int’ is > not a class, struct, or union type > /usr/include/boost/property_map/property_map.hpp:32: error: ‘int’ is > not a class, struct, or union type > /usr/include/boost/property_map/property_map.hpp:33: error: ‘int’ is > not a class, struct, or union type > /usr/include/boost/property_map/property_map.hpp:34: error: ‘int’ is > not a class, struct, or union type > make[2]: *** [CMakeFiles/test.dir/test.cpp.o] Error 1 > make[1]: *** [CMakeFiles/test.dir/all] Error 2 > make: *** [all] Error 2 > > Is it a bug? Thanks a lot! > Nil >

4 5

Fall bug sprint starts tomorrow! (Assuming that the release happens today)
by Marshall Clow 17 Nov '09

17 Nov '09

Here's the deal on the bug sprint: It will run until Sunday, November 29th. For those in the USA, that includes the long Thanksgiving weekend. The goal is to classify and fix bugs, and to close resolve Trac tickets. For those just joining in, the tickets live at https://svn.boost.org The instructions (mostly repeated here) for the last bug sprint are at https://svn.boost.org/trac/boost/wiki/BugSprintJune2009 General information about the ticket process is at What can you do? You can fix bugs • If you are a library maintainer, you can commit your fixes. • If you are not a library maintainer, you can fix a problem on your local machine (and test it!!), and then attach your fix as a patch to the Trac ticket for the library maintainer to commit. If you do this, please provide a test case that exercises the problem; i.e, one that fails w/o your patch, and succeeds afterwards. You can characterize bugs • You can attempt to replicate the bug on your system(s), and attach information that you discover to the Trac ticket. This will make it easier for other people to solve the problem, even if you cannot. • You can write test cases demonstrating the problem, and attach them to the Trac ticket. • You can determine if the bug still exists; sometimes they get fixed, and the ticket never gets closed. In this case, try to identify a checkin that addresses the bug; it's always suspicious when bugs go away "by accident." You can test patches • Some of the tickets in the Trac system contain patches - some are quite old. You can install these patches on your system, and determine if they (a) build, and (b) work. If you do this, please provide detail. Don't just say "works for me" - say "I installed this over revision XXX on system YYY with compiler ZZZ. The patch installed and compiled successfully, and passed all the tests." • You can write test cases for the patches, and attach them to the ticket. You can help implement new features • There are a whole set of "feature requests" in the Trac system. If you have an idea about one of these, you can add comments to the Trac system. You can attempt to implement these new features, and attach the patch to the ticket. (Don't forget the test cases!!). There are several Trac reports that I find helpful for looking at the bugs: Open ticket count by owner: https://svn.boost.org/trac/boost/report/18 (Note that there are 10 tickets that have no owner!) Open ticket count by component: https://svn.boost.org/trac/boost/report/19 Open ticket count by type https://svn.boost.org/trac/boost/report/20 (Overwhelmingly, these are 'bugs') Open ticket count by milestone: https://svn.boost.org/trac/boost/report/21 (Many old milestones here, including 62 tickets w/o a milestone) and, of course, the main report at https://svn.boost.org/trac/boost/report/1 which shows that we currently have 856 open tickets. As soon as the release is final, I will be posting the "release branch" vs. "trunk" comparison as well, so we can find things that need to be merged. -- Marshall

1 0

[GGL] performance
by Barend Gehrels 17 Nov '09

17 Nov '09

Hi list, As there were doubts expressed on this list regarding performance in more challenging environments, an additional test is provided. The resulting figures of GGL are still good, sometimes the best, sometimes the second best. For example, for a union on polygons with many diagonals, holes, and many points on both inputs: ggl 3.6 seconds b.p. 7.8 seconds terralib 8.2 seconds geos 35.9 seconds cgal 77.8 seconds (a cgal-intersection (instead of union) is the fastest in these tests) Polygons have large diagonals, holes are tested, and the number of vertices of both input polygons can be varied in the tests. More information: http://trac.osgeo.org/ggl/wiki/IntersectionStarComb, including references to sources for reproducing the figures, everyone is invited to do that. Regards, Barend

2 4

[serialization] warnings in debug build
by Christoph Duelli 17 Nov '09

17 Nov '09

building vs trunk I got (only in my debug build): boost/archive/basic_binary_oarchive.hpp:89: warning: comparison is always true due to limited range of data type boost/archive/basic_binary_oarchive.hpp:95: warning: comparison is always true due to limited range of data type boost/archive/basic_binary_oprimitive.hpp:163: warning: comparison between signed and unsigned integer expressions boost/archive/basic_binary_iprimitive.hpp:154: warning: comparison between signed and unsigned integer expressions Best regards Christoph

4 7

[Boost.Locale] For Preliminary Review of Boost Community (with Full documentation)
by Artyom 17 Nov '09

17 Nov '09

Hello All, I had written a Boost.Locale library that provides high quality localization facilities in C++ way. This library uses internally ICU library to provide all facilities: - Correct case conversion, case folding and normalization - Collation including support of 4 Unicode collation levels. - Date and time formatting and parsing - Number formatting, spelling and parsing - Monetary formatting and parsing - Powerful message formatting including support plural forms, using GNU catalogs. - Character, word, sentence and line-break boundary analysis. - Codepage conversion - Support of 8-bit character sets like Latin1 and UTF–8 encoded text. - Support of char, wchar_t and C++0x char16_t, char32_t strings and streams. I've tested it with: - Linux GCC 4.1, 4.3. with ICU 3.6 and 3.8 - Windows MSVC-9 (VC 2008), with ICU 4.2 - Windows MingW with ICU 4.2 - Windows Cygwin with ICU 3.8 Documentation: ============== Full tutorials: http://cppcms.sourceforge.net/boost_locale/docs/ Doxygen reference documentation: http://cppcms.sourceforge.net/boost_locale/docs/doxy/html/ Source Code: ============ https://cppcms.svn.sourceforge.net/svnroot/cppcms/boost_locale/trunk Building: ========= At this point I hand't written any JAM files yet and I use CMake. So checkout the code, and run cmake /path/to/boost_locale/libs/locale You need to have ICU 3.6 or above installed on your system. 4.2 and above recommended It is mandatory dependency. Notes for MSVC Users ==================== In many examples I write: wstring tmp=L"שלום עולם!"; // Unicode string In order to make these examples work correctly source code should include BOM to let MSVC know that the text is UTF-8 and not ASCII... So when you try to compile examples convert files in order to make them work correctly. Todo List ========= - Create a good set of unit tests - More boostification (Jam, etc) I'm waiting for your inputs Artyom Beilis

3 4

[Boost] [GGL] Bost.Polygon (GTL) vs GGL - rationale
by Jose 17 Nov '09

17 Nov '09

Hi, I am trying to understand the goemetry libraries. I find the currently under review GGL very useful for non-cartesian geometries (and maybe cartesian too!) but I don't understand what was the rationale for accepting GTL, which is a library mainly used in chip design. The reviews for GTL clearly stated that the library was not appropiate for GIS-like applications as its focus is mainly 2D. So, can anybody explain how these two libraries fit together. I see in which cases GGL will be useful but I am rather confused with Boost.Polygon, as it's optimized for VLSI design (as stated in the BoostCon paper and by some of the reviewers) Did someone decide that it was not possible to have a single GENERIC GEOMETRY library and we should have two ? thanks

5 7

[move][container] Review Request (new versions of Boost.Move and Boost.Container in sandbox and vault)
by Ion Gaztañaga 17 Nov '09

17 Nov '09

Hi to all, I've just uploaded to boost sandbox... http://svn.boost.org/svn/boost/sandbox/move/) and Boost.Vault... http://www.boostpro.com/vault/index.php?action=downloadfile&filename=boost.… ...new versions of libraries Boost.Move and Boost.Container. Online documentation can be found here: Boost.Move http://svn.boost.org/svn/boost/sandbox/move/libs/move/doc/html/index.html Boost.Container http://svn.boost.org/svn/boost/sandbox/move/libs/container/doc/html/index.h… I would like to request a review for both libraries. Boost.Move is quite small (just one header). Boost.Container depends on Boost.Move and Boost.Intrusive so you must download also the post-1.40 version of Intrusive placed in the same package (Boost.Container uses new features from Intrusive not included in Boost 1.40). Just take fresh Boost 1.39 folder and overwrite it with package contents. ************************ What's new in Boost.Move: ************************ ->The library has been modified to support moving rvalues of copyable AND movable types, a limitation of the previous version. This changes where proposed by Klaus Triendl in this thread: http://lists.boost.org/Archives/boost/2009/06/153266.php The library has now new macros to declare copyable and movable types and movable only types. Old macros are still there for backwards compatibility (specially unordered library which was also ported to this move sandbox) but I expect to remove them if current approach is considered better. ->Updated documentation and tests. ************************ What's new in Boost.Container: ************************ -> Added tests and documentation -> Updated the library to the new Boost.Move macros * * * I've tested both libraries in Windows XP using the following compilers: Visual 7.1, Visual 8.0, Visual 9.0, Intel 10.0, Mingw-gcc 4.3 and Mingw-gcc 4.3 in C++0x mode. The library should work on any C++03 conforming compiler Best, Ion

19 95

[random] Compilation errors in linear_congruential.hpp because of min()/max()
by Hartmut Kaiser 17 Nov '09

17 Nov '09

Hi, I'm getting compilations errors against Boost trunk from inside linear_congruential.hpp:+89 on VS2010 because of min()/max() being used without safeguarding them against accidental macro substitution: assert(_x >= min()); assert(_x <= max()); if I replace this with assert(_x >= (min)()); assert(_x <= (max)()); the compiler errors go away. Regards Hartmut ------------------- Meet me at BoostCon http://boostcon.com

2 2

[Function] Interest in boost::functions that take and return references to self-type?
by Tony Astolfi 16 Nov '09

16 Nov '09

Would there be interest in adding a feature to boost::function that allows a function to accept as input and return values of the same type as itself? I think this would be useful for implementing state machines. Something like: typedef boost::function< boost::function_self_type (event const&) > state_transition_fcn; state_transition_fcn initial_state (event const& e); state_transition_fcn second_state (event const& e); state_transition_fcn initial_state (event const& e) { if (/* some predicate on e*/) { // go to new state return &second_state; } else { // stay in old state return &initial_state; } } state_transition_fcn current_state = &initial_state; while (more_events) { current_state = current_state( next_event ); } -Tony Astolfi ps. Please point me in the right direction if such a thing already exists! Thanks!

2 1

[Review] We're halfway through the GGL review
by Hartmut Kaiser 16 Nov '09

16 Nov '09

Hey all, We're halfway through the GGL review (see attached review announcement) it is planned to end November 15th. Now would be a good time to send in your reviews! Regards Hartmut ------------------- Meet me at BoostCon http://boostcon.com

4 6

[Locale] Boost.Locale - preliminary code
by Artyom 16 Nov '09

16 Nov '09

Hello, I'm working on Boost.Locale project. Source code and some basic examples can be found here: https://cppcms.svn.sourceforge.net/svnroot/cppcms/boost_locale/trunk/ It still lacks of documentation and some features. It is mostly tested under Linux gcc-4.3 and it seems to work under MSVC-9 (however I had less tested it). General description: ==================== Boost.Locale is a library that brings high quality localization facilities in C++ way. It uses `std::locale`, and `std::locale` facets in order to pro- vide localization in transparent and C++ aware way to user. C++ has quite good base for localization via existing C++ locale facets: `std::num_put`, `std::ctype`, `std::collate` etc.. But they are very limited and sometimes buggy by design. The support of localization varies between different operating systems and incompatible between them. On the other hand, there is great, well debugged, high quality, widely used ICU library that gives all of the goodies but, it has very old API that mimics Java behavior, it is completely ignores STL and provides useful API only for UTF-16 encoded text, ignoring other popular Unicode encodings like UTF-8 and UTF-16, or limited national character sets like Latin1. Boost.Locale provides the natural glue between C++ locales framework, iostreams and powerful ICU library in following areas: - Correct case conversion, case folding and normalization (DONE) - Collation including support of 4 Unicode collation levels. (DONE) - Date and time formatting and parsing including support of timezones and calendars other then Gregorian. (DONE) - Number formatting, spelling and parsing (DONE) - Monetary formatting and parsing (DONE) - Powerful message formatting including support plural forms, using GNU catalogs. (DONE) - Character, word, sentence and line-break boundary analysis. (TODO) - Support of 8-bit character sets like Latin1 and UTF-8 encoded text. (DONE) - Support of `char`, `wchar_t` and C++0x `char16_t`, `char32_t` strings and streams. (DONE) ------------------------------------------------------------ FYI, It is still work in progress but 90% of the job is done. Please take a look on it --- especially experts in Windows programming. I would glad if someone explains me the process of submission and what should be done for it. I know that it is not yet Boost ready but it goes fast in this direction. Best, Artyom.

4 5

[1.41.0] Beta 1 available
by Beman Dawes 16 Nov '09

16 Nov '09

Boost release 1.40.1 beta 1 is available from SourceForge: http://sourceforge.net/projects/boost/files/ This release contains one new libraries and numerous bug fixes for existing libraries. For details, see http://beta.boost.org/users/news/version_1_41_0for details. To install the beta on your system, see http://beta.boost.org/doc/libs/1_40_0/more/getting_started/index.html Please report any problems with the beta on the developers or users mailing list. Unless unexpected problems arise, the plan is to ship the final 1.40.0 release in a week to ten days. Thanks, --The Boost release team

15 32

[1.41][serialization] MSVC 9 crashes on compilation
by Andrey Semashev 16 Nov '09

16 Nov '09

Hi, I'm building the release branch on MSVC 9 + STLPort 5.1.5. The compiler crashes when compiling Boost.Serialization (more precisely, the xml_wgrammar.cpp file). The command line I use is as follows: bjam.exe -j 2 --toolset=msvc variant=release link=shared threading=multi runtime-link=shared optimization=speed inlining=full stdlib=stlport --without-mpi --without-wave --without-python --without-graph stage Does anyone else have these problems?

8 16

Strict or lax build configuration
by Vladimir Prus 16 Nov '09

16 Nov '09

I have a question how building Boost should behave when the user provides wrong configuration information. E.g., suppose you have this in user-config.jam: using python : 3.20 : /nowhere ; or alternatively, in 1.42 a user runs Boost.Build as follows: ./bjam -sZLIB_INCLUDES=/still-nowhere ... What should happen? - A hard error. After all, the user wanted that specific configuration. And warnings are easy to miss. - A warning. Autodetection will still run and possibly find a random version - A hard error, but also run autodetection and let user know what was found. For example: Could not find Python version 3.20 in /nowhere. There is a valid installation of 2.5 in /usr. Remove the 'using python' statement to use it. Comments? Thanks, Volodya

6 9

[interprocess] compilation errors when BOOST_USE_WINDOWS_H is defined
by Hartmut Kaiser 16 Nov '09

16 Nov '09

If all I have is #define BOOST_USE_WINDOWS_H #include <boost/interprocess/smart_ptr/shared_ptr.hpp> I'm getting a whole bunch of compilation errors related to undefined types like win32_find_data_t, interprocess_filetime, etc. Anything I could do about this (Windows7, VS2005, VS2008, VS2010, Boost SVN trunk)? Regards Hartmut ------------------- Meet me at BoostCon http://boostcon.com

3 3

[exception][mpl] Boost.Exception uses unqualified size_t
by Hartmut Kaiser 16 Nov '09

16 Nov '09

The Boost.Exception library uses size_t without any qualification, which leads to unexpected ADL problems related to mpl::size_t. I'm currently not able to provide a minimal use case as I don't fully understand where this happens, but if I replace size_t with std::size_t in boost/exception/detail/object_hex_dump.hpp my problems go away. I tried to look around in other Boost libraries, but all other occurrences of size_t are qualified as std::size_t. Regards Hartmut ------------------- Meet me at BoostCon http://boostcon.com

7 14

[GGL] Some bits missing from Doxygen docs?
by Phil Endecott 15 Nov '09

15 Nov '09

Dear All, In the GGL docs at http://geometrylibrary.geodan.nl/formal%5Freview/ , if I go: Modules -> geometries -> ggl::linear_ring -> List of all members I see a page that says "This is the complete list of members for ggl::linear_ring< P, V, ClockWise, A >, including all inherited members.", and is then entirely empty. Is this what other people see? Maybe it is something odd with my browser. Thanks, Phil.

3 3

[1.41.0] Release candidates available
by Beman Dawes 15 Nov '09

15 Nov '09

1.41.0 release candidates are available: http://boost.cowic.de/rc/boost_1_41_1_beta1.tar.bz2 - Unix line endings http://boost.cowic.de/rc/boost_1_41_1_beta1.tar.gz - Unix line endings http://boost.cowic.de/rc/boost_1_41_1_beta1.7z - Windows line endings http://boost.cowic.de/rc/boost_1_41_1_beta1.zip - Windows line endings Before pushing the release out to SourceForge, I'd appreciate it if several people would try the release candidate and report success, failure, or any problems encountered. Thanks, --Beman

4 4

[asio] multicore performance
by Isidor Zeuner 15 Nov '09

15 Nov '09

After coming along this ([1]) interesting link, I wonder what the suggested way of building a fully multicore-capable boost::asio tcp server on the linux platform would be. An obvious solution would be to listen on one interface/port per core and solve the problem at the application layer, but I'm looking for a fully transparent solution. I haven't looked more deeply into the boost::asio inner workings. Is it feasible to have modify epoll_reactor to distribute the file descriptors among multiple epoll threads? Best regards, Isidor Zeuner [1] http://stackoverflow.com/questions/1234750/c-socket-server-unable-to-satura…

2 1