April 2010 - Boost - lists.boost.org

[fiber] With MSVC71
by Chard 27 Apr '10

27 Apr '10

I was hoping to try the latest fiber library from the vault, but it currently does not compile successfully with MSVC71. Is it the intention to support this compiler? There are two compilation failures: fiber/strategy.hpp, lines 77-8 template< typename Callable > friend void this_fiber::at_fiber_exit( Callable); I believe this is a VC71 compiler bug; the forward declaration of a template function that is subsequently given friend access to a class does not seem to work. I don't know the workaround. The other can be trivially fixed, and should not cause any issues on other compilers. Here's the patch: --- X:/Old/boost/libs/fiber/src/fiber.cpp Fri Apr 09 19:33:08 2010 +++ X:/New/boost/libs/fiber/src/fiber.cpp Fri Apr 23 11:41:58 2010 @@ -90,7 +90,7 @@ return * this; } -fiber::operator unspecified_bool_type() const +fiber::operator fiber::unspecified_bool_type() const { return info_; } bool

1 0

Re: [boost] VC10 config - some regressions in several libraries
by Brian Wood 27 Apr '10

27 Apr '10

> I've made a note to myself to request that /Za be removed or at least deprecated in VC11 > (our command-line deprecation warnings aren't silenceable, so they're pretty effective). [ snip] Please don't top post and please trim quotes to make things less cumbersome. Regards, Brian Wood Ebenezer Enterprises http://webEbenezer.net

2 1

Color of nodes in Edmonds & Karps max flow algorithms.
by Dan Jiang 27 Apr '10

27 Apr '10

Hi All, I have a question: What is the color(s) of nodes that can still be labeled after max flow is found in the Edmonds and Karp's algorithm? Are they black AND grey or just black? I know black nodes can still be labeled, what about nodes with grey color? -Thanks, Dan _________________________________________________________________ Videos that have everyone talking! Now also in HD! http://go.microsoft.com/?linkid=9724465

2 1

Boost.thread compile error with SunCC 5.9
by Frank Meerkötter 27 Apr '10

27 Apr '10

Hi, I am trying to use boost.thread with the version 5.9 of the sun compiler. I encountered an error during compilation which already has been reported by Urs Stotz some weeks ago. He also supplied a patch for this issue. http://lists.boost.org/Archives/boost/2010/02/161801.php The patch works for me. I wonder if it could be applied for 1.43? Kind Regards, Frank

3 3

[GSoC] Student Proposals
by Andrew Sutton 27 Apr '10

27 Apr '10

Hi All, I'd like to start by thanking all of the students who are interested in working on the Boost C++ Libraries and submitted proposals. Unfortunately, we can't fund every proposal we received (about 85), and the competition for available funding slots (only 10) is very competitive. We carefully reviewed every proposal based on the merits and feasibility of the work described. If your proposal was not accepted, that does not mean that it was without merit. We encourage all students to continue participating with the Boost community as users and developers, regardless of the results of this program, and look forward to seeing new proposals next year. If you are interested in feedback about your proposal's review and how you can improve the competitiveness of your submissions, feel free to contact me directly. Having said that, congratulations to the students whose projects were accepted. I will be sending out instructions soon on getting started and up to speed with the Boost development infrastructure, Subversion, mailing lists, etc. I'm will be posting the instructions to the Boost developers mailing list (not users!). If you are a student whose project was accepted, you need to subscribe to the Boost developer's mailing list [1]. I'd also like to thank all of the mentors for helping with the ranking and assignment of student projects. Thanks again and best regards, Andrew Sutton andrew.n.sutton(a)gmail.com [1] http://lists.boost.org/mailman/listinfo.cgi/boost

1 0

[GSoC] Implementation of arbitrary precision integers (BigInts)
by Adam Smith 27 Apr '10

27 Apr '10

Hi, First of all, I'd like to extend thank you to whoever chose my proposal - it's a great honour, and I'm very excited to contribute to Boost. Just a few questions to start with; these are primarily for my mentor (Rene Rivera), I guess, but of course others could chime in. Will most communication happen on the mailing lists or on #boost or #boost-soc? If the latter, roughly what time of day should I be online? Alternatively, would direct e-mail communication be preferred? Are there any particular documents I should read? Any preliminary tasks that I should do? Thanks again for accepting my proposal! Adam

2 1

[1.43] regex build files
by Richard Hazlewood 27 Apr '10

27 Apr '10

Reminder; you might want to update the stand-alone makefiles in /branches/release/libs/regex/build Is it worth continuing with these? Seems like an unnecessary maintenance overhead, especially given that bjam seems to do it all. Chard.

2 1

fixing the compilation of the math lib on HPUX
by Frank Meerkötter 27 Apr '10

27 Apr '10

Hello, I am currently trying to build boost with the HP aCC (aCC: HP C/aC++ B3910B A.06.12 [Nov 03 2006]). The build fails for the tr1 math lib. [...] acc.compile.c++ bin.v2/libs/math/build/acc/release/threading-multi/copysign.o "libs/math/build/../src/tr1/copysign.cpp", line 16: error #2247: function "copysign(double, double) C" has already been defined extern "C" double BOOST_MATH_TR1_DECL copysign BOOST_PREVENT_MACRO_SUBSTITUTION(double x, double y) ^ 1 error detected in the compilation of "libs/math/build/../src/tr1/copysign.cpp". [...] acc.compile.c++ bin.v2/libs/math/build/acc/release/threading-multi/copysignf.o "libs/math/build/../src/tr1/copysignf.cpp", line 16: error #2247: function "copysignf" has already been defined extern "C" float BOOST_MATH_TR1_DECL copysignf BOOST_PREVENT_MACRO_SUBSTITUTION(float x, float y) 1 error detected in the compilation of "libs/math/build/../src/tr1/copysignf.cpp". [...] "libs/math/build/../src/tr1/fmaxf.cpp", line 16: error #2247: function "fmaxf" has already been defined extern "C" float BOOST_MATH_TR1_DECL fmaxf BOOST_PREVENT_MACRO_SUBSTITUTION(float x, float y) ^ 1 error detected in the compilation of "libs/math/build/../src/tr1/fmaxf.cpp". acc.compile.c++ bin.v2/libs/math/build/acc/release/threading-multi/fminf.o "libs/math/build/../src/tr1/fminf.cpp", line 16: error #2247: function "fminf" has already been defined extern "C" float BOOST_MATH_TR1_DECL fminf BOOST_PREVENT_MACRO_SUBSTITUTION(float x, float y) ^ 1 error detected in the compilation of "libs/math/build/../src/tr1/fminf.cpp". In all cases the code causing the problems looks somewhat like this: #if !(defined(__HP_aCC) && (__HP_aCC >= 61400)) extern "C" float BOOST_MATH_TR1_DECL fminf BOOST_PREVENT_MACRO_SUBSTITUTION(float x, float y) #endif [...] So it looks like someone already took some precautions against this problem. My compiler is slightly older (61200 vs. 61400). When i replace the 61400 at these four places with a 61000 my build completes just fine. ./libs/math/src/tr1/copysign.cpp:#if !(defined(__HP_aCC) && (__HP_aCC >= 61400)) ./libs/math/src/tr1/copysignf.cpp:#if !(defined(__HP_aCC) && (__HP_aCC >= 61400)) ./libs/math/src/tr1/fmaxf.cpp:#if !(defined(__HP_aCC) && (__HP_aCC >= 61400)) ./libs/math/src/tr1/fminf.cpp:#if !(defined(__HP_aCC) && (__HP_aCC >= 61400)) Can anybody more experienced with the aCC tell me if such a change would be a good idea? Would you accept a patch? Frank

2 3

Re: [boost] [1.43][Serialization] Serialization library seriosly broken
by Hartmut Kaiser 27 Apr '10

27 Apr '10

> Hartmut Kaiser wrote: > >> I've just made a change in void_cast.cpp. I tested it here and > >> checked > > into > >> the trunk(61546). > >> > >> Could you test this change ASAP? > > > > I will be able to test tomorrow evening only, but I definitely will > > test. > > > > Regards Hartmut > > You might want to check the change I made as well. The assert exists in 1.42 > - the only difference in the file - void_cast.cpp is a small change I made to fix > a memory leak. I'm not confident that my fix actually fixed it - but of course > everything passes here. I'm sorry to say I recompiled everything against Boost trunk (rev. 61561), but the problem still persists. I don't understand what's going on, but I doubt it's a problem directly in void_cast.cpp. For some reason the type I described gets registered more than once (definitely from different modules, although dynamically loaded, none gets unloaded before the assertion). Did you change anything related to auto-type registration? Regards Hartmut --------------- Meet me at BoostCon www.boostcon.com

4 8

[1.43] Beta build of PDF docs available.
by John Maddock 27 Apr '10

27 Apr '10

A PDF build of the docs for Boost-1.43-beta is now available from https://sourceforge.net/projects/boost/files/boost-docs/1.43.0/boost_pdf_1_… If library authors with quickbook/boostbook docs could take a look and check for SNAFU's that would be great. Enjoy, John.

2 2