- Boost - lists.preview.boost.org

Boost.Thread 1.52 vs 1.51 - g++ 4.7.2
by Jose 20 Nov '12

20 Nov '12

When including thread.hpp (1.52) I get the error below. Same setup but Boost 1.51 works ok. Does anybody know what is the issue ? thanks # g++ -c test.cpp In file included from /usr/include/boost/thread.hpp:24:0, from test.cpp:1: /usr/include/boost/thread/future.hpp:139:26: error: expected ‘;’ at end of member declaration /usr/include/boost/thread/future.hpp:139:32: error: ‘BOOST_THREAD_NOEXCEPT_OR_THROW’ does not name a type /usr/include/boost/thread/future.hpp:139:19: error: looser throw specifier for ‘virtual const char* boost::future_error::what() const’ In file included from /usr/include/boost/thread/exceptions.hpp:21:0, from /usr/include/boost/thread/pthread/thread_data.hpp:10, from /usr/include/boost/thread/thread.hpp:17, from /usr/include/boost/thread.hpp:13, from test.cpp:1: /usr/lib/gcc/x86_64-redhat-linux/4.7.2/../../../../include/c++/4.7.2/stdexcept:70:5: error: overriding ‘virtual const char* std::logic_error::what() const throw ()’

2 1

Graph Algorithm Advice
by J Birch 20 Nov '12

20 Nov '12

This isn't a Boost Graph problem per se, just a graph algorithm question. Let's say I have the following graph: I want to represent this graph in a certain way. In the small legend box above, there are 2 versions of a correct answer. I want to print out each vertex followed by it's immediate adjacent vertex. If it has more than one, then each path from that point should be wrapped in a '()'. Currently, I am using the Ruby RGL package and have tried both a directed and an undirected graph using a depth first search traversal, and have yet to hit on an algorithm that takes every edge case. Anyone out there have some pointers on how to do this? Thanks! Birch

2 3

Failing to build Boost.Test library using Clang 3.1 on Netbeans
by Paul A. Bristow 20 Nov '12

20 Nov '12

Target: i686-w64-mingw32 Thread model: posix) (This is Ruben's mingw64 build) on Netbeans 7.2 on Windows 7. The compilation stage goes OK (but with many warnings In file included from i:/boost-trunk/libs/test/src/test_tools.cpp:16: In file included from i:/boost-trunk\boost/test/impl/test_tools.ipp:23: i:/boost-trunk\boost/test/tree/test_unit.hpp:164:7: warning: 'dllexport' attribute only applies to variables and functions class BOOST_TEST_DECL master_test_suite_t : public test_suite { ^ i:/boost-trunk\boost/test/detail/config.hpp:106:29: note: expanded from macro 'BOOST_TEST_DECL' # define BOOST_TEST_DECL BOOST_SYMBOL_EXPORT ^ i:/boost-trunk\boost/config/platform/win32.hpp:30:42: note: expanded from macro 'BOOST_SYMBOL_EXPORT' # define BOOST_SYMBOL_EXPORT __declspec(dllexport) ^ <built-in>:150:38: note: expanded from macro '__declspec' #define __declspec(a) __attribute__((a)) but the link stage to build the library fails with many Cannot export boost::test_tools::tt_detail::check_impl(boost::test_tools::predicate_result const&, boost::unit_test::lazy_ostream const&, boost::unit_test::basic_cstring<char const>, unsigned int, boost::test_tools::tt_detail::tool_level, boost::test_tools::tt_detail::check_type, unsigned int, ...): symbol not found (full output window message attached). I have also failed to build using bjam (missing <cstddef> so missing the right stdlib?) and I wonder if both these are connected with lack of a suitable boost/config.hpp which caters for Clang *on this configuration*. Where would/should __clang__ be defined in order to select this compiler? My user-config.jam contains (in an attempt to cure ) # Clang compiler using mingw32 dw2 Ruben's release. using clang : 3.1 : "C:/mingw32-dw2/bin/clang++.exe" : <cxxflags>-v <cxxflags>-std="c++11" <cflags>-nostdinc # Do not include MSVC C headers. #<cflag>-c-include-dirs=??? <cxxflags>-nostdinc++ # Do not include MSVC C++ headers. # Use mingw-dw2 for includes <cxxflags>-cxx-include-arch="C:/mingw32-dw2/i686-w64-mingw32" <cxxflags>-cxx-include-32bit-dir=32 <cxxflags>-cxx-include-root="C:/mingw32-dw2/include/c++/4.6.3" <cxxflags>-stdlib=libc++ #<compileflags>-std=gnu++11 #<define>BOOST_STDLIB_CONFIG="i:/boost-trunk/boost/config/stdlib/libstdcpp3.hpp" <linkflags>-stdlib=libc++ ; Any views on my woes? Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow(a)hetp.u-net.com

2 7

Re: [boost] Graph Algorithm Advice
by J Birch 19 Nov '12

19 Nov '12

Sorry, I had pasted it in the client. it is attached now. Thanks! Birch On Nov 19, 2012, at 5:03 PM, Jeremiah Willcock <jewillco(a)osl.iu.edu> wrote: > (off list) > > I don't see your attachment on there. > > -- Jeremiah Willcock > > On Mon, 19 Nov 2012, J Birch wrote: > >> This isn't a Boost Graph problem per se, just a graph algorithm question. Let's say I have the following graph: >> >> >> >> I want to represent this graph in a certain way. In the small legend box above, there are 2 versions of a correct answer. I want to print out each vertex followed by it's immediate adjacent vertex. If it has more than one, then each path from that point should be wrapped in a '()'. Currently, I am using the Ruby RGL package and have tried both a directed and an undirected graph using a depth first search traversal, and have yet to hit on an algorithm that takes every edge case. >> >> Anyone out there have some pointers on how to do this? >> >> Thanks! >> Birch >> >> _______________________________________________ >> Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost >>

1 0

Re: [boost] Countertree
by Leo Goodstadt 19 Nov '12

19 Nov '12

On 17 November 2012 19:44, Francisco Jos? Tapia <fjtapia(a)gmail.com> wrote: > 2012/11/16 Vadim Stadnik <vadimstdk(a)gmail.com> > > On Sat, Nov 17, 2012 at 4:25 AM, Leo Goodstadt <leo.goodstadt(a)llew.org.uk wrote: > > > > > > Project location ( zip file with code and documentation) : > > > > https://dl.dropbox.com/u/8437476/works/countertree_code_doc.zip > > > > > > > > Quick view of documentation with code download : > > > > https://dl.dropbox.com/u/8437476/works/countertree/index.html > > > > > > > > This library is an implementation of a binary red-black counter tree. > > > > This tree have an additional counter in each leaf. This permit the access to > > > > the elements by the position, like in a vector. It is a random access > > > > container with random access iterators > > > > > > > > > Among other things, countertree would allow the boost Statistical > > > Accumulators Library to support rolling median and rolling quantiles > > > efficiently. This would be very useful. > > > > > > I had previously had to hack the tree class in the Policy-Based Data > > > Structures part of the gnu c++ library. > > > (see http://gcc.gnu.org/onlinedocs/libstdc++/ext/pb_ds/tree_order_statistics_nod… ) > > As far as I know, this support is not yet available in countertree project, > > although this is possible in principle. > > About the compatibility with the boost Statistical Accumulators Library, I > didn't thought about it, but it would be useful. I will try to implement, > but I need to examine in deep for to provide you a well founded opinion. > 1) For counter trees, the median is just the N/2-th value in the counter tree, where K is the size of the tree. The Nth-percentile can be calculated similarly. A rolling median with a fixed size window requires there to be an insertion and deletion for each (most!) new value, so I am guessing it should run in O(N log K) time, where N = size of data, K = size of window 2) The alternative is to use two multisets corresponding to values smaller and larger than the current median (or the Nth percentile). This should also have O(N log K) complexity. 3) With the special case where the data consists of integers in a fixed range, we can use a segment tree where the complexity would be O(N log R) where R = the size of the fixed range of data values. This can result in considerable speed ups. The best bit about counter trees are their flexibility, we can calculate multiple Nth-percentiles at the same time (e.g. all the quartiles). As to the actual rather than theoretical speed, I think we need bench-marking.... Leo

2 1

[boostcon12]Trouble with tuples very compiler dependent
by Larry Evans 19 Nov '12

19 Nov '12

This thread is a continuation of that started here: http://article.gmane.org/gmane.comp.lib.boost.devel/235427 It's been renamed because the new name is more topical. The tree_builder benchmarks mentioned here: http://article.gmane.org/gmane.comp.lib.boost.devel/235386 were refactored into: tuple_benchmark.tree_builder.cpp tuple_impl.bcon12_horizontal.hpp tuple_impl.bcon12_vertical.hpp and various other .hpp files. The benchmark was run as shown in the attached filt.compilation. The output after line: :filt_dr= shows with compiler=gcc4_8, the bcon12_horizontal implementation is slower than the bcon12_vertical. This agrees, at least qualitatively, with the page 12 bar chart from Eric's boostcon12 pdf file: trouble_with_tuples.pdf downloadable from here: https://github.com/boostcon/cppnow_presentations_2012/zipball/master and in the mon directory. The variadic bar in the pag12 bar chart corresponds to the: tuple_impl.bcon12_horizontal.hpp tuple implementation in the sandbox slim/test directory. The unrolled bar in the pag12 bar chart corresponds to the: tuple_impl.bcon12_vertical.hpp In contrast to the gcc4_8 compiler, with the clangxx compiler, the relative qualitative performance, is just the opposite. IOW, the bcon12_horizontal implementation is faster than the bcon12_vertical implementation. In fact, the rate of change of the performance difference accelerates as tree depth goes from 2 to 4. The rate of change is so stark that it suggests, at least to me, there may be some bug in clang. Of course that conclusion is based on almost no knowledge, on my part, of the clang implementation. The tuple_benchmark_filt.py can be modified to filter out other parts of the benchmark run output, which is here: http://svn.boost.org/svn/boost/sandbox/variadic_templates/sandbox/slim/test… In conclusion, whether it's best to use vertical or horizontal tuple implementation depends on the compiler. -regards, Larry

2 3

Build errors with msvc/release/pdb
by Christoph Macheiner 19 Nov '12

19 Nov '12

The following libraries generate errors when using msvc, building for release and generating the program debug database: date_time, locale, math (all), thread, wave. Except for math, this seems to be caused by having at least one directory specified in the /Fd option. Also, there are 251 .pdb files in the boost root directory instead of their respective output directory (presumably from serialization, unit_test, math, and most probably others). The command-line is: bjam link=shared runtime-link=shared variant=release --layout=system --stagedir=.\stage stage debug-symbols=on debug-store=database I am at a loss on how to fix this. Can someone help? On a related note, the .pdb files are automatically generated when building for debug. Is there an easy switch somewhere in the sources, so that I can turn this on globally also for release as a work-around? For reference, below is the output for the first file for each library, plus contents of the corresponding .rsp file Thanks, Christoph. ======================================== compile-c-c++ bin.v2\libs\date_time\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\gregorian\greg_month.obj greg_month.cpp libs\date_time\src\gregorian\greg_month.cpp : fatal error C1033: cannot open program database 'd:\dev\boost_1_52_0\gregorian\greg_month.pdb' call "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\date_time\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\gregorian\greg_month.obj.rsp" ---------------------------------------- "libs\date_time\src\gregorian\greg_month.cpp" -Fo"bin.v2\libs\date_time\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\gregorian\greg_month.obj" /Fd"gregorian\greg_month.pdb" -TP /O2 /Zi /Ob2 /W3 /GR /MD /Zc:forScope /Zc:wchar_t /wd4675 /EHs -c -DBOOST_ALL_DYN_LINK=1 -DBOOST_ALL_NO_LIB=1 -DDATE_TIME_INLINE -DNDEBUG "-I." ======================================== compile-c-c++ bin.v2\libs\locale\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\encoding\codepage.obj codepage.cpp libs\locale\src\encoding\codepage.cpp : fatal error C1033: cannot open program database 'd:\dev\boost_1_52_0\encoding\codepage.pdb' call "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\locale\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\encoding\codepage.obj.rsp" ---------------------------------------- "libs\locale\src\encoding\codepage.cpp" -Fo"bin.v2\libs\locale\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\encoding\codepage.obj" /Fd"encoding\codepage.pdb" -TP /O2 /Zi /Ob2 /W3 /GR /MD /Zc:forScope /Zc:wchar_t /wd4675 /EHs -c -DBOOST_ALL_NO_LIB=1 -DBOOST_LOCALE_DYN_LINK=1 -DBOOST_LOCALE_NO_POSIX_BACKEND=1 -DBOOST_SYSTEM_DYN_LINK=1 -DBOOST_THREAD_NO_LIB=1 -DNDEBUG "-I." ======================================== compile-c-c++ bin.v2\libs\math\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\assoc_laguerre.obj assoc_laguerre.cpp libs\math\build\..\src\tr1\assoc_laguerre.cpp(6) : error C2858: command-line option 'program database name (/Fdd:\dev\boost_1_52_0\assoc_laguerre.pdb)' inconsistent with precompiled header, which used '/Fdd:\dev\boost_1_52_0\vc110.pdb' libs\math\build\..\src\tr1\assoc_laguerre.cpp(6) : error C2859: d:\dev\boost_1_52_0\assoc_laguerre.pdb is not the pdb file that was used when this precompiled header was created, recreate the precompiled header. call "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\math\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\assoc_laguerre.obj.rsp" ---------------------------------------- "libs\math\build\..\src\tr1\assoc_laguerre.cpp" -Fo"bin.v2\libs\math\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\assoc_laguerre.obj" /Fd"assoc_laguerre.pdb" -Yu"pch.hpp" -Fp"bin.v2\libs\math\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\pch.pch" -TP /O2 /Zi /Ob2 /W3 /GR /MD /Zc:forScope /Zc:wchar_t /wd4675 /EHs -c -DBOOST_ALL_NO_LIB=1 -DBOOST_BUILD_PCH_ENABLED -DBOOST_MATH_TR1_DYN_LINK=1 -DNDEBUG "-I." "-Ilibs\math\src\tr1" ======================================== compile-c-c++ bin.v2\libs\thread\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\win32\thread.obj thread.cpp libs\thread\src\win32\thread.cpp : fatal error C1033: cannot open program database 'd:\dev\boost_1_52_0\win32\thread.pdb' call "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\thread\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\win32\thread.obj.rsp" ---------------------------------------- "libs\thread\src\win32\thread.cpp" -Fo"bin.v2\libs\thread\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\win32\thread.obj" /Fd"win32\thread.pdb" -TP /O2 /Zi /Ob2 /W4 /GR /MD /Zc:forScope /Zc:wchar_t /wd4675 /EHs -c -DBOOST_ALL_NO_LIB=1 -DBOOST_SYSTEM_DYN_LINK=1 -DBOOST_SYSTEM_NO_DEPRECATED -DBOOST_THREAD_BUILD_DLL=1 -DBOOST_THREAD_DONT_USE_CHRONO -DBOOST_THREAD_THROW_IF_PRECONDITION_NOT_SATISFIED -DNDEBUG "-I." ======================================== compile-c-c++ bin.v2\libs\wave\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\cpplexer\re2clex\aq.obj aq.cpp libs\wave\src\cpplexer\re2clex\aq.cpp : fatal error C1033: cannot open program database 'd:\dev\boost_1_52_0\cpplexer\re2clex\aq.pdb' call "C:\Program Files (x86)\Microsoft Visual Studio 11.0\VC\vcvarsall.bat" x86 >nul cl /Zm800 -nologo @"bin.v2\libs\wave\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\cpplexer\re2clex\aq.obj.rsp" ---------------------------------------- "libs\wave\src\cpplexer\re2clex\aq.cpp" -Fo"bin.v2\libs\wave\build\msvc-11.0\release\debug-store-database\debug-symbols-on\threading-multi\cpplexer\re2clex\aq.obj" /Fd"cpplexer\re2clex\aq.pdb" -TP /O2 /Zi /Ob2 /W3 /GR /MD /Zc:forScope /Zc:wchar_t /wd4675 /EHs -c -DBOOST_ALL_DYN_LINK=1 -DBOOST_ALL_NO_LIB=1 -DNDEBUG -D_CRT_SECURE_NO_DEPRECATE -D_SCL_SECURE_NO_DEPRECATE "-I." ========================================

2 1

User config commands needed to find Clang tools ar and ranlib
by Paul A. Bristow 19 Nov '12

19 Nov '12

I am trying to use the Clang compiler to build a Boost.Test library and have now found the necessary user-config.jam to select the compiler (thanks Jurgen Hunold) , but the link stage tools are not being found. Can anyone suggest what bjam incantation is needed ? Thanks Paul clang-linux.archive i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\th reading-multi\libboost_prg_exec_monitor-clang31-mt-sd-1_53.lib '"ar"' is not recognized as an internal or external command, operable program or batch file. '"ranlib"' is not recognized as an internal or external command, operable program or batch file. "ar" rc "i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\t hreading-multi\libboost_prg_exec_monitor-clang31-mt-sd-1_53.lib" "i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\t hreading-multi\execution_monitor.obj" "i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\t hreading-multi\debug.obj" "i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\t hreading-multi\cpp_main.obj" "ranlib" "i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\t hreading-multi\libboost_prg_exec_monitor-clang31-mt-sd-1_53.lib" ...failed clang-linux.archive i:\boost-trunk\boost\bin.v2\libs\test\build\clang-linux-3.1\debug\link-static\runtime-link-static\th reading-multi\libboost_prg_exec_monitor-clang31-mt-sd-1_53.lib... My user-config.jam contains using clang : #version 3.1 : # compiler "C:/mingw32-dw2/bin/clang++.exe" #"C:/mingw32-dw2/bin/ar.exe" <<<< actual locations of tools #"C:/mingw32-dw2/bin/ranlib.exe" <<<< actual locations of tools : # options <cxxflags>-v # verbose for testing. <cxxflags>-IC:/mingw32-dw2/include/c++/4.6.3 # for stdlib? #e:/dev/apps/mingw64/include/c++/4.7.2 <cxxflags>-IC:/mingw32-dw2/i686-w64-mingw32/include #e:/dev/apps/mingw64/include/c++/4.7.2/x86_64-w64-mingw32 # <cxxflags>-std="c++11" <root>"C:/mingw32-dw2/bin" # attempt to find the link tools ; --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow(a)hetp.u-net.com

1 0

[container] Missing image
by Klaim - Joël Lamotte 19 Nov '12

19 Nov '12

Hi, I thought I pointed it already last year but there is at least one missing important image in the boost container documentation: http://www.boost.org/doc/libs/1_52_0/doc/html/container/non_standard_contai… Joel Lamotte

1 0

[#pragma once]
by Sid Sacek 18 Nov '12

18 Nov '12

I've worked with the MSVC, Sun and GCC compilers. Each one of them supports #pragma once. Is there any reason to wrap this pragma within #ifdefs? Are there any compilers out there today that don't support this #pragma? #if defined( _MSC_VER ) && ( _MSC_VER >= 1020 ) # pragma once #endif Could we change this coding pattern so that it no longer looks like MSVC is being selfish? -Sid Sacek

17 20