January 2008 - Boost-users - lists.boost.org

MPL: Instantiating a template across permutations of types
by cp now 29 Jan '08

29 Jan '08

Hello. I am trying to instantiate a template taking two parameters across a series of types contained in mpl::vector's, and place the resulting types in an mpl::vector. template <class A, class B> class T {}; typedef mpl::vector<int, double, char> Atypes; typedef mpl::vector<double, std::string> Btypes; template <template <class, class> class TEMPLATE, class ATYPE, class BTYPES, class CLASSES> struct unary_fold_inner : mpl::fold< BTYPES, CLASSES, mpl::push_back<mpl::_1, TEMPLATE<ATYPE, mpl::_2> > > {}; template <template <class, class> class TEMPLATE, class ATYPES, class BTYPES> struct unary_fold : mpl::fold< ATYPES, mpl::vector<>, typename detail::unary_fold_inner<TEMPLATE, mpl::_2, BTYPES, mpl::_1>::type > {}; typedef unary_fold< T, Atypes, Btypes >::type CLASSES; I expect CLASSES will be an mpl::vector containing 2x3=6 types as follows: T<int, double> T<int, std::string> T<double, double> T<double, std::string> T<char, double> T<char, std::string> in other words, instantiating template T across all permutations of Atypes and Btypes. What I see instead is puzzling. Classes is indeed an mpl::vector, and indeed contains 6 types, but the types are simply: T<double, double> T<std::string, std::string> T<double, double> T<std::string, std::string> T<double, double> T<std::string, std::string> The iteration I'm trying to accomplish is all wrong. I think it has to do with how I'm creating my metafunctions. I'm still unclear about when the ::type suffix is needed (for the last parameter to mpl::fold<> for example) and whether I need an "apply" template somewhere in my struct's. Any help appreciated. v

2 2

boost::filesystem::rename - behavior/code vs. docs
by Andriy Gapon 29 Jan '08

29 Jan '08

[repost] This is what documentation on boost::filesystem::rename states: http://www.boost.org/libs/filesystem/doc/tr2_proposal.html template <class Path1, class Path2> void rename(const Path1& from_p, const Path2& to_p); Requires: Path1::external_string_type and Path2::external_string_type are the same type. Effects: Renames from_p to to_p, as if by POSIX rename(). Postconditions: !exists(from_p) && exists(to_p), and the contents and attributes of the file originally named from_p are otherwise unchanged. [Note: If from_p and to_p resolve to the same file, no action is taken. Otherwise, if to_p resolves to an existing file, it is removed. A symbolic link is itself renamed, rather than the file it resolves to being renamed. -- end note] Actual effect is that an exception is thrown when renaming a file to an existing file on the same fs. rename(2) performs an 'atomic' rename in this case without problems, but in the boost code there is a specila check against "too permissive" posix rename(): http://boost.cvs.sourceforge.net/boost/boost/libs/filesystem/src/operations… 1126 rename_api( const std::string & from, const std::string & to) 1127 { 1128 // POSIX is too permissive so must check 1129 fs::system_error_type dummy; 1130 if ( fs::exists( status_api( to, dummy ) ) ) 1131 return EEXIST; 1132 return std::rename( from.c_str(), to.c_str() ) != 0 1133 ? errno : 0; 1134 } And rename testing code also checks for the above behavior. I think that this is a huge difference between the docs and the code, and this is very counter-intuitive too. -- Andriy Gapon

1 0

[pre-processor ] help needed
by Ovanes Markarian 29 Jan '08

29 Jan '08

Hello *, I have some pre-processor sequence in form of: (X1) (X2) (X3) (X4) which is expanded by BOOST_PP_SEQ_FOR_EACH to template parameters: some_template<X1, X2, X3, X4> ... My problem now is that some of Xn parameters can be a template instantiation and that actually causes PP to think, that there are more parameters to the function applied by ..._SEQ_FOR_EACH. Example: some_other_template<T1, T2> cause preprocessor to think that there are 2 params some_other_template<T1 and T2>. Putting the parameter into additional braces like ((X4)) results in the template: some_template<X1, X2, X3, (X4)> and that results in a compiler error. I know this construct can switch off ADL, but I don't know why it result here in a compiler error, since I fully qualify the X4 (from which namespace it is). Is there any way to treat Xn as a single paramer but during the final pre-processing stage get rid of additional braces? Many thanks, Ovanes

2 4

boost regex, stlport and icu on linux
by Michele Bini 29 Jan '08

29 Jan '08

Hello, I'm building boost 1.34.1 on linux (with redhat 9 gcc 3.2.2-5) with STLport 5.0 and ICU. This is the bjam invocation: bjam -q -j4 --toolset=gcc \ --build-dir=~/boost/build \ --stagedir=~/boost/stage --prefix=~/build-gecko \ -sNO_BZIP2=1 \ -sSTLPORT_VERSION=5 -sSTLPORT_5_PATH=~/build-gecko/lib \ -sSTLPORT_INCLUDE_DIRECTORY=../include/stlport \ -sSTLPORT_LIB_DIRECTORY=. \ -sHAVE_ICU=1 -sICU_PATH=~/build-gecko \ --without-python \ runtime-link=shared variant=release \ instruction-set=i686 threading=multi stage install This simple example: #include <iterator> #include <boost/regex.hpp> #include <iostream> int main(int argc, const char** argv) { std::string xml_str = "<?xml version=\"1.0\"?><test/>\n"; boost::regex e1("<\\?xml version=\"[^\"]+\"\\?>", boost::regex::extended); std::string sanitized_xml = boost::regex_replace( xml_str, e1, "", boost::regex_constants::match_default | boost::regex_constants::format_literal); return 0; } compiles ok with c++ -c -o ./test.o -mcpu=pentiumpro -march=pentiumpro -pipe -O2 \ -g3 -pthread -fforce-addr -fstrict-aliasing -Wall -W -Wcast-align \ -Wconversion -Wpointer-arith -Wsign-compare -Wwrite-strings \ -ftemplate-depth-48 -fno-dollars-in-identifiers -fno-operator-names \ -ffor-scope -Wctor-dtor-privacy -Wnon-virtual-dtor \ -Woverloaded-virtual -Wsign-promo -Wsynth -D__linux__ \ -D__OSVERSION__=2 -D__LITTLE_ENDIAN__ -D_REENTRANT -D_GNU_SOURCE \ -DHAVE_CONFIG_H=1 -DDEBUG -DFATAL_ASSERTS -DXP_UNIX \ -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 \ -D_FILE_OFFSET_BITS=64 -D_LARGE_FILES -DNO_GCC_PRAGMA \ -I${HOME}/build-gecko/include/stlport \ -I${HOME}/build-gecko/include \ -I${HOME}/build-gecko/include/boost-1_34_1 \ ./test.cc # Yes, I know the command line is longer than the example but fails when linking with c++ -pthread -Wl,-rpath-link,${HOME}/build-gecko/lib \ -o ./test ./test.o \ ~/build-gecko/lib/libboost_regex-gcc32-mt-1_34_1.so \ ~/build-gecko/lib/libicuuc.so \ ~/build-gecko/lib/libicui18n.so \ ~/build-gecko/lib/libicudata.so \ ~/build-gecko/lib/libstlport.so The full linker error is: test.o(.gnu.linkonce.t._ZN5boost12regex_searchIPKcN8stlp_std9allocatorINS_9sub_matchIS2_EEEEcNS_12regex_traitsIcNS_16cpp_regex_traitsIcEEEEEEbT_SC_RNS_13match_resultsISC_T0_EERKNS_11basic_regexIT1_T2_EENS_15regex_constants12_match_flagsESC_+0x9d): In function `bool boost::regex_search<char const*, stlp_std::allocator<boost::sub_match<char const*> >, char, boost::regex_traits<char, boost::cpp_regex_traits<char> > >(char const*, char const*, boost::match_results<char const*, stlp_std::allocator<boost::sub_match<char const*> > >&, boost::basic_regex<char, boost::regex_traits<char, boost::cpp_regex_traits<char> > > const&, boost::regex_constants::_match_flags, char const*)': /home/nas/michele/build-gecko/include/stlport/stl/_alloc.c:273: undefined reference to `boost::re_detail::perl_matcher<char const*, stlp_std::allocator<boost::sub_match<char const*> >, boost::regex_traits<char, boost::cpp_regex_traits<char> > >::construct_init(boost::basic_regex<char, boost::regex_traits<char, boost::cpp_regex_traits<char> > > const&, boost::regex_constants::_match_flags)' test.o(.gnu.linkonce.t._ZN5boost9re_detail12perl_matcherIPKcN8stlp_std9allocatorINS_9sub_matchIS3_EEEENS_12regex_traitsIcNS_16cpp_regex_traitsIcEEEEE11match_matchEv+0xd4): In function `boost::re_detail::perl_matcher<char const*, stlp_std::allocator<boost::sub_match<char const*> >, boost::regex_traits<char, boost::cpp_regex_traits<char> > >::match_match()': /home/nas/michele/build-gecko/include/boost-1_34_1/boost/regex/v4/regex_iterator.hpp:48: undefined reference to `boost::match_results<char const*, stlp_std::allocator<boost::sub_match<char const*> > >::maybe_assign(boost::match_results<char const*, stlp_std::allocator<boost::sub_match<char const*> > > const&)' test.o(.gnu.linkonce.t._ZN5boost9re_detail11raise_errorINS_20regex_traits_wrapperINS_12regex_traitsIcNS_16cpp_regex_traitsIcEEEEEEEEvRKT_NS_15regex_constants10error_typeE+0x9a): In function `void boost::re_detail::raise_error<boost::regex_traits_wrapper<boost::regex_traits<char, boost::cpp_regex_traits<char> > > >(boost::regex_traits_wrapper<boost::regex_traits<char, boost::cpp_regex_traits<char> > > const&, boost::regex_constants::error_type)': /home/nas/michele/build-gecko/include/boost-1_34_1/boost/regex/v4/perl_matcher_common.hpp:861: undefined reference to `boost::re_detail::raise_runtime_error(stlp_std::runtime_error const&)' collect2: ld returned 1 exit status I googled a bit and found this: http://lists.boost.org/boost-users/2006/02/17394.php but there was no follow up. Is there some mistake in my example? Does somebody have any suggestions or workarounds? I know that gcc 3.2.2 is quite old but I have used it for many years and always convinced it to do its work (and no, I cannot easily switch to a newer compiler). Any help will be really appreciated; cheers, Michele

2 1

bjam help modifying a jam file
by Derrick Hathaway 28 Jan '08

28 Jan '08

Hello,I recently inquired about building Boost System which is included with the Boost Asio distribution as a universal binary. I think my question was too comlicated, so let me simplify it. What I _think_ I need to do is add my build environment information into the Boost.System Jamfile. Something like: ### libs/system/build/Jamefile # Boost System Library Build Jamfile # (C) Copyright Beman Dawes 2002, 2006 # Distributed under the Boost Software License, Version 1.0. # (See accompanying file LICENSE_1_0.txt or www.boost.org/LICENSE_1_0.txt) # See library home page at http://www.boost.org/libs/system ##### this section added # Compiler configuration using darwin : 4.0 : g++-4.0 : <cflags>"-isysroot /Developer/SDKs/MacOSX10.5.sdk -arch ppc -arch i386" <linkflags>"-Wl,-syslibroot,/Developer/SDKs/MacOSX10.5.sdk -arch ppc -arch i386" ; project boost/system : source-location ../src : usage-requirements # pass these requirement to dependents (i.e. users) <link>shared:<define>BOOST_SYSTEM_DYN_LINK=1 <link>static:<define>BOOST_SYSTEM_STATIC_LINK=1 ; SOURCES = error_code ; lib boost_system : $(SOURCES).cpp : <link>shared:<define>BOOST_SYSTEM_DYN_LINK=1 <link>static:<define>BOOST_SYSTEM_STATIC_LINK=1 ; ## end Jamfile Any Ideas? Is this the wrong place to get help on my non-universal linking problem? Thanks. -Derrick

2 1

A question about Boost.RegEx
by Chengyuan Ma 28 Jan '08

28 Jan '08

I have a text file in the following format: (sparse data format, index:value) 0.0 1:0.269474 2:0.145364 3:0.067149 4:0.112643 5:0.212212 6:0.244601 7:0.181663 8:0.238227 9:0.848362 10:0.266284 11:0.058374 12:0.071349 13:0.192308 14:0.20059 15:0.256923 16:0.385338 17:0.123268 18:0.119405 19:0.350768 20:0.187007 21:0.369464 22:0.056476 23:0.059463 24:0.07298 25:0.158566 26:0.192542 27:0.315876 28:0.503185 29:0.216059 30:0.122681 31:0.228612 32:0.116034 33:0.12488 34:0.171623 35:0.222429 36:0.278741 37:0.170732 38:0.404539 39:0.078273 40:0.201989 41:0.367349 42:0.310658 43:0.176915 44:0.215489 45:0.207045 46:0.267294 47:0.158534 48:0.114389 49:0.085446 50:0.141968 51:0.11669 52:0.804789 53:0.533344 54:0.112373 55:0.173574 56:0.495218 57:0.122419 58:0.091748 59:0.209178 60:0.100954 61:0.168572 62:0.130615 63:0.080905 64:0.552943 65:0.208904 66:0.072037 67:0.166432 68:0.539735 69:0.186302 70:0.161657 71:0.135055 72:0.131747 73:0.434487 74:0.235148 75:0.119409 76:0.137161 77:0.186354 78:0.182466 79:0.105231 80:0.049308 81:0.199764 82:0.275725 83:0.369274 84:0.222261 85:0.1464 86:0.396967 87:0.937 88:0.983 90:0.983 91:1.0 92:1.0 This one is just one line, i have 64000 lines like this one. What's the best way to load the data? I use Boost::RegEx and Boost::lexical_cast to do this. But It takes 2 minutes to read all the data. Is there a better way to do this? bool LibFile::ReadFile(const string &fileName) { ifstream fin(fileName.c_str(), ios::in) ; boost::regex elabel("^([0-9]+\\.?[0-9]+)", boost::regbase::icase); boost::regex eitem("(\\d+):([-+]?[0-9]*\\.?[0-9]+)", boost::regbase::icase); while (fin.good()) { string buffer ; getline(fin, buffer) ; if ( buffer.length() > 0) { Instance *pinstance = new Instance() ; pinstance->tag = "notag" ; pinstance->vector = new double[featureDim] ; for ( int ii = 0 ; ii < featureDim; ii ++ ) { pinstance->vector[ii] = 0.0 ; } boost::smatch what; string::const_iterator itb = buffer.begin() ; string::const_iterator ite = buffer.end() ; double label = 0.0 ; if ( boost::regex_search( itb, ite, what, elabel) ) { label = boost::lexical_cast<double>(what[1].str()) ; itb = what[0].second ; } while ( boost::regex_search( itb, ite, what, eitem) ) { int index = boost::lexical_cast<int>(what[1].str()) ; double val = boost::lexical_cast<double>(what[2].str()) ; if ( index <= featureDim) { pinstance->vector[index - 1] = val ; } itb = what[0].second ; } vecInstance.push_back(pinstance) ; instanceNum ++ ; } } fin.close() ; return true ; }

2 1

[interprocess] nptl compatibility
by Szilard Pataki 28 Jan '08

28 Jan '08

Hello Ion, I would like to use the interprocess library and some NPTL features (pthread_setaffinity_np => assign a thread to a cpu) at the same time. The pthread_setaffinity_np function is declared in /usr/include/nptl/pthread.h (on my developer platform), but throughout the interprocess/boost headers I can find #include <pthread.h> ... which makes me think that /usr/include/pthread.h is included. Can I simply swap in the interprocess .hpp files the include statements to this: #include <nptl/pthread.h> ... or do you know of a cleaner solution? I'm asking you this, because I'm using a trimmed down boost library (interprocess + bare minimum from the rest of boost) and only in the interprocess headers do I see pthread.h inclusion. Furthermore, I will have to link against /lib64/tls/libpthread-2.3.4.so... do you foresee a problem with that? Many thanks, Szilard

2 1

Boost.Test
by Hernán Leoni 28 Jan '08

28 Jan '08

Hi, I'm beginning to use Boost.Test, I decided to create a test suite (with BOOST_AUTO_TEST_SUITE) for each module to test, and various test cases (with BOOST_AUTO_TEST_CASE) in each suite. I was wondering is it exist any macro to define functions to be executed at: - test suit initialization: execute a function before any test in the test suite - test suit finalization: execute a function after any test in the test suite - test setup: execute a function before each test - test tear down: execute a function after each test Is there an easy way to do what I need? Many thanks, Hernan

2 1

[boost-users] [ptr_container and output iterator]
by chun ping wang 28 Jan '08

28 Jan '08

Hi, is their going to be an overload for ptr_container to be able to work well with back_insert_iterator, front_insert_iterator and insert_iterator. Like i can copy from ptrVectorA to ptrVectorB without worrying about the range of ptrVectorB.

2 10

linking errors with VC8
by jhfjgfhjgf＠Safe-mail.net 28 Jan '08

28 Jan '08

Hello, I am using the current boost (1.35) from the SVN, built with Visual Studio 2008 in 32 bit mode. The debug build of my project works fine, but unfortunately there is a linker error when building my application with the release configuration. 1>LINK : fatal error LNK1104: cannot open file 'libboost_iostreams-vc90-mt-s-1_35.lib' The boost library path is correct - and there is definitely no 'libboost_iostreams-vc90-mt-s-1_35.lib'. I build use using bjam: bjam --toolset=msvc-9.0 -j4 -sZLIB_INCLUDE=..\..\..\..\zlib-1.2.3 -sZLIB_SOURCE=..\..\..\..\zlib-1.2.3 link=static,shared debug release install bjam --toolset=msvc-9.0 -j4 -sZLIB_INCLUDE=..\..\..\..\zlib-1.2.3 -sZLIB_SOURCE=..\..\..\..\zlib-1.2.3 link=static,shared debug release clean What's wrong? Thank you -Daniel

2 1