- Boost-users - lists.boost.org

[Q] regular expressions change the 'substitue with' value
by V S P 14 Jul '09

14 Jul '09

Looking for some help with how to use regular expression library I have a string "!1!" I am replacing it with "\temp\i\otrq92rq40100__db1003.xls" ( a windows path to a file) This is how I am doing it std::string repl=subst_designator+ boost::lexical_cast<std::string>(counter)+ subst_designator; boost::algorithm::replace_all_regex(new_arg, boost::regex(repl),*res_it); Where: subst_designator = '!' counter = 1 new_arg = "!1!" *res_it="\temp\i\otrq92rq40100__db1003.xls" after the above code, the new_arg contains a strange string that appears to have back slashes removed and the first \t replaced with something like a 'tab' new_arg=" empiotrq92rq40100__db1003.xls" I am thinking that this is wrong, because the regular expression should copy without modifications the value of *res_it into the new_arg I am using 1.38 on windows with VS9. My question is , how can make it so the string in *res_it substitutes !1! thank you in advance, Vlad -- Vlad P author of C++ ORM http://github.com/vladp/CppOrm/tree/master -- http://www.fastmail.fm - IMAP accessible web-mail

3 7

Linking to release boost libs form debug build
by johnnyk427 14 Jul '09

14 Jul '09

I've noticed my boost libs are slowing my game down incredibly in my debug builds. I'd like to link to release boost libs even in Debug, is that possible and will it cause any problems? Heres what I've got so far: I've managed to stop my project from auto-linking to the generated lib names (by adding #define BOOST_ALL_NO_LIB), and listed the libs I want to include manually, but now I get link errors like this: libcmt.lib(invarg.obj) : error LNK2005: __initp_misc_invarg already defined in Libcmtd.lib(invarg.obj) It wants to link to the release RunTime, but my project is linking to the debug RunTime. So it seems that I want to create my boost libs so that they are release (and optimized), but link to the debug runtime. Is it even possible to build the libs like that? I'm studying boost jam now but having a hard time understanding how I would make libs as such. Thanks for any help you can offer, -John Thanks! -John -- View this message in context: http://www.nabble.com/Linking-to-release-boost-libs-form-debug-build-tp2445… Sent from the Boost - Users mailing list archive at Nabble.com.

2 2

Date Time I/O formatting question
by Polder, Matthew J 13 Jul '09

13 Jul '09

I've found the Date Time library incredibly useful for dealing with timestamps, but I can't seem to get the output formatted the way I want. I've followed the example for the Date Time IO Tutorial (http://www.boost.org/doc/libs/1_39_0/doc/html/date_time/date_time_io.html#d…) but the output always comes out the same way. I've tried the following code with Visual Studio Express 2008 and g++ 4.2 on Solaris 10. I'm using Boost 1.39.0. Any help appreciated. matthew #include <fstream> #include <functional> #include <iostream> #include <sstream> #include "boost/date_time/gregorian/gregorian.hpp" #include "boost/date_time/posix_time/posix_time.hpp" #include "boost/date_time/local_time/local_time.hpp" using namespace std; int main(int argc, char** argv) { stringstream ss; ifstream in( argv[1] ); string inputFormat = argv[2]; string outputFormat = argv[3]; //I want my own formatting. boost::local_time::local_time_facet* output_facet = new boost::local_time::local_time_facet(); boost::local_time::local_time_input_facet* input_facet = new boost::local_time::local_time_input_facet(); ss.imbue(locale(locale::classic(), output_facet)); ss.imbue(locale(ss.getloc(), input_facet)); output_facet->format( outputFormat.c_str() ); input_facet->format( inputFormat.c_str() ); while ( !in.eof() ) { string s; getline( in, s ); ss<<s; boost::posix_time::ptime p; ss>>p; //This works fine. string s1 = boost::posix_time::to_simple_string(p); cout<<s1<<"\n"; ss.str(""); ss<<p; //I don't get my format. string s2 = ss.str(); cout<<s2<<"\n"; std::ws(in); ss.str(""); } } ----Input File---- 05:01:2007_23:08:02 06:01:2007_23:09:04 07:01:2007_16:32:58 07:01:2007_18:51:42 07:01:2007_19:51:27 ----Output---- 2007-Jan-05 23:08:02 2007-Jan-05 23:08:02 2007-Jan-06 23:09:04 2007-Jan-06 23:09:04 2007-Jan-07 16:32:58 2007-Jan-07 16:32:58 2007-Jan-07 18:51:42 2007-Jan-07 18:51:42 2007-Jan-07 19:51:27 2007-Jan-07 19:51:27

2 1

[statechart] Access to asynchronous_state_machine
by Bill Clark 13 Jul '09

13 Jul '09

We're using an asynchronous state machine. You can't construct these directly; you have to create them indirectly using a scheduler. Is there any "official" way to get access to the state machine after it's constructed? Or, alternatively, is there a way to pass a parameter to its constructor? The scenario I have in mind is one where the client needs to supply the state machine with some information at run time, ideally immediately upon construction, but in any case before any reactions occur. So far I have not found an easy way for the client to get access to the state machine itself. I can work around this by assuming that the state machine is a singleton, and setting a publicly visible static instance pointer. But this is far from ideal. Is there a better way? Thanks, Bill

1 1

POD boost aligned storage?
by Joseph Garvin 13 Jul '09

13 Jul '09

I was trying to figure out if boost::aligned_storage was POD so I looked at the header. It looks like the internal implementation type is a POD, but the actual aligned_storage itself is not. However, that implementation type is exposed through a public typedef. Is it safe for me to use the typedef if I really do need a POD type and don't mind losing the safety of being noncopyable? Is opting out of that tradeoff what the typedef is there for?

2 1

[asio]Winsock.h
by Robert Jones 13 Jul '09

13 Jul '09

What's the deal with Winsock.h & Asio? I can see there's been chatter on this topic before, but not what the result is. Thanks, Rob.

6 7

[Pool] Overhead in using singleton_pool?
by Julius Ziegler 13 Jul '09

13 Jul '09

Hello list, I would like to know if there is any overhead in using singleton_pool instead of a simple pool object. Especially, I guess there could be some overhead due to synchronisation, as the singleton interface is guaranteed to be thread safe. I currently have some trouble with deallocation order of statics at program exit (I am using a static pool object), but everything works fine when I use singleton_pool instead. Are there any disadvantages (runtime) when using singleton_pool? Thanks! Julius

1 0

Re: [Boost-users] [Program Options] Difficulty compiling for static runtime library
by Lindley M French 13 Jul '09

13 Jul '09

> Lindley M French wrote: > > I have a large project that I'd like to use Boost.Program > Options with. Most of the other Visual Studio 2005 projects are > set to link against "Multithreaded" or "Multithreaded Debug" > rather than the DLL variants of same; so of course, I need to > build Program Options accordingly. > > > > I tried this: > > bjam.exe" release --runtime-link=static program_options > > > > But that did not produce the library that VS wants, which is: > > > > libboost_program_options-vc80-mt-s-1_39.lib > > > > Try > bjam release runtime-link=static --with-program_options > > In Christ, > Steven Watanabe I tried that, it produced the following output: bjam.exe release --runtime-link=static --with-program_options WARNING: No python installation configured and autoconfiguration failed. See http://www.boost.org/libs/python/doc/building.html for configuration instructions or pass --without-python to suppress this message and silently skip all Boost.Python targets warning: No toolsets are configured. warning: Configuring default toolset "msvc". warning: If the default is wrong, your build may not work correctly. warning: Use the "toolset=xxxxx" option to override our guess. warning: For more configuration options, please consult warning: http://boost.org/boost-build2/doc/html/bbv2/advanced/configuration.html Building C++ Boost. After the build, the headers will be located at C:\Documents and Settings\lfrench\Desktop\MADCAT\atg-madcat-dev\im_nav\projects\madcat\third-party-software\boost_1_39_0 The libraries will be located at C:\Documents and Settings\lfrench\Desktop\MADCAT\atg-madcat-dev\im_nav\projects\madcat\third-party-software\boost_1_39_0\stage\lib Use 'bjam install --prefix=<path>' if you wish to install headers and libraries to a different location and remove the source tree. ...patience... ...found 523 targets... ...updating 15 targets... common.mkdir bin.v2\libs\program_options\build\msvc-8.0\release\link-static common.mkdir bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\cmdline.obj cmdline.cpp libs\program_options\src\cmdline.cpp(272) : warning C4800: 'int' : forcing value to bool 'true' or 'false' (performance warning) compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\config_file.obj config_file.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\options_description.obj options_description.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\parsers.obj parsers.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\variables_map.obj variables_map.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\value_semantic.obj value_semantic.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\positional_options.obj positional_options.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\utf8_codecvt_facet.obj utf8_codecvt_facet.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\convert.obj convert.cpp compile-c-c++ bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\winmain.obj winmain.cpp msvc.archive bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\libboost_program_options-vc80-mt-1_39.lib common.copy stage\lib\libboost_program_options-vc80-mt-1_39.lib bin.v2\libs\program_options\build\msvc-8.0\release\link-static\threading-multi\libboost_program_options-vc80-mt-1_39.lib 1 file(s) copied. ln-NT stage\lib\libboost_program_options-vc80-mt.lib "NT symlinks not supported yet, making copy" 1 file(s) copied. ...updated 15 targets... It doesn't appear to be using the correct filename.

2 1

boost autolinkig issue - library names on vc9?
by Maróy Ákos 13 Jul '09

13 Jul '09

Hi, I'm having an issue with boost autolinking on Visual Studio 9 (2008), with the boost-pything library. the boost build process generates libraries of the name: libboost_XXX-vc90-mt-gd-1_39.lib and the like, but interestingly autolinkig fails with the following message: 1>LINK : fatal error LNK1104: cannot open file 'boost_python-vc90-mt-gd-1_39.lib' am I doing something wrong? Akos

3 4

Linking to release libs in debug build
by John Krajewski 13 Jul '09

13 Jul '09

I've noticed my boost libs are slowing my game down incredibly in my debug builds. I'd like to link to release boost libs even in Debug, is that possible and will it cause any problems? Secondly, I'm not sure how to do this. Boost is somehow auto-linking to the correct lib for my build (I've downloaded the libs precompiled for visual studio). Does anyone know how this auto-linking process works and how I can override it to link to release libs instead of debug? Thanks! -John

2 1