December 2006 - Boost-users

threads / scoped_lock
by Goroll, Torsten 14 Dec '06

14 Dec '06

Hi, as I understand the scoped_lock, it locks every ressource in a code block. What if I dont want to lock a ressource all the time the method is running? Do I have to create a new method? Example: boost::mutex a_mutex; ... { boost::mutex::scoped_lock lock(a_mutex); cou<<"bla"; //much of code where cout dont has to be locked ... } Torsten -----Ursprüngliche Nachricht----- Von: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org]Im Auftrag von boost-users-request(a)lists.boost.org Gesendet: Donnerstag, 14. Dezember 2006 14:55 An: boost-users(a)lists.boost.org Betreff: Boost-users Digest, Vol 1117, Issue 1 Send Boost-users mailing list submissions to boost-users(a)lists.boost.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.boost.org/mailman/listinfo.cgi/boost-users or, via email, send a message with subject or body 'help' to boost-users-request(a)lists.boost.org You can reach the person managing the list at boost-users-owner(a)lists.boost.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Boost-users digest..."

5 5

Re: [Boost-users] [bjam]BorlandC++BuilderX problem staging Boost
by Jerry Schreiber 14 Dec '06

14 Dec '06

I have experimented further with bjam input syntax, trying to follow the examples in http://www.boost.org/tools/build/v1/build_system.htm. Since I am currently interested in the filesystem library only, I added "--with-filesystem". This reduced the amount of output so that the CMD window's capacity was not exceeded. I discovered that the first thing going wrong was: borland-C++-action bin\boost\libs\filesystem\build\boost_filesystem.dll\borland- 5_6_4\debug\exception.obj C:\Boost\boost_1_33_1\libs\filesystem\build\../src/exception.cpp: Fatal F1003 C:\Boost\boost_1_33_1\boost/filesystem/config.hpp 23: Error directive: Dynamic linking Boost.Filesystem does not work for Borland; use static linking instead *** 1 errors in Compile *** I have tried variations on -sBUILD=runtime-link-static and -sBUILD=<runtime-link>static, but to no avail; I keep getting the message that the borland-C++-action is requesting Dynamic linking, which does not work with Borland. What is the precise syntax to get only static linking from bjam? Jerry Schreiber

1 0

building 64-bit Windows using MVC 8.0
by Rusty Jones 14 Dec '06

14 Dec '06

Hi, I have found some information on building 64-bit versions of boost along with people's experience, but I don't seem to have a clear understanding of the state of 64-bit support. 1. How do you build boost 1.33.1 to build as a 64-bit version in MVC 8.0? Can you use bjam or do you have to do something else? 2. Will version 1.34 support 64-bit windows? The regression tests for RC_1_34_0 used to have columns for 64-bit windows but I don't see them there anymore ( http://engineering.meta-comm.com/boost-regression/CVS-RC_1_34_0/developer/s… ). Any information would be appreciated. I need to make our software 64-bit compatible and getting 64-bit compatible boost is essential to do this. Cheers, Rusty

1 0

mapped_file_source
by Victor Morales-Duarte 14 Dec '06

14 Dec '06

Hi all, I'm trying to use boost::iostream::mapped_file_source directly, but I can't close and then re-open the file without getting an exception. Here is an example. ********************************************************************************************** #include <boost/iostreams/device/mapped_file.hpp> int main( void ) { std::string fileName( "SomeFile.ext" ); int osAlignment = boost::iostreams::mapped_file_source::alignment(); boost::iostreams::mapped_file_source fileMap( fileName, 2 * osAlignment, 0 ); // Load data to application memory using fileMap.data(); fileMap.close(); // The following line throws because fileMap.is_open() returns true fileMap.open( fileName, 1024 * osAlignment, 1024 * osAlignment ); // Load more data to application memory using fileMap.data(); return 0; } ***************************************************************************************** What I truly don't understand is why a call to is_open() woud return true immediately after a call to close(). What do I need to do to be able to map different parts of the same file? I can't map the whole file because the number of files that I need to process is large enough that I'd run out of allocation space for my process. Thanks guys. Victor I'm using boost 1.33.1 with Visual Studio 2003 on WinXP SP2. --------------------------------- Cheap Talk? Check out Yahoo! Messenger's low PC-to-Phone call rates.

1 0

Re: [Boost-users] threads / scoped_lock
by Goroll, Torsten 14 Dec '06

14 Dec '06

But why so simple if I could do it intricately too? ;o) Thank you -----Ursprüngliche Nachricht----- Von: Ovanes Markarian [mailto:om_boost@keywallet.com] Gesendet: Donnerstag, 14. Dezember 2006 15:20 An: boost-users(a)lists.boost.org Betreff: Re: [Boost-users] threads / scoped_lock Hi! you could define a scope where to lock ;) Just like this: boost::mutex a_mutex; ... { { boost::mutex::scoped_lock lock(a_mutex); //lock cout<<"bla"; }//end of scope => unlock //much of code where cout dont has to be locked ... } With Kind Regards, Ovanes Markarian On Thu, December 14, 2006 15:12, Goroll, Torsten wrote: > Hi, > > as I understand the scoped_lock, it locks every ressource in a code block. What if I dont want to > lock a ressource all the time the method is running? Do I have to create a new method? > > Example: > > boost::mutex a_mutex; > ... > { > boost::mutex::scoped_lock lock(a_mutex); > cou<<"bla"; > > //much of code where cout dont has to be locked > ... > } > > Torsten > > -----Ursprüngliche Nachricht----- > Von: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org]Im Auftrag von > boost-users-request(a)lists.boost.org > Gesendet: Donnerstag, 14. Dezember 2006 14:55 > An: boost-users(a)lists.boost.org > Betreff: Boost-users Digest, Vol 1117, Issue 1 > > > Send Boost-users mailing list submissions to > boost-users(a)lists.boost.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.boost.org/mailman/listinfo.cgi/boost-users > or, via email, send a message with subject or body 'help' to > boost-users-request(a)lists.boost.org > > You can reach the person managing the list at > boost-users-owner(a)lists.boost.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Boost-users digest..." > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

1 0

BOOST_ASSERT and others
by abir basak 14 Dec '06

14 Dec '06

Hi, Hi, if any assert is there like BOOST_ASSERT or assert where I can pass a custom message to print on assert failure ? i.e. something like BOOST_ASSERT(condition, message) Also, is there any debug print version for boost? Something like, I want a message should be printed only when the application is running in DEBUG version (whatever flag the system provides for that). The message can be cout or printf kind of thing. A little safety is better sometimes like debug print can't change the value of the variable etc. Thanks -- Abir Basak, Member IEEE Software Engineer, Read Ink Technologies B. Tech, IIT Kharagpur email: abir(a)abirbasak.com homepage: www.abirbasak.com

2 1

A default constructor for Boost.MPI environment?
by Bo Peng 14 Dec '06

14 Dec '06

Dear boost users, I am using Boost.MPI for a python module so there is no argc, argv that can be provided to boost::parallel::mpi::environment(int & argc, char ** & argv) Because an environment object has to exist during the life time of the module, I have to define a global variable in an awkward way: int fake_argc = 0; char * fake_argv = ""; char ** ptr_argv = &fake_argv; boost::parallel::mpi::environment env(fake_argc, ptr_argv); Can a default constructor be provided? Bo

5 5

[bjam]BorlandC++BuilderX problems staging Boost
by Jerry Schreiber 14 Dec '06

14 Dec '06

I am entirely new to Boost. I used the following command: bjam -sTOOLS=Borland stage I got a large amount of output with lots of errors indicated, but I did get quite a bit created in C:\Boost\boost_1_33_1\bin\boost\libs. As I am currently interested in using the filesystem library, I attempted to build the simple_ls.cpp example file and got the following error on the link step: ilink32 -D -ap -Tpe -x -Gn -v -LC:\CBuilderX\lib;C:\Boost\boost_1_33_1 c0x32.objwindows\Debug_Build\simple_ls.obj,C:\Boost\boost_1_33_1\libs\filesystem\windows\Debug_Build\simple_ls.exe,, cw32.lib import32.lib,, Turbo Incremental Link 5.65 Copyright (c) 1997-2002 Borland Fatal: Unable to open file 'LIBBOOST_FILESYSTEM-BCB-SD-1_33_1.LIB' Is C++BuilderX supported by Boost, and if so, what is the proper command for bjam? Is there anything else I need to do in order to get the link to work? Thanks. Jerry Schreiber

2 1

Re: [Boost-users] HowTo Pass arguments to boost threads?
by Gabriel Ki 13 Dec '06

13 Dec '06

Hi, I am just trying the boost threads example from the link. #include <boost/thread/thread.hpp> #include <boost/thread/mutex.hpp> #include <boost/thread/tss.hpp> #include <iostream> boost::mutex io_mutex; boost::thread_specific_ptr<int> ptr; struct count { count(int id) : id(id) { } void operator()() { if (ptr.get() == 0) ptr.reset(new int(0)); for (int i = 0; i < 10; ++i) { (*ptr)++; boost::mutex::scoped_lock lock(io_mutex); std::cout << id << ": " << *ptr << std::endl; } } int id; }; int main(int argc, char* argv[]) { boost::thread thrd1(count(1)); boost::thread thrd2(count(2)); thrd1.join(); thrd2.join(); return 0; }// Output $ ./testme 10591400: 1 10591400: 2 10591400: 3 10591400: 4 10591400: 5 10591400: 6 10591400: 7 10591400: 8 10591400: 9 10591400: 10 Segmentation fault I have a few questions. 1) Why is id copied incorrectly in operator()? 2) Why does the count object get copied so many times? (i tried to write a copy constructor into count to see the id value) 3) I tried to add setstacksize in boost code and recompile. Is it a better way to do it? Thanks in advance. Gabriel ----- Original Message ---- From: Moshe Matitya <Moshe.Matitya(a)Kayote.com> To: boost-users(a)lists.boost.org Sent: Tuesday, December 12, 2006 3:40:03 AM Subject: Re: [Boost-users] HowTo Pass arguments to boost threads? Forever Kid wrote: > > What is the method boost prefers to pass arguments to boost > threads? (such as a pointer to some class > object) > > I saw the class thread_specific_ptr. > Is thread_specific_ptr designed for this purpose? > > Is there a recommended site with boost examples to check out? > Boost has good documentation, although I prefer to learn by > example and use strict documentation as I become more experienced. > > Thanks much, > Graham Take a look this introductory article on Boost.Threads by Bill Kempf: http://www.ddj.com/184401518 It addresses the question you're asking in a very clear, straightforward manner, with examples. Moshe _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users ____________________________________________________________________________________ Do you Yahoo!? Everyone is raving about the all-new Yahoo! Mail beta. http://new.mail.yahoo.com

2 1

program_options variable_value as cast explodes
by Edward Z. Yang 13 Dec '06

13 Dec '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm using program_options in a few programs that I'm doing, and I'd like to clarify what's going on at certain points. vm.count("key") appears to tell us the number of times a flag is specified, but the documentation here: <http://www.boost.org/doc/html/variables_map.html> doesn't seem to say anything about it. Where is it specified? Does abstract_variables_map actually inherit off of std map? Why don't we use !vm["key"].empty() instead? (slightly more verbose, but it's clearer about what it's doing). The vm["term"].as<int>() template seems to combust spectacularly (as in cause a stack dump) when it turns out that vm["term"] doesn't actually exist. This doesn't seem very robust, and according to the documentation, an exception ought to be thrown. Am I just misunderstanding the docs? What is the preferred method determining that no parameters were passed? I'm currently using argc == 1, which seems a bit hacky, and doesn't catch errant parameters. Thanks. This might seem a little pedantic, but I'm trying to do things right. :-) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFFf2sIqTO+fYacSNoRAqz5AJ9nBOEz87v8H4HjRB6mM2NSvNSllQCaA+SE s0kcqXiQXy7+ID7ugotbDKc= =NvIj -----END PGP SIGNATURE-----

2 1