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

[Boost String Algorithms library] ireplace_all bug with character arrays
by Richard Turrin 18 Jan '08

18 Jan '08

Hi, I've encountered a bug with ireplace_all from the Boost String Algorithms library for version 1.35. If I pass a character array for the Format argument, the function ignores the null terminator and instead uses the entire array minus the last character to replace the occurrences of the Search string in the Input parameter. However, if I provide a pointer to a character array, the string replacement is done as expected. When I stepped into the code, I came across the following lines (excerpt from /trunk/boost/range/as_literal.hpp): template< class Char, std::size_t sz > inline iterator_range<const Char*> as_literal( const Char (&arr)[sz] ) { return boost::make_iterator_range( arr, arr + sz - 1 ); } There is no check for a null terminator. Nor did I find one after this function returned. I think the author here assumed that a caller would always pass in a character array of the right size. I've attached a sample that demonstrates the problem. I performed the test in Visual Studio 2008. Let me know if you need more information. Thanks. Rich Turrin

4 15

the Switch library by Steven Watanabe has been accepted provisionally
by Stjepan Rajko 18 Jan '08

18 Jan '08

The Switch library, submitted by Steven Watanabe, has been accepted provisionally pending a mini-review to be scheduled when possible. The purpose of the mini-review will be to assure that the suggestions outlined below have been adequately addressed. I would like to again thank Steven for submitting a very useful library, and for diligently working with the reviewers on improving its design. I would also like to thank Tobias Schwinger, Joel de Guzman, Darren Cook, Daniel (dherring@...), Felipe Magno de Almeida, and Paul Bristow for providing reviews, votes and half votes. Finally, thanks also to Dan Marsden, Larry Evans, Alexander Nasonov, Richard (legalize@...) and Mathias Gaunard for contributing to the discussion. SUMMARY OF VOTES: Tobias Schwinger and Joel de Guzman have submitted full reviews with positive votes: http://article.gmane.org/gmane.comp.lib.boost.user/32637 http://article.gmane.org/gmane.comp.lib.boost.user/32634 (Joel eventually converted his vote to "yes") One "yes" (half) vote was submitted by Paul Bristow: http://article.gmane.org/gmane.comp.lib.boost.devel/170069 One "no" (half) vote was submitted by Felipe Magno de Almeida: http://article.gmane.org/gmane.comp.lib.boost.devel/170051 Another "no" came privately - the point there was that the library didn't undergo enough scrutiny before submission, and that it should be resubmitted. Another privately submitted "no" came soon after the review opened, stating that the reviewer was not able to deduce any usability/usefulness from the library documentation, and suggested there should be a motivating example on the first page. RATIONALE FOR ACCEPTANCE: I believe the low number of full, positive reviews is justified by the relatively narrow scope that this library covers. Within that scope, however, reviewers indicated that the library is extremely useful. Also, the reviewers submitting positive votes have indicated (and proven) significant expertise in the area, leading me to place much weight on their votes. On the other hand, the negative votes did not point out any concrete flaws in the library itself that would merit rejection. RATIONALE FOR REQUIRING A MINI-REVIEW The "no" votes did, on the other hand, indicate some skepticism of the quality of the library/documentation as submitted, and the progress of the redesign that was happening during the review. Also, as a result of the review, the potential scope of the library has expanded significantly from what was originally proposed (which mostly focused on the ability to choose the number of cases at run-time). I believe that a combination of these two give sufficient reason for a mini-review - doubts about the library can be resolved, and those that might become prospective users of the library under the expanded scope will get a chance to give their opinion. I volunteered to manage the mini-review, so once the library is ready it will hopefully not remain in the review queue for too long (as it seems that finding review managers is the biggest problem ATM). To keep the burden on the community low, I propose to treat the mini-review as a partial continuation of the current review - I will consider the "yes" votes by Tobias and Joel to be left standing unless they state otherwise, and will require any "no" votes to be (re)justified given the new condition of the library. SUGGESTED CHANGES TO THE LIBRARY: * Documentation All reviewers have noted that the documentation could be improved. Specific suggestions were: * motivating example on the first page * additional examples * performance stats * more explanation of the types involved and how they are used. As the scope for the library has increased as a consequence of the review, a gentler, more detailed documentation approach might benefit the library - I believe that many people might find the new library useful, given the right motivation and examples in the documentation. * Return type Tobias Schwinger suggested that the return type should be specified explicitly in the switch_ call, rather than inferring it from the function object, and this seems to be the status quo in the newer designs. Joel de Guzman also suggested that if a version of switch_ should infer the return type, that return type should be a Boost.Variant of the types returned by each of the cases. * Default default behavior If the default behavior is left unspecified, switch_ should invoke the default constructor of the return type in the default case, rather than throwing an exception. This was suggested by Tobias Schwinger after the issue was raised by Dan Marsden. Alexander Nasonov has also indicated that a throwing default case can cause inefficiency, even when it is guaranteed not to be invoked. * Fall-through Alexander Nasonov pointed out the lack of fall-through capability in the submitted implementation. Fall-through has been incorporated into the recent designs being discussed. * Allowing void-returning function objects for the default case with a non-void returning switch Tobias Schwinger mentioned the possibility of allowing default case function objects of void return type even when the switch return type is non-void, as long as the function object call does not return. This would be useful, for example, if the function object is guaranteed to throw. After some discussion on whether determining if the return type is void can be done efficiently, it looks like the proposed solution is to use a special return type that indicates that the call will not return. Since Tobias indicated that this feature is a minor point, it could be implemented or not, depending on how well it fits into the final design. * Design There has been a lengthy discussion on the underlying design of the library, what it should offer, and whether the name of the library is appropriate. There are two types of designs that have been put forward as being appropriate and/or necessary for a Boost.Switch library. On one hand, there is the submitted design (labeled "A" in recent postings), which takes the switch cases in an MPL sequence and a single function object containing implementations of all cases. Joel de Guzman has proposed another design (labeled "B"), in which the switch cases are specified separately and allow a different function to be used for each case. Design "A" is better suited for situations in which the number of cases is determined by a compile time constant, while "B" is more appropriate for many use cases, and provides more flexibility and a more familiar syntax. The result of the discussion was that the Boost.Switch library should cover both situations, and an interface is now being developed that can be used in different ways for different uses (based on a common concept capturing case sequences). Thanks again to all that have contributed to this review, and congratulations to Steven for the accepted library. As others have stated, there is much confidence that Steven will meet the challenges successfully for the mini-review, which we will schedule as soon as the library is ready. Regards, Stjepan

2 2

Error using native_file_string()
by Eswar K 17 Jan '08

17 Jan '08

Hi, I am newbie to boost. I was earlier using boost-1.33.1 and my code was working fine. In FC6 i am compiling my code and shows me the following error. * error: âclass boost::filesystem::basic_directory_entry<boost::filesystem::basic_path<std::basic_string<char, std::char_traits<char>, std::allocator<char> >, boost::filesystem::path_traits> >â has no member named ânative_file_stringâ* Is there any difference in the libraries ? Am I missing something? My function is something like this. *void recurse_directory( const boost::filesystem::path& directory ) { boost::filesystem::directory_iterator end; for( boost::filesystem::directory_iterator dir_itr( directory ); dir_itr != end; ++dir_itr ) { if( boost::filesystem::is_directory( *dir_itr ) ) { try { recurse_directory( *dir_itr ); } catch (boost::filesystem::filesystem_error &e) { std::cerr << "Error recursing directory: " << (*dir_itr).native_file_string() << " (" << e.what() << ")" << std::endl; continue; } } else { add_file( (*dir_itr).native_file_string() ); // Get error here.... } } } * I am getting the error on the highlighted line. I am stuck here. Any help would be deeply appreciated. Thanks in advance. Regards, Eswar

2 1

Examples of boost::thread?
by Robert Dailey 17 Jan '08

17 Jan '08

Hi, I want to make a simple worker thread in my WTL application. However, looking through the boost::thread documentation, I don't see any samples of how to create a thread. Would anyone be so kind as to point me in the right direction? Thank you.

2 1

Boost.regex :: u32regex does not compile with latest STLPort
by Amit 17 Jan '08

17 Jan '08

Hi, Not sure if this is an issue on boost side or STLPort side, but compiling u32regex match givess an error in v4/perl_matcher.hpp, line 85. the call is essentially vector<int>::assign(int, int) since char type is int for ICU strings. This is being resolved to template<iter> assign(iter, iter) member, hence the error. I have fixed it in my local copy by changing the line to s1.assign(static_cast<std::size_t>(1), col); But don't know if this is a good fix. This used to work on Dinkum STL, so I am guessing STLPort is messing up somewhere. Do I need to build regex libraries again after this change? Thanks Amit

2 2

MinGW Build Issues
by Chris Miller 17 Jan '08

17 Jan '08

Why on earth I try and support Windows is beyond my comprehension. It's just something I do. I'm having some massive difficulties making Boost build with MinGW so I can start using it from Code::Blocks. I've tried using Boost's native Jam system, but first I have to build that. And that doesn't work. It keeps telling me this: C:\USERS\LORDSA~1\DESKTOP\BOOSTJAM>build.bat mingw ### ### Using 'msvc' toolset. ### C:\Users\LORDSA~1\DESKTOP\BOOSTJAM>rd /S /Q bootstrap C:\Users\LORDSA~1\DESKTOP\BOOSTJAM>md bootstrap C:\Users\LORDSA~1\DESKTOP\BOOSTJAM>cl /nologo /GZ /Zi /MLd /Fobootstrap/ /Fdboot strap/ -DNT -DYYDEBUG kernel32.lib advapi32.lib user32.lib /Febootstrap\jam0 co mmand.c compile.c debug.c execnt.c expand.c filent.c glob.c hash.c hdrmacro.c he aders.c jam.c jambase.c jamgram.c lists.c make.c make1.c newstr.c option.c outpu t.c parse.c pathunix.c regexp.c rules.c scan.c search.c subst.c timestamp.c vari able.c modules.c strings.c filesys.c builtins.c pwd.c class.c w32_getreg.c nativ e.c modules/set.c modules/path.c modules/regex.c modules/property-set.c modules /sequence.c modules/order.c 'cl' is not recognized as an internal or external command, operable program or batch file. C:\USERS\LORDSA~1\DESKTOP\BOOSTJAM>g++ G++: no input files C:\USERS\LORDSA~1\DESKTOP\BOOSTJAM> It keeps trying to use the MSVC toolset, even though I'm telling it not to. Any help here would be nice! I'm putting this all into an instruction set for potential programmers on my project here: http://www.fsdev.net/index.php?option=com_openwiki&Itemid=22&id=code:kw_pro… -- Registered Linux Addict #431495 http://profile.xfire.com/mrstalinman John 3:16! If Microsoft is the Wal*Mart of the Software World, then Linux is the Home Depot

5 15

ms vc80 works well but intel compiler10 fail
by 杨留安 17 Jan '08

17 Jan '08

here I have some class to test the boost::any_cast,usually I use intel compiler 10.0,but I got a message the application must be close. but the ms vc 8.0 compiler works well. My operator is windows xp,and I use vs2005 with inter compiler 10.0 here is my code: ----------------------------------------------------------------------------------------------------------------------------------------- class CA{public: void tellme() { cout<<"I am ca"<<endl; }}; class CB{public: void tellme() { cout<<"I am cb"<<endl; }}; class CC{public: void tellme() { cout<<"I am cc"<<endl; }}; class dotellme{public: void operator()(boost::any& ra) { if(CA* p = boost::any_cast<CA>(&ra)) { p->tellme(); } else if (CB* p = boost::any_cast<CB>(&ra)) { p->tellme(); } else if (CC* p = boost::any_cast<CC>(&ra)) { p->tellme(); } else if(string* p = boost::any_cast<string>(&ra)) { cout<<*p<<endl; } else { int a; try { a = boost::any_cast<int>(ra); cout<<"it's int:"<<a<<endl; } catch (boost::bad_any_cast& e) { cout<<"OOPS"<<endl; } } }}; //void tellme(boost::any& ra)//{// dotellme()(ra);//}int _tmain(int argc, _TCHAR* argv[]){ vector<boost::any> va; va.push_back(CA()); va.push_back(CB()); va.push_back(CC()); va.push_back(string("haha,it a string")); va.push_back(dotellme()); va.push_back(5); dotellme dt; for_each(va.begin(),va.end(),dotellme()); va.clear(); return 0;} _________________________________________________________________ 手机也能上 MSN 聊天了，快来试试吧！ http://mobile.msn.com.cn/

1 0

Boost 1.35 Release Date
by Chauhan, Vikas 17 Jan '08

17 Jan '08

Hello all, Following on the original post. Does anybody have anymore updates on BOOST 1.35 (rc1) release date? Is there a location on Boost website where I can get this information? Thanks and regards, Vikas > Message: 6 > Date: Wed, 28 Nov 2007 16:37:43 +0100 > From: "Tarjei Knapstad" <tarjei.knapstad(a)gmail.com> > Subject: Re: [Boost-users] Boost 1.35 release date > To: boost-users(a)lists.boost.org > Message-ID: > <ff5e09cc0711280737g1a1b919cw9633f00e0a1346de(a)mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > On 26/11/2007, Chauhan, Vikas <Vikas.Chauhan(a)anritsu.com> wrote: > > > > Hello, > > > > Can you please advise the planned release date for BOOST > 1.35 version. > > > > >From Beman's post on the main Boost list: > <quote> > * Merge for release complete: Sunday, November 25th. > * Start testing release candidate build process: Monday, December 3rd. > * Release Candidate 1: Monday, December 10th. > </quote> > > So I'm guessing early next year if there's not too much > trouble with the RC's. My guess is as good as yours though :) > > -- > Tarjei >

4 5

Re: [Boost-users] [boost] [switch_] Compound Cases [was Case Concept]
by Steven Watanabe 17 Jan '08

17 Jan '08

AMDG Joel de Guzman wrote: > Steven Watanabe wrote: > > >> Also, one thing that needs to be nailed down is: What Is I? >> Tobias was using it as the index into the list of labels: >> mpl::int_<0>, mpl::int_<1>, mpl::int_<2>, ... >> In my original interface and as you seem to be using it I is one of: >> mpl::at_c<labels, 0>::type, mpl::at_c<labels, 1>::type, >> mpl::at_c<labels, 2>::type >> > > It's in the concept I presented: > > I An MPL Integral Constant type > > Here: > > http://boost.org/libs/mpl/doc/refmanual/integral-constant.html > > Anything special I am missing? > What I meant was where they come from and thus how they should be interpreted. If you have mpl::vector_c<int, 4,2, 7> do you get mpl::int_<0>, mpl::int_<1>, and mpl::int_<2> or mpl::int_<4>, mpl::int_<2>, and mpl::int_<7>? In Christ, Steven Watanabe

2 1

Jamfiles and project
by Mojmir Svoboda 17 Jan '08

17 Jan '08

Hello, I'm having trouble assembling Jamfiles.v2 correctly: Let's assume i have following hierarchy: ~ - devel $DEVEL_ROOT +- boost $BOOST_ROOT +- mt +- flog Jamfile.v2, Jamroot +- doc +- test +-test1 - Jamfile.v2 . +-testN - Jamfile.v2 Q1: How do I include headers/project with absolute path in flog/test/test1/Jamfile.v2? <include>$DEVEL_ROOT or <library>$DEVEL_ROOT/mt/mt Right now it's relative to ~/devel/flog. Q2: How do I include something conditionaly based on threading model? something like if (threading==multi) { include project mt } To be honest, as a new user i'd appreciate some helloworldish project sample, from which the usage of building framework should be obvious. Many thanks, Mojmir

1 0

problems linking boost_date_time
by jerome.wacongne＠fxoverlay.com 17 Jan '08

17 Jan '08

Hi, Here is what I get when I try to link my unit tests agains boost_date_time. g++ -LC:\dev\workspace\fxoverlayFundation\Debug -LC:\dev\workspace\fxoverlayBusiness\Debug -LC:\dev\boost_1_34_1\stage\lib -ofxoverlayBusinessTests.exe emptyTests.o allTests.o InterestRateTests.o InterestCurveTests.o -lboost_date_time-mgw34-d-1_34_1 -lboost_unit_test_framework-mgw34-d-1_34_1 -lfxoverlayBusiness -lfxoverlayFundation C:\dev\workspace\fxoverlayBusiness\Debug/libfxoverlayBusiness.a(InterestCurve.o): In function `ZSt10_ConstructISt4pairIKSsN3com9fxoverlay8business12InterestRateEES6_EvPT_RKT0_': C:/dev/msys/MinGW/bin/../lib/gcc/mingw32/3.4.5/../../../../include/c++/3.4.5/bits/stl_tree.h:(.text$_ZN5boost9date_time15month_formatterINS_9gregorian10greg_monthENS0_13simple_formatIcEEcE12format_monthERKS3_RSo[boost::date_time::month_formatter<boost::gregorian::greg_month, boost::date_time::simple_format<char>, char>::format_month(boost::gregorian::greg_month const&, std::basic_ostream<char, std::char_traits<char> >&)]+0x38): undefined reference to `boost::gregorian::greg_month::as_short_string() const' C:/dev/msys/MinGW/bin/../lib/gcc/mingw32/3.4.5/../../../../include/c++/3.4.5/bits/stl_tree.h:(.text$_ZN5boost9date_time15month_formatterINS_9gregorian10greg_monthENS0_13simple_formatIcEEcE12format_monthERKS3_RSo[boost::date_time::month_formatter<boost::gregorian::greg_month, boost::date_time::simple_format<char>, char>::format_month(boost::gregorian::greg_month const&, std::basic_ostream<char, std::char_traits<char> >&)]+0x57): undefined reference to `boost::gregorian::greg_month::as_long_string() const' Both "libboost_date_time-mgw34-d-1_34_1.a" and "liboost_unit_test_framework-mgw34-d-1_34_1.a" are in "C:\dev\boost_1_34_1\stage\lib" Here are my includes : #include <boost/date_time/posix_time/ptime.hpp> #include <boost/date_time/posix_time/time_parsers.hpp> #include <boost/date_time/posix_time/time_formatters.hpp> using namespace boost::posix_time; The functions I actualy use are : from_iso_string() and to_simple_string() (wich build a ptime from a string and return a string from a ptime) I would apreciate any suggestion to make my program compile and link. Regards, Jérôme This message and any attachments (the "message") is intended solely for the addressees and is confidential. If you receive this message in error, please delete it and immediately notify the sender. Any use not in accord with its purpose, any dissemination or disclosure, either whole or partial, is prohibited except formal approval. The internet can not guarantee the integrity of this message. Overlay AM (and its subsidiaries) shall (will) not therefore be liable for the message if modified. --------------------------------------------- Ce message et toutes les pieces jointes (ci-apres le "message") sont etablis a l'intention exclusive de ses destinataires et sont confidentiels. Si vous recevez ce message par erreur, merci de le detruire et d'en avertir immediatement l'expediteur. Toute utilisation de ce message non conforme a sa destination, toute diffusion ou toute publication, totale ou partielle, est interdite, sauf autorisation expresse. L'internet ne permettant pas d'assurer l'integrite de ce message, Overlay AM (et ses filiales) decline(nt) toute responsabilite au titre de ce message, dans l'hypothese ou il aurait ete modifie.

2 2

Re: [Boost-users] Memory Leak Detection output format
by Jamie Cook 17 Jan '08

17 Jan '08

Gennadiy, thanks for the quick reply. I've followed your advice but with little luck :( I have a very simple test project which is not still giving me the same old output my std_afx.h (precompiled headers are turned off) // ----------------------------------------------------------------------------------------------------------- #pragma once #define _CRTDBG_MAP_ALLOC #include <crtdbg.h> // #ifdef _DEBUG // #include <new> // #define BOOST_TEST_DEBUG_NEW new( _CLIENT_BLOCK, __FILE__, __LINE__) // #define new BOOST_TEST_DEBUG_NEW // #else // #define BOOST_TEST_DEBUG_NEW // #endif #define BOOST_TEST_MAIN #include <boost/test/unit_test.hpp> #include <boost/test/floating_point_comparison.hpp> #include <boost/filesystem.hpp> // Windows Header Files: #define WIN32_LEAN_AND_MEAN // Exclude rarely-used stuff from Windows headers #include <windows.h> // ----------------------------------------------------------------------------------------------------------- my test cpp file // ----------------------------------------------------------------------------------------------------------- #include "std_afx.h" #include <iostream> BOOST_AUTO_TEST_CASE( SimpleTestCase ) { int i = 9; i++; int * j = new int[101]; } // ----------------------------------------------------------------------------------------------------------- and this generates the following output when i run the executable... ================================================ Running 1 test case... *** No errors detected Detected memory leaks! Dumping objects -> {203} normal block at 0x00368D40, 404 bytes long. Data: < > CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD {164} normal block at 0x00368158, 1280 bytes long. Data: <x 6 lationStatis> 78 83 36 00 6C 61 74 69 6F 6E 53 74 61 74 69 73 Object dump complete. ================================================ As you can see the test case introduces an easy to spot memory leak of 101 integers (at 4 bytes a pop = 404 bytes). I've done this with your debug block uncommented and the output is exactly the same... One thing I did have to do to get that block to compile was add the #include <new>, if that wasn't there I got an error from stlport stating that c:\program files\stlport-5.1.4\stlport\stl\_new.h(47) : fatal error C1189: #error : Cannot include native new header as new is a macro.-- I think the problem is that even though I've compiled stlport with settings telling it to use regular old new/delete and not it's own super-duper new/delete, that it is still using it's own internal versions of 'regular' new and delete... I'll try fiddling with some options in stlport unless you can think of any other reason it would continue to have this behaviour? Jamie Cook Research Engineer ………………………………………..………………… Veitch Lister Consulting Pty Ltd. Ground Floor, 80 Jephson St, PO Box 1054, Toowong, QLD 4066 AUSTRALIA Ph: +61 7 3870 4888 Fax: +61 7 3870 4446 Email: jamie.cook(a)veitchlister.com.au Web: www.veitchlister.com.au ………………………………………..…………………

2 1

Unserialize throws an exception
by Lorenzo Cavallaro 17 Jan '08

17 Jan '08

Dear all, I'm trying to unserialize a boost::archive::text_iarchive. The code used to work, but know (for no apparent reasons) it's calling terminate throwing the following exception. terminate called after throwing an instance of 'boost::archive::archive_exception what(): stream error Aborted The piece of code is very simple: std::ifstream ifs(filename); boost::archive::text_iarchive ia(ifs); ia >> map; where map is an STL hash_multimap. Any ideas? Btw, this is one of the first thing I'm doing in the code. Nonetheless to say, another code of mine which serialize and unserialize works like a charm... Boost library version is 1.34.1. TIA, bye Lorenzo -- Lorenzo `Gigi Sullivan' Cavallaro <sullivan(a)security.dico.unimi.it> GPG key at http://security.dico.unimi.it/~sullivan/sullivan.asc Until I loved, life had no beauty; I did not know I lived until I had loved. (Theodor Korner) See the reality in your eyes, when the hate makes you blind. (A.H.X)

3 4

[signals] problem when disconnecting slots
by raimar wagner 16 Jan '08

16 Jan '08

Hello! I'm trying to use the boost signals library to implement a callback system, but my first test-application (see boost_test_app.cpp in attachment) won't compile. The connect method works as expected, but if I'm trying to disconnect my slot (in line 28: " mySignal.disconnect(slot);") a compiler error message appears. The error message: /path_to_boost/include/boost-1 _34_1/boost/boost/function_equal.hpp:17: error: no match for 'operator==' in 'f == g' /path_to_boost/include/boost-1_34_1/boost/function/function_base.hpp:577: note: candidates are: bool boost::operator==(const boost::function_base&, boost::detail::function::useless_clear_type*) /path_to_boost/include/boost-1_34_1/boost/function/function_base.hpp:589: note: bool boost::operator==(boost::detail::function::useless_clear_type*, const boost::function_base&) My compile command: g++ -I/path_to_boost/include/boost-1_34_1/ -g -o boost_test_app boost_test_app.cpp -L/path_to_boost/lib -lboost_signals-gcc41 I'm using boost 1.34.1 with g++ (GCC) 4.1.0 (SUSE Linux) as compiler. Best regards Raimar

4 3

Casting a shared_ptr as a prototyped (void *)?
by Elli Barasch 16 Jan '08

16 Jan '08

I'd like to pass a shared_ptr as the entry argument to a thread function via pthread_create. However, the entry argument is prototyped as void *. I can't simply cast the shared pointer as a (void *). How do I go about this?

6 10

Re: [Boost-users] Boost.Test Framework to test a DLL
by Plonka Christian 16 Jan '08

16 Jan '08

Hi Ovanes, I found the problem. The console_test_runner and my NetCtrld.dll used different boost_unit_test_framework-....dll's. So I had two different master test suites, one in each DLL. Now it seems to be working. Thanks, Christian ________________________________ Von: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] Im Auftrag von Ovanes Markarian Gesendet: Dienstag, 15. Jänner 2008 18:40 An: boost-users(a)lists.boost.org Betreff: Re: [Boost-users] Boost.Test Framework to test a DLL Is there any chance that Boost.Test can not load (find) your DLL? Try to copy console_test_runner into the directory which contains your DLL? Debug DLLs under VC++ have usually a suffix 'D' before .dll. The letter NetCtrl->d<-.dll is probably this suffix. If not, add it, so that the name is NetCtrldD.dll With Kind Regards, Ovanes On Jan 15, 2008 9:55 AM, Plonka Christian <christian.plonka(a)amst.co.at> wrote: Hi, I'm planning to change from CppUnit to Boost.Test for unit testing since it is now possible to integrate unit tests also in a DLL. Nevertheless I do encounter some problems here. The code is (I guess you'll recognize it): //---------------------------------------------------------------------- --- #define BOOST_TEST_DYN_LINK #include <boost/test/unit_test.hpp> using namespace boost::unit_test; BOOST_AUTO_TEST_CASE( test1 ) // void test1() { int i = 0; BOOST_CHECK_EQUAL( i, 2 ); } BOOST_AUTO_TEST_CASE( test2 ) { BOOST_CHECKPOINT("About to force division by zero!"); int i = 1, j = 0; i = i / j; } extern "C" { __declspec(dllexport) bool init_unit_test() { printf("Initializing\n"); framework::master_test_suite().p_name.value = "Test runner test"; // framework::master_test_suite().add( BOOST_TEST_CASE( &test1 ) ); return true; } } //---------------------------------------------------------------------- --- I the sample (test_runner_test.dll) it works fine but when I use it in my DLL I get: C:\Projekte\Test\NetControl\bin>console_test_runner --test NetCtrld.dll Initializing Test setup error: test tree is empty I also tried the commented version using master_test_suite().add() without any change. Any hint what I can try or how I can debug this? Thanks, Christian _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

2 1

[program options] support for sections
by Ovanes Markarian 16 Jan '08

16 Jan '08

Hello! In my last project I remeber to see somewhere that boost::program_options library supports config files sections. But I can not find any hints to this feature in the current doc. Does anyone has an experience with sections in the config file? With Kind Regards, Ovanes

1 1

Re: [Boost-users] Boost.Test Framework to test a DLL
by Plonka Christian 16 Jan '08

16 Jan '08

Hi Ovanes, this is, why I added 'printf("Initializing\n"); to init_unit_test() in my DLL. As this is indeed printed before the error message, I think that the lib is found and loaded. Furthermore, if the lib cannot be found, I get: C:\Projekte\Test\NetControl\bin>console_test_runner --test NetCtrl.dll Boost.Test framework internal error: unknown reason So it seems that the DLL is loaded and init_unit_test() is invoked, but the test cases are not registered ... Christian ________________________________ Von: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] Im Auftrag von Ovanes Markarian Gesendet: Dienstag, 15. Jänner 2008 18:40 An: boost-users(a)lists.boost.org Betreff: Re: [Boost-users] Boost.Test Framework to test a DLL Is there any chance that Boost.Test can not load (find) your DLL? Try to copy console_test_runner into the directory which contains your DLL? Debug DLLs under VC++ have usually a suffix 'D' before .dll. The letter NetCtrl->d<-.dll is probably this suffix. If not, add it, so that the name is NetCtrldD.dll With Kind Regards, Ovanes On Jan 15, 2008 9:55 AM, Plonka Christian <christian.plonka(a)amst.co.at> wrote: Hi, I'm planning to change from CppUnit to Boost.Test for unit testing since it is now possible to integrate unit tests also in a DLL. Nevertheless I do encounter some problems here. The code is (I guess you'll recognize it): //---------------------------------------------------------------------- --- #define BOOST_TEST_DYN_LINK #include <boost/test/unit_test.hpp> using namespace boost::unit_test; BOOST_AUTO_TEST_CASE( test1 ) // void test1() { int i = 0; BOOST_CHECK_EQUAL( i, 2 ); } BOOST_AUTO_TEST_CASE( test2 ) { BOOST_CHECKPOINT("About to force division by zero!"); int i = 1, j = 0; i = i / j; } extern "C" { __declspec(dllexport) bool init_unit_test() { printf("Initializing\n"); framework::master_test_suite().p_name.value = "Test runner test"; // framework::master_test_suite().add( BOOST_TEST_CASE( &test1 ) ); return true; } } //---------------------------------------------------------------------- --- I the sample (test_runner_test.dll) it works fine but when I use it in my DLL I get: C:\Projekte\Test\NetControl\bin>console_test_runner --test NetCtrld.dll Initializing Test setup error: test tree is empty I also tried the commented version using master_test_suite().add() without any change. Any hint what I can try or how I can debug this? Thanks, Christian _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

Applying Boost license to non boost projects
by Roshan Naik 16 Jan '08

16 Jan '08

I am interested in adopting the Boost Software license for an open source C++ library (Castor: http://www.mpprogramming.com/2.html) The Boost FAQ describes how to apply the license to Boost header files. It basically states that the foll. lines should be included: // Copyright Joe Coder 2004 - 2006. // Distributed under the Boost Software License, Version 1.0. // (See accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt) - Does this same guideline apply to non boost projects ? The need for stating Boost.org there might confuse some users of non boost prj to think that the library is supplied by Boost. - Do all header files in my prj (which is a pure header library) need to include this text at the top/somewhere in order to adopt the license. -Roshan

2 1

[iostreams and serialization] serialize via iostreams
by Pooyan McSporran 15 Jan '08

15 Jan '08

I'm trying to use a combination of the iostreams library and the serialization library in order to be able to serialise objects and send them over a socket. I've written a class called "SocketDevice" which implements a bidirectional iostreams instance, and I'm trying to use this with serialization, in particular the binary_oarchive and binary_iarchive archive models. The problem is that when I do this, the code ends up calling a copy constructor (which doesn't exist) on my SocketDevice class. However I don't want to implement a copy constructor for this class as it doesn't make sense (as it manages a socket and a buffer for that socket, etc). The interface of my SocketDevice class looks like this: class SocketDevice : public boost::iostreams::device<boost::iostreams::bidirectional> { public: typedef char char_type; SocketDevice (int sock); ~SocketDevice (); std::streamsize read (char *s, std::streamsize n); std::streamsize write (const char *s, std::streamsize n); private: const int sock; char *buffer; size_t bufferSize; }; and I'm trying to do the object sending/receiving via template methods, eg: template<class C> void sendObject (const C& data, int sock) { boost::iostreams::stream_buffer<SocketDevice> buf (sock); std::ostream s (&buf); boost::archive::binary_oarchive oa (s); oa << data; } The copy constructor appears to be synthesised when the 'buf' object is instantiated in the above template method. My guess is that I'm doing something wrong, but I can't see what. I've also tried splitting my SocketDevice class from a bidirectional one to a pair of input & output ones, but with the same result. I've also tried making sendObject() a normal (non-template) method, this makes no difference. I realise that there are issues with trying to combine socket communications with object serialisation, but I'm still curious as to why I have this copy constructor issue.

2 2

Re: [Boost-users] templates w/ non-type arguments
by Hicham Mouline 15 Jan '08

15 Jan '08

I realize my question was not complete! Is it possible to write the 3 x 3 x 3 explicit instantiations, not with 27 lines of code, but perhaps in some kind of loop? Rds, -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Michael Gopshtein Sent: 15 January 2008 08:51 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] templates w/ non-type arguments "Hicham Mouline" <hicham(a)mouline.org> wrote in message news:20080114191355.0EA71200C2@ws6-7.us4.outblaze.com... > Hello, > > I have: > enum ParamType { FLAT, CURVE1D, CURVE2D }; > > template<ParamType r, ParamType q, ParamType v> > class TIP: > { > }; > > I would like to explicitly write all the possible full specializations of > TIP... > template class TIP<FLAT, FLAT, FLAT>; template<> class TIP<FLAT, FLAT, FLAT> {}; template<> class TIP<FLAT, FLAT, CURVE1D> {}; etc. > ... > > regards,

2 1

Boost.Test Framework to test a DLL
by Plonka Christian 15 Jan '08

15 Jan '08

Hi, I'm planning to change from CppUnit to Boost.Test for unit testing since it is now possible to integrate unit tests also in a DLL. Nevertheless I do encounter some problems here. The code is (I guess you'll recognize it): //---------------------------------------------------------------------- --- #define BOOST_TEST_DYN_LINK #include <boost/test/unit_test.hpp> using namespace boost::unit_test; BOOST_AUTO_TEST_CASE( test1 ) // void test1() { int i = 0; BOOST_CHECK_EQUAL( i, 2 ); } BOOST_AUTO_TEST_CASE( test2 ) { BOOST_CHECKPOINT("About to force division by zero!"); int i = 1, j = 0; i = i / j; } extern "C" { __declspec(dllexport) bool init_unit_test() { printf("Initializing\n"); framework::master_test_suite().p_name.value = "Test runner test"; // framework::master_test_suite().add( BOOST_TEST_CASE( &test1 ) ); return true; } } //---------------------------------------------------------------------- --- I the sample (test_runner_test.dll) it works fine but when I use it in my DLL I get: C:\Projekte\Test\NetControl\bin>console_test_runner --test NetCtrld.dll Initializing Test setup error: test tree is empty I also tried the commented version using master_test_suite().add() without any change. Any hint what I can try or how I can debug this? Thanks, Christian

2 1

MetaComm and Boost Consulting SVN daily snapshot services
by Joaquín Mª López Muñoz 15 Jan '08

15 Jan '08

Hello, Some time ago, there were two places where users could download daily snapshots of the Boost repository: * http://engineering.meta-comm.com/boost.aspx (linked from boost.org) * http://www.boost-consulting.com/boost.tar.bz2 Alas, none of them are being updated anymore. It'd be great if at least one could be put back in business, as this is a very useful service for casual Boost users without SVN tools. Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

1 0

[DateTime] format flag to print single-digit hour?
by Boris 15 Jan '08

15 Jan '08

I'm using the Boost.DateTime IO system to create customized date/time strings. While the library supports quite a lot of format flags out of the box I'm missing a format flag to print a single-digit hour. There is the format flag %k though (see http://www.boost.org/doc/html/date_time/date_time_io.html#date_time.format_…) - however it adds a preceding blank for single-digit hours. I guess there is currently no format flag for single-digit hours? Is it possible then to extend the Boost.DateTime IO system somehow to add your own format flag? While I could use another library like Boost.Format additionally I prefer to stick to one (it might make more sense then to switch to Boost.Format completely)? Boris

1 0

templates w/ non-type arguments
by Hicham Mouline 15 Jan '08

15 Jan '08

Hello, I have: enum ParamType { FLAT, CURVE1D, CURVE2D }; template<ParamType r, ParamType q, ParamType v> class TIP: { }; I would like to explicitly write all the possible full specializations of TIP... template class TIP<FLAT, FLAT, FLAT>; ... regards,

2 1