June 2004 - Boost-users - lists.boost.org

RE: [Boost-users] protassign
by Nat Goodspeed 24 Jun '04

24 Jun '04

> -----Original Message----- > From: Erik Thiele [mailto:erik@thiele-hydraulik.de] > Sent: Wednesday, June 23, 2004 11:28 AM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] protassign > > now i can do: > > void foo() > { > protassign<bool> foorunning_assigner(foorunning, true); > dojobs(); > dofoooo(); > } > > and that's all. is there one such thing in boost already??? [Nat] In the forthcoming Boost serialization library: http://www.rrsd.com/boost/ there's a state_saver template class that does what you're talking about: http://www.rrsd.com/boost/libs/serialization/doc/state_saver.html This seems useful enough to promote to a library more general than the serialization library.

3 3

failures and featurerequest in/for numeric_cast
by Peter Piehler 24 Jun '04

24 Jun '04

failures and featurerequest in/for numeric_cast I use boost in a little project for my study. In some cases the software must convert numeric datatypes. So I tried to convert with boost-library numeric_cast and found the following problems (gcc version 3.2.3 20030422): -the maximum could not be handled correctly, for example: +numeric_cast<int>(numeric_limits<unsigned int>::max()); doesn't throw a exception -no precession handling, example: +numeric_cast<float>(numric_limits<double>::epsilon()+1); doesn't throw a exception -no NaN, bounded/unbounded and exact handling. -no specialization. I need numeric_cast specializiation for: +numeric_limits-interface can't use for datatypes with caracteristics only known at runtime. +datatyps without static_cast support (example: gmp-library) So I have extended the boost::numeric_cast for my special requirements. You can download it under the following link: http://www.pienet.de/source/numeric_cast.tar.gz . Compile and run numeric_cast_tescases.cpp with USE_BOOST_CAST to test ::boost::numeric_cast. The output is: Running 2 test cases... numeric_cast_testcases.cpp(72): error in "standard_cast_test": exception bad_numeric_cast is expected numeric_cast_testcases.cpp(91): error in "standard_cast_test": exception bad_numeric_cast is expected numeric_cast_testcases.cpp(93): error in "standard_cast_test": exception bad_numeric_cast is expected numeric_cast_testcases.cpp(135): error in "standard_cast_test": exception bad_numeric_cast is expected *** 4 failures detected in test suite "numeric_cast Test" I think the line static_cast<X>(static_cast<Y>(x)) != x; is a tricky code. For X/Y==char/unsigned char,short/unsigned short etc. is always false. The compiler convert it only if the datatypes have not the same size/bytes or are between integers and floating points. sorry for my bad english and code, it's the first time that i ever have published. Peter Bestellen Sie Y! DSL und erhalten Sie die AVM "FritzBox SL" für 0. Sie sparen 119 und bekommen 2 Monate Grundgebührbefreiung. http://de.adsl.yahoo.com

2 1

bug in date_time? cannot initialize ptime to not_a_date_time
by Erik Thiele 24 Jun '04

24 Jun '04

hi. #include <boost/date_time/posix_time/ptime.hpp> int main() { boost::posix_time::ptime x(boost::date_time::not_a_date_time); } # g++ --version 2.95.4 #g++ -Wall -c foo.cpp -Iboost_1_31_0 (compiling into cache) foo.cpp: In function `int main()': foo.cpp:5: no matching function for call to `boost::posix_time::ptime::ptime (boost::date_time::special_values)' boost/boost/date_time/posix_time/ptime.hpp:39: candidates are: boost::posix_time::ptime::ptime(boost::gregorian::date, boost::posix_time::time_duration) boost/boost/date_time/posix_time/ptime.hpp:42: boost::posix_time::ptime::ptime(boost::gregorian::date) boost/boost/date_time/posix_time/ptime.hpp:45: boost::posix_time::ptime::ptime(const boost::posix_time::int64_time_rep &) boost/boost/date_time/posix_time/ptime.hpp:48: boost::posix_time::ptime::ptime(const boost::posix_time::ptime &) foo.cpp:5: warning: unused variable `class boost::posix_time::ptime x' -- Erik Thiele

2 1

protassign
by Erik Thiele 23 Jun '04

23 Jun '04

hi in some cases i do assign values to variables just for the life of my function: void foo() { bool oldfoorunning = foorunning; foorunning = true; dojobs(); dofoooo(); foorunning = oldfoorunning; } now the problem arises when dojobs throws an exception. in order to be safe i have to do: void foo() { bool oldfoorunning = foorunning; foorunning = true; try { dojobs(); dofoooo(); } catch (...) { foorunning = oldfoorunning; throw; } foorunning = oldfoorunning; } this is ugly and it can get more ugly by having more than one such variable. then you stack try/catch into each other and code gets really ugly and error prone. for that reason i created a class as follows: template<class X> class protassign : public boost::non_copyable { public: inline protassign (X &var_, const X &newvalue) : var(var_) { oldvalue = var; var = newvalue; } inline ~protassign () { try { var = oldvalue; } CATCHDESTRUCTOR; //<-- my special destructor error catching macro, //forget it. (is there an alternative in boost?) } private: X &var; X oldvalue; }; now i can do: void foo() { protassign<bool> foorunning_assigner(foorunning, true); dojobs(); dofoooo(); } and that's all. is there one such thing in boost already??? cu erik -- Erik Thiele

1 0

Re: Re: compiling boost with mingw in winxp
by Alexis H. Rivera-Rios 23 Jun '04

23 Jun '04

Hi, 1) g++ is in the path. Doing the g++ --version displays the version information: C:\opt\boost_1_31_0>g++ --version g++ (GCC) 3.2.3 (mingw special 20030504-1) Copyright (C) 2002 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. 2) Mingw is installed in the default directory: c:\mingw Does the fact that I got a warning about spawnvp function when I built bjam has something to do with this? C:\opt\boost-jam-3.1.10>.\bootstrap.mingw\jam0 -f build.jam --toolset=mingw "--t oolset-root= " ...found 83 targets... ...updating 4 targets... .cc. bin.ntx86\yyacc.exe .cc. bin.ntx86\mkjambase.exe .cc. bin.ntx86\jam.exe execnt.c: In function `execcmd': execnt.c:651: warning: passing arg 3 of `spawnvp' from incompatible pointer type execnt.c:682: warning: passing arg 3 of `spawnvp' from incompatible pointer type .link. bin.ntx86\bjam.exe 1 file(s) copied. ...updated 4 targets... I read somewhere something about different signatures for the spawn function and this indicates the one used by bjam is not the one that mingw supports. Why would spawn fail? Is this an OS issue or a compiler thing? Is bjam the program that fails? Or is it a python script that fails? How can I trace this error? I really appreciate your help, Alex ===== Programming Tutorial: In Python: To do this, do this In Perl: To do this, do this or this or this or this... In C: To do this, do this, but be careful In C++: To do this, do this, but don't do this, be careful of this, watch out for this, and whatever you do, don't do this __________________________________ Do you Yahoo!? Yahoo! Mail - 50x more storage than other providers! http://promotions.yahoo.com/new_mail

3 3

Re: [Boost-users] Re: compiling boost with mingw in winxp
by Alexis H. Rivera-Rios 22 Jun '04

22 Jun '04

Hi, This is the output of running the command line with the d+4 and d+2 option respectively. I'm running the command from the dos shell (cmd.exe). As far as the filesystem. The filesystem is NTFS. Thanks, Alex C:\opt\boost_1_31_0>bjam -sTOOLS=mingw -sPYTHON_ROOT=c:\Python23 -sPYTHON_VERSIO N=2.3 -d+4 Executing raw command directly Executing command [g++] [-Wl,--exclude-symbols,_bss_end__:_bss_start__:_data_end __:_data_start__] [-Wl,--out-implib,bin\boost\libs\date_time\build\boost_date_ti me.dll\mingw\debug\boost_date_time-mgw-d-1_31.lib] [-g] [-shared] [-mno-cygwin] [-o] ["bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debug\boost_date _time-mgw-d-1_31.dll"] ["bin\boost\libs\date_time\build\boost_date_time.dll\ming w\debug\greg_month.obj"] ["bin\boost\libs\date_time\build\boost_date_time.dll\mi ngw\debug\greg_weekday.obj"] ["bin\boost\libs\date_time\build\boost_date_time.dl l\mingw\debug\date_generators.obj"] spawn: Invalid argument C:\opt\boost_1_31_0>bjam -sTOOLS=mingw -sPYTHON_ROOT=c:\Python23 -sPYTHON_VERSIO N=2.3 -d+2 gcc-Link-action bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debug\b oost_date_time-mgw-d-1_31.dll bin\boost\libs\date_time\build\boost_date_time.dll \mingw\debug\boost_date_time-mgw-d-1_31.lib g++ -Wl,--exclude-symbols,_bss_end__:_bss_start__:_data_end__:_data_start_ _ -Wl,--out-implib,bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debu g\boost_date_time-mgw-d-1_31.lib -g -shared -mno-cygwin -o "bin\boost\libs\dat e_time\build\boost_date_time.dll\mingw\debug\boost_date_time-mgw-d-1_31.dll" "bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debug\greg_month.obj" "bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debug\greg_weekday.obj " "bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debug\date_generator s.obj" spawn: Invalid argument ===== Programming Tutorial: In Python: To do this, do this In Perl: To do this, do this or this or this or this... In C: To do this, do this, but be careful In C++: To do this, do this, but don't do this, be careful of this, watch out for this, and whatever you do, don't do this __________________________________ Do you Yahoo!? Yahoo! Mail - 50x more storage than other providers! http://promotions.yahoo.com/new_mail

4 4

Dynamic bitset help
by Michael Short 21 Jun '04

21 Jun '04

Hello please anyone some one must have a example or something using the dynamic bitset? Has anyone got a few good examples of the dynamic bitset been used with member functions like append and block type? If so could you please send in some thanks. Michael

1 0

Re: [Boost-users] Newbie: What is the actual return type?
by Doug Gregor 21 Jun '04

21 Jun '04

[Sorry about the top-post; switching e-mail accounts] On Sun, 20 Jun 2004, Tim Rowe wrote: > > You're actually getting back an (Lvalue) property map. Think of it > > like an std::map<vertex_descriptor, std::string>, because it maps > > from the vertices in the graph g to the vertex name (which is an > > std::string here). The difference is that the property map is fast > > (O(1)) whereas the std::map would be slow (O(log |V|)). > > > > Interior properties will get a bit easier with the next Boost > > release. > > > > > > Doug > > Thanks. The property_map documentation says that "the get() function > takes a property map object [...] and a key object"; is that the get() > that's used there? Well, no :(. The get() being used there is actually part of the PropertyGraph concept, documented here: http://www.boost.org/libs/graph/doc/PropertyGraph.html When you have an interior property p (like vertex_name_t in the example graph) and graph g, the expression get(p, g) gives you a property map for that particular property and graph, that can be "indexed" with the appropriate key (e.g., a vertex or edge descriptor). > So vertex_name is a map of maps, with a Graph as a key? This analogy actually works (much to my surprise!), so long as you don't push it too far :) So yes, get(vertex_name, g) returns a map from vertices in g to the name of the vertex. > And where is the interface to property_map defined? > boost_1_31_0\libs\graph\doc\using_property_maps.html says that it's > defined in boost_1_31_0\libs\property_map\property_map.html, but that > just > contains another example, and I can't find which header declares it -- I > can't see it in property_map.hpp which is where I expected it to be! The actual struct template property_map is used (only!) to get the return type of, e.g., get(vertex_name, g). It is defined in boost/graph/properties.hpp, but the definition won't help much because I don't think that's quite what you're asking for. When you use get(vertex_name, g) you get back a property map. There's no single type that defines a property map, because it's a concept. To find out what you can do with a property map, check out: http://www.boost.org/libs/property_map/property_map.html Specifically, you want the links to the ReadablePropertyMap, WriteblePropertyMap, and LvaluePropertyMap concepts. Doug

1 0

Re: [Boost-users] compiling boost with mingw in winxp
by Alexis H. Rivera-Rios 21 Jun '04

21 Jun '04

Janusz, I'm using version 3.1.0-1 of mingw (its basically the current release that you can download at www.mingw.org) THanks, Alex PS. Sorry for posting in another thread, I don;t know how to reply in the current thread. Alexis H. Rivera-Rios writes: > Hi, > > I downloaded boost 1.31.0 and I'm using bjam version > 3.1.10. What mingw version are you using? Regards, Janusz ===== Programming Tutorial: In Python: To do this, do this In Perl: To do this, do this or this or this or this... In C: To do this, do this, but be careful In C++: To do this, do this, but don't do this, be careful of this, watch out for this, and whatever you do, don't do this __________________________________ Do you Yahoo!? Yahoo! Mail - You care about security. So do we. http://promotions.yahoo.com/new_mail

4 4

Newbie: What is the actual return type?
by Tim Rowe 20 Jun '04

20 Jun '04

In the Kevin Bacon example in the Boost::Graph documentation, we declare: typedef adjacency_list < vecS, vecS, undirectedS, property < vertex_name_t, std::string >, property < edge_name_t, std::string > > Graph; Graph g; and later declare and call: typedef property_map < Graph, vertex_name_t >::type actor_name_map_t; actor_name_map_t actor_name = get(vertex_name, g); What are we actually getting back here? I can see that we can index it (we do later), but is it just an array? Or a vector? Or what? I assume I am /supposed/ to know (not like Ada's "type foo is private"), so that I know what I can do with it! TIA, Tim Rowe

2 2

Re: staticially linked STLport and Boost
by Lukáš Krejčí 20 Jun '04

20 Jun '04

Hi, I have encountered similar problem that you solved after "some fiddling". Can you specify what exactly you did to successfully compile Boost.Python? Thanks, L. >> try >> >> bjam "-sTOOLS=msvc-stlport" "-sBUILD=debug <runtime-link>static ><stlport-iostream>on" > >The default Boost.Python settings seem to conflict with ><runtime-link>static, but after some fiddling I worked out how to fix >it. I thought it wouldn't need this, but I also needed to force >"<threading>multi" to get it passed the final set of linking problems. > Of course, it seems to work at the moment, and I'll probably be back >here if/when I start getting strange crashes :) > > Thanks for the pointer in the right direction :) >

1 0

BGL on VC6.0
by Boris Kagan 20 Jun '04

20 Jun '04

Hi All, I have a problem in compiling my very first example in BGL - quick_tour.cpp, using boost-1_31, VC 6.0 and native Microsoft stl. I've got the following error messages: c:\boost\include\boost-1_31\boost\graph\detail\adjacency_list.hpp(1055) : error C2244: 'bidirectional_graph_helper_with_property<Config>::remove_edge' : unable to resolve function overload c:\boost\include\boost-1_31\boost\graph\detail\adjacency_list.hpp(1057) : error C2954: template definitions cannot nest c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(45) : error C2143: syntax error : missing ';' before '<' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(45) : error C2059: syntax error : '<' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(45) : error C2039: 'type' : is not a member of '`global namespace'' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(45) : error C2238: unexpected token(s) preceding ';' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(52) : error C2653: 'mpl' : is not a class or namespace name c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(52) : error C2143: syntax error : missing ';' before '<' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(52) : error C2059: syntax error : '<' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(56) : error C2238: unexpected token(s) preceding ';' c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(57) : see reference to class template instantiation 'is_incrementable_::impl<T>' being compiled c:\boost\include\boost-1_31\boost\detail\is_incrementable.hpp(69) : fatal error C1903: unable to recover from previous error(s); stopping compilation My question is : May I use BGL 1_31 using VC6.0 with all restrictions of VC6.0 compiler? Thanks a lot for your advice, Kagan Boris.

1 0

compiling boost with mingw in winxp
by Alexis H. Rivera-Rios 18 Jun '04

18 Jun '04

Hi, I downloaded boost 1.31.0 and I'm using bjam version 3.1.10. I want to compile the code with Mingw by doing the following: bjam -s"TOOLS=mingw" and I get the following message: C:\opt\boost_1_31_0>bjam -s"TOOLS=mingw" install ...found 8817 targets... ...updating 3260 targets... MkDir1 C:\Boost MkDir1 C:\Boost\lib gcc-Link-action bin\boost\libs\date_time\build\boost_date_time.dll\mingw\debug\b oost_date_time-mgw-d-1_31.dll bin\boost\libs\date_time\build\boost_date_time.dll \mingw\debug\boost_date_time-mgw-d-1_31.lib spawn: Invalid argument I have no idea were to start looking to debug the code. If it may help, I downloaded the source code for bjam and compiled it and get the following messages: C:\opt\boost-jam-3.1.10>build mingw ### ### Using 'mingw' toolset. ### C:\opt\boost-jam-3.1.10>rd /S /Q bootstrap.mingw C:\opt\boost-jam-3.1.10>md bootstrap.mingw C:\opt\boost-jam-3.1.10>gcc -DNT -o bootstrap.mingw\jam0.exe command.c compile. c execnt.c execunix.c execvms.c expand.c filent.c fileos2.c fileunix.c filevms.c glob.c hash.c hdrmacro.c headers.c jam.c jambase.c jamgram.c lists.c make.c mak e1.c newstr.c option.c parse.c pathunix.c pathvms.c regexp.c rules.c scan.c sear ch.c subst.c timestamp.c variable.c modules.c strings.c filesys.c builtins.c pwd .c class.c w32_getreg.c native.c modules/set.c modules/path.c modules/regex.c m odules/property-set.c modules/sequence.c modules/order.c execnt.c: In function `execcmd': execnt.c:651: warning: passing arg 3 of `spawnvp' from incompatible pointer type execnt.c:682: warning: passing arg 3 of `spawnvp' from incompatible pointer type C:\opt\boost-jam-3.1.10>.\bootstrap.mingw\jam0 -f build.jam --toolset=mingw "--t oolset-root= " clean ...found 1 target... ...updating 1 target... .rm. clean ...updated 1 target... C:\opt\boost-jam-3.1.10>.\bootstrap.mingw\jam0 -f build.jam --toolset=mingw "--t oolset-root= " ...found 83 targets... ...updating 4 targets... .cc. bin.ntx86\yyacc.exe .cc. bin.ntx86\mkjambase.exe .cc. bin.ntx86\jam.exe execnt.c: In function `execcmd': execnt.c:651: warning: passing arg 3 of `spawnvp' from incompatible pointer type execnt.c:682: warning: passing arg 3 of `spawnvp' from incompatible pointer type .link. bin.ntx86\bjam.exe 1 file(s) copied. ...updated 4 targets... The executables are generated, but I suspect the spawn warning has to do with the build error. I'll appreciate your help, Alex ===== Programming Tutorial: In Python: To do this, do this In Perl: To do this, do this or this or this or this... In C: To do this, do this, but be careful In C++: To do this, do this, but don't do this, be careful of this, watch out for this, and whatever you do, don't do this __________________________________ Do you Yahoo!? Read only the mail you want - Yahoo! Mail SpamGuard. http://promotions.yahoo.com/new_mail

2 1

INI File Parser?
by Duane Murphy 18 Jun '04

18 Jun '04

A while back, I noted that there was an INI file parser as part of boost. It was obscure, but I think it was in the program_options library. I note references to program_options searching the web site, but I'm not finding anything in the latest builds. I'm specifically interested in parsing INI files (I'm on a Macintosh) but the program_options library also seemed interesting. Is there anything available? What's the status of the program_options library? Thanks for any pointers, ...Duane

4 4

How to Compile boost::function With a NoArg Member Function ?
by Eduardo Bezerra 17 Jun '04

17 Jun '04

Hi, I need to use boost::function in a context like this one: struct Widget { string ToString() {return "Widget"; } }; int main() { function<string () f(&Widget::ToString); } What's the correct syntax to sucessfully compile this example ? Best regards, Eduardo _________________________________________________________________ MSN Messenger: instale grátis e converse com seus amigos. http://messenger.msn.com.br

2 1

Newbie question -- graph library example
by Tim Rowe 17 Jun '04

17 Jun '04

In the example examples/file_dependencies.cpp, I can't see what either of the typedefs for Color do: typedef property_map<Graph,vertex_color_t>::type Color; If I comment then out everything seems still to work. What are they there for? TIA, Regards, Tim

2 1

[Tuple] broken for gcc-3.4.0
by Jürgen Hunold 17 Jun '04

17 Jun '04

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi ! After updating my cvs snapshot at 12.15 , the tuple library is broken for gcc 3.4.0. My own program stopped compiling and when going to libs/tuple/test and running bjam I get: In file included from /ive/hunold/packages/boost/boost/tuple/tuple.hpp:38, from ../../../libs/tuple/test/tuple_test_bench.cpp:19: /ive/hunold/packages/boost/boost/tuple/detail/tuple_basic.hpp: In function `typename boost::tuples::access_traits<typename boost::tuples::element<N, boost::tuples::cons<HT, TT> >::type>::non_const_type boost::tuples::get(boost::tuples::cons<HT, TT>&)': /ive/hunold/packages/boost/boost/tuple/detail/tuple_basic.hpp:212: error: `get_class' undeclared (first use this function) (full output of failing tests attached) So I think this is really something which gcc 3.4.0 does no handle. Yours, Jürgen - -- * Dipl.-Math. Jürgen Hunold ! Institut für Verkehrswesen, Eisenbahnbau * voice: ++49 511 762-2529 ! und -betrieb, Universität Hannover * fax : ++49 511 762-3001 ! Appelstrasse 9a, D-30167 Hannover * hunold(a)ive.uni-hannover.de ! www.ive.uni-hannover.de -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFA0ZJ7ljbJ/LLrxrYRAhxCAJ9xnzG7bfPORsLVU9pjhD90RjP5CQCbBgL6 g154KH2PULNGi8pBFIDsiTM= =XvKx -----END PGP SIGNATURE-----

2 3

boost::filesystem::path corruption
by Stuart Bingë 17 Jun '04

17 Jun '04

Hello all, I've recently started using some of the Boost libraries, one of which being boost::filesystem. I have however encountered a problem while using the library, which as far as I can tell is a bug. I'm not sure if anyone is aware of this problem, or if is not an issue with the library itself but rather with my configuration - in either case a description of the problem follows: Whenever I instantiate a boost::filesystem::path object and try to print out a text representation of the path using a method such as .string(), the path itself is printed out, however it looks as though the string is not terminated properly as a whole bunch of trailing garbage appears after the path. The following sample program produces the problem for me: #include <iostream> #include <boost/filesystem/path.hpp> int main(int c, char** v) { boost::filesystem::path testpath("test/path"); std::cout << testpath.native_file_string() << std::endl; return 0; } When I run this code I get output similar to the following: test/pathpÿÿÿÿtest/ÿÿÿÿdataÿÿÿÿpath... etc (a whole bunch of other random stuff like days of the week & month, as well as garbage character strings) Can anyone shed some light on this for me? I'm running boost version 1.31.0 on Gentoo Linux x86 (kernel 2.6.5-gentoo-r1) with GCC version 3.3.2, for anyone who is interested. Regards, Stuart Bingë

2 1

limiting searches in BGL
by Jason Crosswhite 16 Jun '04

16 Jun '04

Hello, I am just starting to use the BGL, and have been beating my head upon a problem. I am needing to find an irregularly bounded set of points in three dimensional space. The points are within a convex hull and the graph edges are edges in a delaunay triangulation. I want to start at a node known to be on the hull, and crawl through the inside with either depth first or breadth first search. If you are on the hull, you would not allow the search to continue outside the hull. I thought of doing this with a simple visitor, and color the unwanted vertices black. However, from inside the scope of the visitor, I do not believe I can have access to the colors. Several ugly hacks come to mind, but I would like to know if there is an easy way of doing this first? -Jason Crosswhite

3 4

Re: limiting searches in BGL
by Vladimir Prus 16 Jun '04

16 Jun '04

Hi Jason, > I am just starting to use the BGL, and have been beating my head upon a > problem. I am needing to find an irregularly bounded set of points in > three dimensional space. The points are within a convex hull and the > graph edges are edges in a delaunay triangulation. I want to start at a > node known to be on the hull, and crawl through the inside with either > depth first or breadth first search. If you are on the hull, you would > not allow the search to continue outside the hull. I think you might want to use 'TerminatorFunc' argument to depth_first_visit. That's a functor which returns true if the search should not continue after a certain vertex. - Volodya

1 0

dynamic boost::array and performance/size of STL vector
by Bo Peng 15 Jun '04

15 Jun '04

Dear list, I frequently need to use arrays with undetermined (at compile time) but constant (user specify size) size. I.e., I need something like boost::array<int> arr(5); without arr.insert etc, rather than boost::array<int,5> arr; which only accept constant array length. Of course I can use 'vector<int> arr(5)' in this case but I am worry about the performance and size overhead, specifically: 1. How is size() implemented in vector<>? I am concerned with this function since icc (intel c++ compiler) can not vectorize 'for(i=0;i<arr.size();i++)' unless arr.size() is an inlined reference to a variable. 2.Since I will have a lot of such arrays, I would also like to know how many additional variables vector<int> keeps. I.e. exactly how big is vector<int> arr(5)? It might be trivial to copy boost::array and add constructor/destructor but I do not want to re-invent the wheel. Thanks. Bo

4 5

RE: [Boost-users] Regex lookahead/lookbehind
by James Gunn 15 Jun '04

15 Jun '04

*** Before acting on this e-mail or opening any attachment you are advised to read the disclaimer at the end of this e-mail *** Hi John, > -----Original Message----- > From: John Maddock [mailto:john@johnmaddock.co.uk] > Sent: 15 June 2004 12:40 > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Regex lookahead/lookbehind > > > > Hi all, > > > > Is it possible to do lookahead and/or lookbehind within > boost::regex? > > Lookahead is supported at present, lookbehind will be in the > next revision (currently in cvs in the branch called regex5). > > John. Thanks for the quick response. In the end I managed to solve my current problem without needing lookahead or lookbehind. I used your (earlier) advice of checking for word boundaries at the beginning and end of my regular expression and this has helped loads! The latest version of my postcode validation expression is shown below: (\bGIR\s0AA\b)| (\b[^QVX\W0-9][^IJZ\W\s]([0-9]{1,2}?|[ABCDEFGHJKSTUW0-9]?|[0-9][ABEHMNPRVWXY ]?) \s+\d[^CIKMOV]{2}\b) I'm certain that the expression could be made simpler and that it could also be made more efficient but I'm not sure where. Cheers, James ********************************************************************** Please note: This e-mail and its attachments contain only the opinions of the sender and do not necessarily reflect the policy(s) of the communisis group in general. Employees of the communisis group are required not to make any defamatory statements and not to infringe or authorise any infringement of copyright or any other legal right by e-mail. Any such communication is therefore outside the scope of employment of the individual concerned. The communisis group will not accept any liability in respect of such a communication. Confidentiality: This e-mail and any attachments, together with their contents, are confidential unless otherwise explicitly stated in writing by the sender of this e-mail and are for the intended recipient only. If they have come to you in error you must not take any action in respect of them, which includes but is not limited to reproducing, sending or storing them, other than to notifying the sender immediately of the mistake, and deleting the e-mail, any attachments and any reproductions made by replying to it. Viruses: This e-mail and any attachments have been scanned for viruses but we cannot guarantee that they are virus free. The recipient should check this e-mail and any attachments for viruses. The communisis group accepts no responsibility for any damage caused by any virus transmitted by this e-mail or any of its attachments. In the event of any unauthorised copying or forwarding, the recipient will be required to indemnify the communisis group against any claim for loss or damage caused by any viruses or otherwise. ********************************************************************** ______________________________________________________________________ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email ______________________________________________________________________

1 0

Regex lookahead/lookbehind
by James Gunn 15 Jun '04

15 Jun '04

*** Before acting on this e-mail or opening any attachment you are advised to read the disclaimer at the end of this e-mail *** Hi all, Is it possible to do lookahead and/or lookbehind within boost::regex? Thanks in advance James Gunn Software Developer --LongSig Computer Bureau Communisis DM Manston Lane Crossgates Leeds LS15 8AH Telephone +44 (0)113 225 5306 Fax +44 (0)113 225 5921 Email James.Gunn(a)communisis-dm.co.uk ********************************************************************** Please note: This e-mail and its attachments contain only the opinions of the sender and do not necessarily reflect the policy(s) of the communisis group in general. Employees of the communisis group are required not to make any defamatory statements and not to infringe or authorise any infringement of copyright or any other legal right by e-mail. Any such communication is therefore outside the scope of employment of the individual concerned. The communisis group will not accept any liability in respect of such a communication. Confidentiality: This e-mail and any attachments, together with their contents, are confidential unless otherwise explicitly stated in writing by the sender of this e-mail and are for the intended recipient only. If they have come to you in error you must not take any action in respect of them, which includes but is not limited to reproducing, sending or storing them, other than to notifying the sender immediately of the mistake, and deleting the e-mail, any attachments and any reproductions made by replying to it. Viruses: This e-mail and any attachments have been scanned for viruses but we cannot guarantee that they are virus free. The recipient should check this e-mail and any attachments for viruses. The communisis group accepts no responsibility for any damage caused by any virus transmitted by this e-mail or any of its attachments. In the event of any unauthorised copying or forwarding, the recipient will be required to indemnify the communisis group against any claim for loss or damage caused by any viruses or otherwise. ********************************************************************** ______________________________________________________________________ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email ______________________________________________________________________

2 1

Regex failing to match postcodes
by James Gunn 15 Jun '04

15 Jun '04

*** Before acting on this e-mail or opening any attachment you are advised to read the disclaimer at the end of this e-mail *** Hi all, I'm having some difficulty getting my regular expression working. Basically, I need to make sure that a UK postcode is valid. The postcode that is passed to my function sometimes has extra things with it such as: Wakefield, WF1 3RD Shrewsbury SY2 5PT Shropshire It now seems to be failing to find the postcode in the above examples. Also, when I pass my function a postcode that I know is invalid, such as JG2 7L5 it matches it as G2 7L5 instead of failing to do the match. The regular expression I'm using is below: (?:(?:(^|\s)+ A[BL]|B[ABDHLNRST]?| C[ABFHMORTVW]|D[ADEGHLNTY]|E[CHNX]?|F[KY]|G[LUY]?| H[ADGPRSUX]|I[GMPV]|JE|K[ATWY]|L[ADELNSU]?|M[EKL]?| N[EGNPRW]?|O[LX]|P[AEHLOR]|R[GHM]|S[AEGKLMNOPRSTWY]?| T[0ADFNQRSW]|UB|W[ACDFNRSV]?|YO|ZE) \d(?:\d|[A-Z])?\s+\d[A-Z]{2}($|\s)+) Can anyone tell me whats wrong with my expression? BTW I'm using boost 1.31.0 on VC++ 7.1, Windows XP. James Gunn Software Developer --LongSig Computer Bureau Communisis DM Manston Lane Crossgates Leeds LS15 8AH Telephone +44 (0)113 225 5306 Fax +44 (0)113 225 5921 Email James.Gunn(a)communisis-dm.co.uk ********************************************************************** Please note: This e-mail and its attachments contain only the opinions of the sender and do not necessarily reflect the policy(s) of the communisis group in general. Employees of the communisis group are required not to make any defamatory statements and not to infringe or authorise any infringement of copyright or any other legal right by e-mail. Any such communication is therefore outside the scope of employment of the individual concerned. The communisis group will not accept any liability in respect of such a communication. Confidentiality: This e-mail and any attachments, together with their contents, are confidential unless otherwise explicitly stated in writing by the sender of this e-mail and are for the intended recipient only. If they have come to you in error you must not take any action in respect of them, which includes but is not limited to reproducing, sending or storing them, other than to notifying the sender immediately of the mistake, and deleting the e-mail, any attachments and any reproductions made by replying to it. Viruses: This e-mail and any attachments have been scanned for viruses but we cannot guarantee that they are virus free. The recipient should check this e-mail and any attachments for viruses. The communisis group accepts no responsibility for any damage caused by any virus transmitted by this e-mail or any of its attachments. In the event of any unauthorised copying or forwarding, the recipient will be required to indemnify the communisis group against any claim for loss or damage caused by any viruses or otherwise. ********************************************************************** ______________________________________________________________________ This email has been scanned by the MessageLabs Email Security System. For more information please visit http://www.messagelabs.com/email ______________________________________________________________________

3 2

[uBLAS] Compilation problem with VC 7.1
by kjfh kjnk 15 Jun '04

15 Jun '04

Hi, This simple code creates 2 ublas vector and computes the inner_product. It does not compile with VC 7.1: --------------------------------------------------- #include <iostream> #include <string> #include <boost/concept_check.hpp> #include <boost/numeric/ublas/matrix.hpp> namespace ublas = boost::numeric::ublas; int main(int argc, char **argv) { ublas::vector<unsigned short> toto(3); ublas::vector<double> tata(3); double titi = ublas::inner_prod(toto, tata); return 0; } --------------------------------------------------- Here is the compilation: *************************************************** C:\Temp\ublas>cl -I"C:\Program Files\boost_1_31_0" /EHsc ublas-test.cpp Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.3052 for 80x86 Copyright (C) Microsoft Corporation 1984-2002. All rights reserved. ublas-test.cpp ublas-test.cpp(12) : error C2440: 'initializing' : cannot convert from 'boost::numeric::ublas::vector_scalar_binary_traits<E1,E2,F>::result_type' to 'double' with [ E1=boost::numeric::ublas::vector<unsigned short>, E2=boost::numeric::ublas::vector<double>, F=boost::numeric::ublas::vector_inner_prod<unsigned short,double,boost::mpl::if_c<false,unsigned short,boost::numeric::ublas::promote_traits<unsigned short,double>::builtin_promote_type>::type> ] No user-defined-conversion operator available that can perform this conversion, or the operator cannot be called C:\Temp\ublas> **************************************************** It compiles well with with VC 6.0, but the result is not good... (with a proper initialisation) Anybody has an idea ? JD ___________________________________________________________ALL-NEW Yahoo! Messenger - sooooo many all-new ways to express yourself http://uk.messenger.yahoo.com

2 2