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

Creating a copy of a subset of a graph
by Adam Teasdale Hartshorne 12 Feb '06

12 Feb '06

Hi All, I want to form a new graph B from an existing graph A, copying out only edges(and their all their vertices) given that an edge contains at least one of the vertices in a given list (i.e. I am going to end up with a graph with more than just the vertices in the list and their adjacency information. With some extra vertices, where an edge exists between them and vertices in the defined list.) I know I should be using filter_graph and copy_graph, but not 100% sure how. Any help/code snippets would be much appreciated, I am new to boost graph library so please make any explanations as simple as possible, Adam

1 0

Template metaprogramming for compile-time initialization of complex types?
by Arno Schödl 12 Feb '06

12 Feb '06

Hello, here is my problem: I have a complex data type, like a std::vector< std::string > and need some global constant values of this type. The type is also used non-constant throughout the program, so I do not want to use dumb types, but rather STL containers. In an ideal world, initialization of these constants would not happen at runtime, but at compile time, essentially generating a constant memory image of this type via meta-programming. The exact memory layout is compiler-dependent, of course, but the difficult, but portable part would be the programmer's interface, in particular for nested types and user-defined ones. Why do I want this? Start-up time is a big problem for larger apps, and CRT init code filling std::vectors with stuff causes lots of page faults slowing down the start. This is particularly annoying if the constant never happens to be used in that particular run. And of course it is inelegant to keep data around twice, once in dumb constant arrays, and once on the heap after the containers have been filled from these arrays. Why can't I use POD arrays for constant data instead of STL containers? Because the same data type is often used both const (for default values, to list possible choices in menus, etc.) and non-const (for user-defined values). Unifying the access would be awfully convenient. I have implemented this approach with the boost preprocessor library for MSVC 8 std::vector< integral type >, which works, but is nowhere general enough. Has anyone else made attempts in this direction? Arno

1 0

Boost, STLport, and Visual Studio 2005
by Robert S. Grimes 11 Feb '06

11 Feb '06

Hi all, Trying to get Boost 1.33.1 and STLport 5.0.1 to co-exist under Micro$oft Visual Studio 2005. I got them both to apparently compile and install okay, but when I attempt a Boost::Serialization test, I run into a problem: LINK : fatal error LNK1104: cannot open file 'libboost_serialization-vc80-mt-gdp-1_33_1.lib' >From the docs, I indeed find that the "-gdp-" library is non-existent; specifically, the 'p' for STLport is missing. How do I build the STLport runtimes? Thanks! -RSG

2 1

Minor spurious warning message -- Very Trivial
by Robert S. Grimes 11 Feb '06

11 Feb '06

Hi all, Using Boost 1.33.1 serialization library with Micro$oft Visual Studio 2005, I get two minor warnings while compiling the demo_xml.cpp program. They are in boost/serialization/collections_save_imp.hpp. Lines 36 and 37 in the function template save_collection are the culprit, resulting in warnings of the form: warning C4267: 'initializing' : conversion from 'size_t' to 'unsigned int', possible loss of data These are the problem lines: unsigned int count = s.size(); ar << make_nvp("count", const_cast<const unsigned int &>(count)); Changing the "unsigned int" to "size_t" fixes the warnings, like this: size_t count = s.size(); ar << make_nvp("count", const_cast<const size_t &>(count)); Trivial problem, I admit, but warnings are warnings... Cheers, -RSG

2 1

Boost thread library wrapper
by boostguy＠kapiladhanvi.com 10 Feb '06

10 Feb '06

Hi all This mail is in continuation to the earlier mail I had written asking for help in writting a thread wrapper for boost.threads. We already have an existing interface to which we would like our wrapper for boost.threads to adhere to. We have a runnable class which is inherited by the applications wanting to create threads and they override the run method. This is how our application currently creates threads. Thanks Peter for your earlier mail explaining on how to use boost::shared_ptr and boost::bind Your code for a Runnable interface > struct Runnable > { > virtual ~Runnable() {} > virtual void run() = 0; > }; We have a class Runnable with nearly the same code as above. To make a dummy implementation, I made the run methd a normal function. Nothing derives from it.. run is what I want to execute in a thread. > > void execute_in_thread( boost::shared_ptr<Runnable> pr ) > { > boost::thread th( boost::bind( &Runnable::run, pr ) ); > } I am creating a thread in MyThread class using the following code. --- MyThread.h ---- typedef boost::shared_ptr<Runnable> myPtr; class MyThread { public: MyThread( ); ~MyThread( ); private: myPtr p1; } --- MyThread.cpp ---- MyThread::MyThread( ) : p1( new Runnable() ) { cout << "MyThread::MyThread( ) constructor " << endl; boost::thread th( boost::bind( &Runnable::run, p1) ); } ---- Runnable.cpp --- // this is the only function in Runnable.h void Runnable::run() { cout << "Runnable::run() YUPPIE! "<< endl; } ------ end ----------- The code compiles, but nothing happens.. I am missing something here ?? Is there something else I need to do to start my thread?? Thanks Dhanvi

4 6

newbie question
by Lawrence Lim 10 Feb '06

10 Feb '06

Hello, I'm trying to compile the following: #include <boost/type_traits/is_integral> int main() { return 0; } I'm getting a long_long_type is undeclared in namespace boost error. I worked around it by: namespace boost { typedef long long long_long_type; typedef unsigned long long ulong_long_type; } #include <boost/type_traits/is_integral> int main() { return 0; } Is there a better way to fix this issue? My compiler is gcc3.2.3 in Red Hat Linux. Thanks, Lawrence

1 0

[Unit Test Framework]
by Evan Carew 10 Feb '06

10 Feb '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I'm trying out Unit Test for the first time & decided to try out the first example listed in the getting started docs. It compiled fine on my g++ v4.0 boost v1_32 setup, but I get the following output which looks wrong to me: [root@r3-mobile src]# ./test_watchdog Running 1 test case... /root/projects/IBS/test_watchdog/src/test_watchdog.cpp(38): error in "free_test_function": test 2 == 1 failed Exception in "free_test_function": signal: memory access violation /usr/src/build/570409-i386/BUILD/boost_1_32_0/libs/test/build/../src/unit_test_suite.cpp(150): fatal error in "free_test_function": Testing aborted *** 1 failure detected (1 failure expected) in test suite "Unit test example 1" Is there an easily explained (and implemented) way to get the output for this test app to work? Thanks, Evan Carew -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFD7NkKpxCQXwV2bJARAmHfAJ40gh9/WXZe+sXCNgCZzPfea4nciQCgojG1 9L8mgvqBJxLAxE9V1UAy2tw= =KfmW -----END PGP SIGNATURE-----

2 1

Re: [Boost-users] [program_options] Bug in program_options with vc-8_0 build target when displaying the help descriptions
by boost＠nogga.de 10 Feb '06

10 Feb '06

Hi Will, thanks for the answer. I still ask myself what is the correct way of handling these situations with the new standard library. I had a similar issue in my own software and I didn't liked the solution to fall back to manually track the array bounds. The only way to write this line safe would be: > if (par_end - line_begin < line_length - indent) in order to compute a distance between the two involved iterators. But is this really better? Is the old line really a safety problem? Or is this type of problem the penalty one have to pay for the enhanced security in other areas? Best regards Dirk Will Bryant schrieb: > Hmm, Vladimir Prius doesn't seem to be responding on the list at the > moment, but FWIW this is a known bug - see my November 2005 posts > about it in the mailing list archive. Vladimir was working on a fix, > but the last version he sent still had the problems... > > Will > > > boost(a)nogga.de wrote: > > Hello, > > > > there is a bug in the help output generated from > > > > void format_paragraph(std::ostream& os, std::string par, unsigned > > first_column_width, unsigned line_length) > > > > function. The error message is cryptic and something like that: > > > > _Myptr + _Off <= (((_Mystring *)this->_Mycont)->_Myptr() + > > ((_Mystring *)this->_Mycont)->_Mysize) && _Myptr + _Off >= > > ((_Mystring *)this->_Mycont)->_Myptr(): > > > > The point is that the new vc8.0 compiler (or better the vs2005 > > standard library) performs a lot of additional iterator checking. > > In the function format_paragraph the "line_begin" iterator is > > checked, whether there is one more line to indent in the help > > output or whether this is the last line. During this process, the > > "line_begin" iterator is incremented behind the end of the current > > string "par.end()" and this will trigger the > > _SCL_SECURE_VALIDATE_RANGE warning from the new library. > > > > if (line_begin + (line_length - indent) > par_end) { line_end = > > par_end; } > > > > You should be able to trigger this behavior with every application > > when printing the help text for an option that is longer than one > > line. > > > > This new iterator checking has it's pros and cons. While it is good > > to find any sort of error, how would you safely write the case > > above without reverting to index like operations? > > > > Best regards Dirk _______________________________________________ > > Boost-users mailing list Boost-users(a)lists.boost.org > > http://lists.boost.org/mailman/listinfo.cgi/boost-users > > > >

1 0

Re: [Boost-users] pd darwin Symbol not found
by RIVASSEAU Jean Noel 10 Feb '06

10 Feb '06

Do you develop in XCode ? I remember that while developing on XCode there is an option that does fast linking (it is called Zero Link) and if you check that, it does not really link your application at link time but only at run time. So I think this is in fact a linking problem, disable the Zero Link option under XCode and your code will probably fail to link. As to why it does not link, I do not know, which library (boost) are you trying to link to ? Jean-Noël _____ De : boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] De la part de Locatelli Jérôme Envoyé : jeudi 9 février 2006 14:14 À : boost-users(a)lists.boost.org Objet : [Boost-users] pd darwin Symbol not found I'am writing a application on MacOS 10.4.4 using darwin 8 with gcc 4, I compile and link my application without problem but when I run my application , I've got the error bellow. Has anybody got an idear?? Thanks. dyld: lazy symbol binding failed: Symbol not found: __ZN5boost6detail3tss4initEPNS_9function1IvPvSaINS_13function_baseEEEE Referenced from: /Users/locatelli/Projets/ics/IcsServer/obj.ics/../lib/libIcsServerCommandSystem.0 Expected in: flat namespace dyld: Symbol not found: __ZN5boost6detail3tss4initEPNS_9function1IvPvSaINS_13function_baseEEEE Referenced from: /Users/locatelli/Projets/ics/IcsServer/obj.ics/../lib/libIcsServerCommandSystem.0 Expected in: flat namespace Trace/BPT trap --------------------------------------------------- Locatelli Jérôme Institut Laue-Langevin Service Contrôle des instruments 6 rue Jules Horowitz, BP 156 38042 Grenoble cedex 9 http://www.ill.fr/ Tel: 33 (0)4 76 20 76 27 Fax: 33 (0)4 76 20 77 00 ---------------------------------------------------

3 2

bcp: towards a library dependency list
by John Maddock 10 Feb '06

10 Feb '06

I've just committed some changes to bcp that give it a shortened list option: bcp --list-short truncates the list so you just see directory names where appropriate, for example: bcp --list-short shared_ptr.hpp Gives the dependency list: boost/assert.hpp boost/checked_delete.hpp boost/config.hpp boost/config boost/current_function.hpp boost/detail boost/mpl boost/non_type.hpp boost/preprocessor boost/shared_ptr.hpp boost/static_assert.hpp boost/throw_exception.hpp boost/type.hpp boost/type_traits boost/version.hpp which is reasonably nice and compact :-) Now, here's the question: what did the folks who were asking for this functionality actually want to use the information for? I should stress that it's overkill for extracting subsets of Boost, because many libraries will depend on just one or two files inside boost/mpl or boost/preprocessor, so using the list for that purpose will result in an even "thicker" dependency than copying with bcp already produces. It's possible to contract the above list even further: for example by contracting to library names where possible, so for example contracting all of: boost/regex.hpp boost/regex libs/regex to just "regex" However there would still be a lot of Boost headers that can't automatically be assigned to any library name, so it's not clear to me how much use this would be? If folks can give me an idea of what they were looking to use a dependency list for it'll help me decide whether to stop here or to continue trying to contract the list further. John.

1 0