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

RTTI for VC 6
by Samuel 16 Aug '04

16 Aug '04

VC 6 has an option for Run-Time Type Information (RTTI) that is normally off. I was able to compile and execute a sample program that had a problem. I noticed that Boost uses RTTI (such as the dynamic_cast operator), so I checked to see if the option was on; it was not. I turned the option on and the program worked. That problem is probably a known problem, but something such as the following can be added somewhere to not allow a VC 6 program to even compile if the RTTI option is off. I think it could make things a little easier for some beginners. # if !defined(_CPPRTTI) # error "Boost requires use of RTTI" # endif

6 22

Re: [Boost-users] Re: RTTI for VC 6
by Samuel 16 Aug '04

16 Aug '04

[The message I am replying to was sent from a newsgroup and I am sending my reply (this) as a message to the list.] "Edward Diener" <eddielee(a)tropicsoft.com> wrote in message news:cfmhi0$4fd$1@sea.gmane.org... > > What Mr. Dimov is saying is that it is impossible to place the code you > would like to use above in Boost because it would affect other Boost > libraries which do not need RTTI to work. That's no problem. The Boost headers already have more than 400 (including comment) lines that test for the "BOOST_MSVC" preprocessor symbol. In other words, there is already an abundance of code that is relevant only to MS VC. > Furthermore if it were placed in > the header file of a library which did need RTTI to work, just including > that header file, without actually using code in it that needed RTTI such as > dynamic_cast, would cause an immediate compile time error. Perhaps. I hope, though, that the possibility that it might not be useful won't prevent someone from communicating the idea to someone familiar with the relevant headers, since there is a possibility that it might be useful. I have more confidence in the developers that a solution is possible. I might have missed something, but I don't understand why it is relevant to say that it won't work. It might not work, but it might work. Yet you are saying it definitely won't work. That seems innapropriate to me.

2 2

RE: [Boost-users] RTTI for VC 6
by Bjorn.Karlsson＠readsoft.com 16 Aug '04

16 Aug '04

Enough of this. There will be no more insults, and no more blame, in this thread. Please try again, in a more productive manner. Thanks, Bjorn Karlsson Moderator [Everything snipped]

1 0

Boost.Python 'Fatal Error'
by Ed Fear 16 Aug '04

16 Aug '04

Hello there, I've just started using Boost.Python, and find it absolutely invaluable. For the moment, my main task is embedding the python interpreter within an application of mine, and I have successfully built a small application that can load a module, execute a Python function and then collect the result. The next logical step was to extend the Python script somewhat into objects. I wrote some extra code in the script which I checked with the command line interpreter and verified to work. Before changing the C++ code to actually /use/ these new objects, I decided to just see what would happen if I kept exactly the same C++ code but with the new Python script (which still contains the function used for the last example). It ran fine - producing the correct output - but then exited abruptly with the following message: "Fatal Python error: PyThreadState_Get: no current thread". The changes to the original script file were the addition of two new classes - OptionsGUI and Button - and, oddly, commenting out Button makes the program work again (I say oddly because it's simply a class with a constructor, nothing more). Evidently I'm using the library incorrectly, and have searched through the documentation for things I might have done wrong but alas, I do find it rather daunting. I also noticed that my code is quite different from examples - is this a correct way to be doing things? And yes, at present, I'm not doing any exception or error handling - I'm a little unclear on it but I'm sure I will be able to do it given time :) For those wishing to look at the source, I have put the two files - EmbeddedPythonFunc.cpp and script2.py - in my webspace to prevent everyone having to download them as attachments. They are available at the following addresses: http://kutsurogu.pwp.blueyonder.co.uk/EmbeddedPythonFunc.cpp http://kutsurogu.pwp.blueyonder.co.uk/script2.py (incidentally, I didn't include stdafx.h because it simply includes iostream, tchar.h and boost/python.hpp). I hope I have sent my help request to the right place! Apologies if I have not. Thank you in advance, Ed Fear

2 1

polymorphic container/boost::any
by goochrules! 16 Aug '04

16 Aug '04

I'm quite sure that this problem has been addressed many times before, so I suspect it as a better name and I suspect that boost::any could be the solution. I want to store related types in the same stl container, for instance, this is the class heirachy I want: class tree { public: void grow() =0; }; class oak : public tree { public: void grow() { std::cout << "oak grows slowly" << std::endl; }; class pine : public tree { public: void grow() { std::cout << "pine grows quickly" << std::endl; }; Then, I try to put them in a std::list: std::list< tree > forest; forest.push_back(oak()); forest.push_back(pine()); ...and I find out that I can't define tree::grow as abstract. So I give it a dummy implentation -- void grow() { std::cout << "tree grows" << std::endl; --, and try to iterate over the objects: for(std::list< tree >::iterator iter = forest.begin(); forest.end() != iter; ++iter) { iter->grow(); } ...and I don't get what I expect: tree grows tree grows I then think about trying to use a std::list< tree& > and discover that it won't compile. I think briefly consider std::list< tree* > but decide this dosen't meet requirements of memory management. So I do some searching for 'polymorphic container' and 'polymorphic iterator' and come up with boost::any, but I've no luck in making it work. Attached is my code. In the end, I'm looking for a way to add related types to a stl container and not have to do anything complicated to call a method on the type's abstract interface and to not have to wory about memory management. Maybe this question is better suited for comp.lang.c++[.moderated]? -- Matthew Peltzer -- goochrules(a)gmail.com

4 6

problem with static linking on linux gcc 2.95.3
by Daniel Huber 16 Aug '04

16 Aug '04

Hi, I am wondering if any of you boost gurus can help with this problem. I have compiled and installed boost using gcc 2.95.3 on a linux redhat 7 machine. The install actually fails on some targets. I get this message at the end of the install: ...failed updating 24 targets... ...skipped 48 targets... ...updated 2343 targets... But I am guessing that this is expected behavior for an old compiler. Anyway, I am interested in the filesystem library at this time, and so I tried compiling the simple_ls.cpp example. When I compile using the default linking (which uses shared objects), the program works fine, but when I compile using the -static flag to limit to static linking, I get a lot of undefined reference errors (see below). Unfortunately, I haven't been able to figure out why this would happen. Any suggestions on how to debug this or as to what might be going wrong would be greatly appreciated. Sincerely, Daniel /tmp/ccU0ehTU.o: In function `main': /tmp/ccU0ehTU.o(.text+0xc): undefined reference to `boost::filesystem::initial_path(void)' /tmp/ccU0ehTU.o(.text+0x2c): undefined reference to `boost::filesystem::native(basic_string<char, string_char_traits<char>, __default_alloc_template<true, 0> > const &)' /tmp/ccU0ehTU.o(.text+0x3e): undefined reference to `boost::filesystem::path::path(char const *, bool (*)(basic_string<char, string_char_traits<char>, __default_alloc_template<true, 0> > const &))' /tmp/ccU0ehTU.o(.text+0x50): undefined reference to `boost::filesystem::system_complete(boost::filesystem::path const &)' /tmp/ccU0ehTU.o(.text+0xe8): undefined reference to `boost::filesystem::exists(boost::filesystem::path const &)' /tmp/ccU0ehTU.o(.text+0x10d): undefined reference to `boost::filesystem::path::native_file_string(void) const' /tmp/ccU0ehTU.o(.text+0x188): undefined reference to `boost::filesystem::is_directory(boost::filesystem::path const &)' /tmp/ccU0ehTU.o(.text+0x1ad): undefined reference to `boost::filesystem::path::native_directory_string(void) const' /tmp/ccU0ehTU.o(.text+0x257): undefined reference to `boost::filesystem::is_directory(boost::filesystem::path const &)' /tmp/ccU0ehTU.o(.text+0x289): undefined reference to `boost::filesystem::path::leaf(void) const' /tmp/ccU0ehTU.o(.text+0x2f5): undefined reference to `boost::filesystem::path::leaf(void) const' /tmp/ccU0ehTU.o(.text+0x427): undefined reference to `boost::filesystem::path::native_file_string(void) const' /tmp/ccU0ehTU.o(.text+0x5ab): undefined reference to `boost::filesystem::path::leaf(void) const' /tmp/ccU0ehTU.o: In function `boost::filesystem::directory_iterator::directory_iterator(boost::filesystem::path const &)': /tmp/ccU0ehTU.o(.boost::filesystem::directory_iterator::gnu.linkonce.t.(boost::filesystem::path const &)+0x26): undefined reference to `boost::filesystem::detail::dir_itr_init(boost::shared_ptr<boost::filesystem::detail::dir_itr_imp> &, boost::filesystem::path const &)' /tmp/ccU0ehTU.o: In function `boost::filesystem::directory_iterator::dereference(void) const': /tmp/ccU0ehTU.o(.boost::filesystem::directory_iterator::gnu.linkonce.t.dereference(void) const+0x12): undefined reference to `boost::filesystem::detail::dir_itr_dereference(boost::shared_ptr<boost::filesystem::detail::dir_itr_imp> const &)' /tmp/ccU0ehTU.o: In function `boost::filesystem::directory_iterator::increment(void)': /tmp/ccU0ehTU.o(.boost::filesystem::directory_iterator::gnu.linkonce.t.increment(void)+0x12): undefined reference to `boost::filesystem::detail::dir_itr_increment(boost::shared_ptr<boost::filesystem::detail::dir_itr_imp> &)' collect2: ld returned 1 exit status

1 0

Re:Re: Regex help needed, please help
by Kosta Adarchenko 16 Aug '04

16 Aug '04

Message: 8 Date: Sun, 15 Aug 2004 10:31:50 +0100 From: "John Maddock" <john(a)johnmaddock.co.uk> Subject: Re: [Boost-users] Regex help needed, please help. To: <boost-users(a)lists.boost.org> Message-ID: <00a601c482ab$78c9dc10$d8ec0352@fuji> Content-Type: text/plain; charset="iso-8859-1" > John, I have stated the exact regex, it's \d\d\d\d\d\d , a simple regex matching 6 numbers, but the same happens even with one \d regex. I don't see a way to refactor the \d expression. OK, in that case I don't know what the problem is - are you able to generate a test case (mail me privately if the files are large)? And again, from where is the exception being thrown (in other words is it regex that's at fault)? John. ---------------------------------- The exception is being thrown from the search function, I generated a test case and sent it to your email at john(a)johnmaddock.co.uk, files are fairly large compressed 246kb, opened the take about 25MB (Love to see zip on text files). Thanks ahead Kosta. --------------------------------- Do you Yahoo!? New and Improved Yahoo! Mail - 100MB free storage!

1 0

Regex help needed, please help.
by Kosta Adarchenko 15 Aug '04

15 Aug '04

boost-users-request(a)lists.boost.org wrote: Send Boost-users mailing list submissions to boost-users(a)lists.boost.org To subscribe or unsubscribe via the World Wide Web, visit http://www.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..." Today's Topics: 1. Re: Regex problems, please help (John Maddock) ---------------------------------------------------------------------- Message: 1 Date: Fri, 13 Aug 2004 11:39:54 +0100 From: "John Maddock" Subject: Re: [Boost-users] Regex problems, please help To: Message-ID: <01d401c48125$ee4e79f0$3c520352@fuji> Content-Type: text/plain; charset="iso-8859-1" > After some seven thousands hits the match algo throughs the memory exhausted exception. > So I cought the exception and moved the iterator by one, printed the location it points to in my searched file to view the problematic section for the regex find algorithm. > I have noticed that the problem happens not with all large files, but with quite a bit of them. > After I have received the problematic section in my file, I ran the program on it and it did not raise any exceptions. > Please , help me I have a Graduate project to complete within 3 weeks from now, and this is one of the major bugs I have not resolved. You don't say *exactly* what the regex is that's causing the problem - it makes a difference - problems like this are usually caused by an expression that behaves badly with specific input (because regexes are NP-Hard to match in the general case). Boost.Regex will throw an exception rather than go on thrashing forever trying to find a match, but to determine whether this is the case here, you need to trap the exception at the point it's being thrown, and see if the regex lib is indeed the culprit, if it is, then refactoring the expression that you're using so that it's less ambiguous is usually the answer. John. ------------------------------ _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://www.lists.boost.org/mailman/listinfo.cgi/boost-users End of Boost-users Digest, Vol 279, Issue 2 ******************************************* John, I have stated the exact regex, it's \d\d\d\d\d\d , a simple regex matching 6 numbers, but the same happens even with one \d regex. I don't see a way to refactor the \d expression. --------------------------------- Do you Yahoo!? New and Improved Yahoo! Mail - Send 10MB messages!

2 1

[boost-users]_BGL:_copy_component_bugs
by Erik Jenelius 13 Aug '04

13 Aug '04

Hi. In boost/graph/copy.hpp, besides the copy_graph function, I noticed a nice, undocumented function called copy_component which uses breadth-first search to copy a component from a graph to a new graph. Unfortunately, it doesn't work. I think there are at least two bugs in it: 1. The vertices are copied at the examine_vertex event point and the edges are copied at the examine_edge event point. However, the target of the edge has not yet been examined at that point, which leads to strange results. If the edges are copied at the black_target event point instead, the algorithm seems to work correctly. 2. The std::vector orig2copy is templated on the vertex descriptor type of the input graph, while it should be the output graph. The function copy_graph is correct here. Also, both copy_graph and copy_component by default assume the vertices have a vertex_all property and the edges an edge_all property, which are copied to the output graph. Wouldn't it be more intuitive to not assume and copy any properties by default? - Erik Jenelius

1 0

Regex problems, please help
by Kosta Adarchenko 13 Aug '04

13 Aug '04

Here is the problem: I want to find a simple regular _expression something like \\d\\d\\d\\d\\d\\d in a large file (15 mb) using VC++ 7.1 . I use Spirit's File Iterator as the input iterator for the search algorithm. After some seven thousands hits the match algo throughs the memory exhausted exception. So I cought the exception and moved the iterator by one, printed the location it points to in my searched file to view the problematic section for the regex find algorithm. After the "problematic section" of some 300 chars ends the algorithm keeps working fine. here is the code. //------------------------------------------------------------------------------------------- void TheFunction() { //print the problematic section here ofstream of; of.open("d:\\err1.txt"); boost::spirit::file_iterator< > begin("d:\\log1.txt"); boost::spirit::file_iterator<> beginning_saved(begin); //save the beginning in order to compare later boost::spirit::file_iterator< > ending; ending= begin.make_end(); boost::match_results<boost::spirit::file_iterator<char > > what; boost::match_flag_type flags = boost::match_default; long matchescounter=0; std::exception e; int i; bool found; found=false; try { found=boost::regex_search(begin, ending, what, _expression, flags); if (found) { begin =what[0].second; cout<<what[0]; matchescounter=1; } } catch (std::exception &e) { cout<<e.what(); } while(found) { try { found=boost::regex_search(begin, ending, what, _expression, flags); begin =what[0].second; if (found) { cout<<matchescounter<<"----------"<<what[0]<<"-------------------"<<what[0].second-beginning_saved<<"\n"; matchescounter++; } } catch (std::exception &e) { cout<<*begin; of<<*begin; begin+=1; } } cout<<"---------------the end------------\n"; cout<<matchescounter; of.close(); } //--------------------------------------------------------------------------------------- I have noticed that the problem happens not with all large files, but with quite a bit of them. After I have received the problematic section in my file, I ran the program on it and it did not raise any exceptions. Please , help me I have a Graduate project to complete within 3 weeks from now, and this is one of the major bugs I have not resolved. Yours truely, Kosta. Desparatly waiting for your reply. --------------------------------- Do you Yahoo!? Y! Messenger - Communicate in real time. Download now.

2 1