February 2006 - Boost-users

progress_timer writes two newlines?
by Daniel Lidström 08 Feb '06

08 Feb '06

Hello! Why does boost::progress_timer write two newlines on destruction? One would suffice for me. Hälsningar, Daniel

1 0

[Regex] : Unknown character sequence.
by Sandeep Bantia 08 Feb '06

08 Feb '06

Hi, I'm new to regex. This is basically what I was trying. #include <iostream> #include <boost/regex.hpp> int main() { std::string input = "sand._12"; boost::regex reg(".*\._[0-9]+"); ...... } I get a warning which says "Unknown escape sequence '\.' . Its the same case with any other special character or even '\s' . I guess I'm missing something very simple. Any ideas ? Thanks in advance. -- -Sandeep.

2 1

Re: [Boost-users] [Regex] : Unknown character sequence.
by paul.floyd 08 Feb '06

08 Feb '06

> Hi, > > I'm new to regex. This is basically what I was trying. > > #include <iostream> > #include <boost/regex.hpp> > > int main() > { > std::string input = "sand._12"; > boost::regex reg(".*\._[0-9]+"); Hi In C++, you need to double your escape characters, so that in order to pass '.*\._[0-9]+' to the RE, you need to pass '.*\\._[0-9]+' to the compiler. With the string that you use, the compiler will strip the sole backslash, and the RE will get '.*._[0-9]+', which isn't what you want. A+ Paul Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34 /mn) ; tél : 08 92 68 13 50 (0,34/mn)

1 0

Re: [Boost-users] Multi_index Container
by Soni, Aashit 08 Feb '06

08 Feb '06

Hi Joaquín It worked fine. Thanks for the quick answer. I really appreciate this. now, I can go ahead and use it in my project. regards, aashit ________________________________ From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Joaquín Mª López Muñoz Sent: Wednesday, February 08, 2006 1:16 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Multi_index Container "Soni, Aashit" ha escrito: Hi Joaquín, Éric I appreciate your response, Before that I got some workaround from the web and the snippet I have written to port it to my existing project worked until I face real problem. My observation is like when I have Ordered_index with std::string, the code I have written compiles really well and works also. But when I remove this ordered_index, the code does not compile.Here is the code that compiles: <<Multi_index_container_working.txt>>Here is the code that does not compile: <<Multi_index_container_not_working.txt>>If you see the difference I have only two lines commented. Yes, I'm able to reproduce the problem you report under MSVC 6.0. Two remarks: * The problem does not show with the CVS version of Boost.MultiIndex (to be shipped in Boost 1.34), as it includes a number of optimizations alleviating the choking of MSVC 6.0 with very long symbol names, which is the core reason for this weird behavior. * Until you switch to Boost 1.34 (and even when you do), you can try avoiding the use of tags and using numeric identifiers instead. Tags add some stress to such a weak compiler as MSVC 6.0. The attached file is a variation of your Multi_index_container_not_working.txt file where tags have been suppressed and numerical identifiers used instead. I have checked it out to compile and work OK under MSVC 6.0. I hope this helps, please keep me informed otherwise, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

1 0

Re: [Boost-users] Multi_index Container
by Soni, Aashit 08 Feb '06

08 Feb '06

Hi Joaquín, Éric I appreciate your response, Before that I got some workaround from the web and the snippet I have written to port it to my existing project worked until I face real problem. My observation is like when I have Ordered_index with std::string, the code I have written compiles really well and works also. But when I remove this ordered_index, the code does not compile. Here is the code that compiles: <<Multi_index_container_working.txt>> Here is the code that does not compile: <<Multi_index_container_not_working.txt>> If you see the difference I have only two lines commented. ... //ordered_non_unique<tag<name>,BOOST_MULTI_INDEX_MEMBER(employee,std::string,name) >, .... .... //typedef boost::multi_index::index<employee_set,name>::type employee_set_by_name;... Same boost::multi_index::index<employee_set,ssnumber>::type is not working if the std::string is not indexed. right now I am working with dummy string. But I need to scrap the string and have some real code in place that should go into my project. If I am not able to solve this. I can not use multi_index_container and it is real loss for me. Help always appreciated. Thanks in advance. aashit ps: please consider these txt files are cpp files. ________________________________ From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Joaquín Mª López Muñoz Sent: Tuesday, February 07, 2006 9:39 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Multi_index Container Hello Aashit, "Soni, Aashit" ha escrito: Hi all, I am newbie to Boost library. I am trying to use multi_index_container library in my project. The problem I have at my hands needs One Hash index and one ordered indexed. I refer the Boost tutorial and Advance topic and tried to write the following snippet. But when I compiled it gives compilation error. I searched lot of different options and tried to fix it but I could not able to do it so. If you can guide me how to shut the compilation errors it will be great. If anyone of you is having any decent example tried out for Hash having ordered indexes please send it to me so I can go through and build my logic on top of that.... Code Follows. I am using MSVC SP6 on Win XP . This is primarily the problem: To make MSVC 6.0 compile a program using Boost.MultiIndex you've got to observe some stricter rules than in the normal case. Please read the section on MSVC 6.0 at http://boost.org/libs/multi_index/doc/compiler_specifics.html#msvc_60 typedef employee_set::index<employee_set,ssnumber>::type employee_set_by_ssn; Problem #1: MSVC 6.0 does not support the "index" nested typedef. Instead, write typedef index<employee_set,ssnumber>::type employee_set_by_ssn; employee_set_by_ssn& ssn_index=es.get<2>(); Problem #2: MSVC 6.0 does not support the get member function, use the global version instead: employee_set_by_ssn& ssn_index=get<2>(es); With these changes, and also tweaking /Zm and /ZI (as explained in the aforementioned section), your program compiles. Bare in mind that your compiler is very old and broken and making Boost.MultiIndex work for it can be quite a challenge, although it is doable. Good luck with your project, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 1

[program_options]allow_unregistered() ignored?
by Merrill Cornish 08 Feb '06

08 Feb '06

I have this problem that while program options works just fine for processing registered options, an unregistered option triggers a MS runtime exception (I'm using MinGW, so it uses MS runtime) that can't be caught at the C++ level. My store() function is shown below: po::store(po::command_line_parser(argc, argv). options(all_verbose). positional(pd). run(), vm); I saw the allow_unregistered() feature. I thought this would be a way to get all illegal command line options "out of the way" so they couldn't trigger the exception. So I modified the above to be: po::store(po::command_line_parser(argc, argv). options(all_verbose). positional(pd). allow_unregistered(). run(), vm); However, putting an illegal argument on the command line, say --x, still gives the message, "This application has requested the Runtime to terminate it in an unusual way." Is an unregistered option triggering some sort of NULL access? Merrill

2 1

Creating static libraries of boost
by boostguy＠kapiladhanvi.com 08 Feb '06

08 Feb '06

Hi all, I am trying to compile boost into static libraries on windows, but I am not able to create static libraries. Can someone point out what I am missing ? I am on a windows XP machine with mingW installed and in path. This is a simple task but somehow I am not getting static libraries. The command line I am using on my command prompt is : bjam "-sTOOLS=gcc" "-sBUILD=release static multi" stage The output libraries created are .lib and .dll, but the filenames are not starting with libboost-*.lib , they are just boost-*.lib. According to the http://boost.org/more/getting_started.html#Build_Install , windows static libraries should have lib*.lib as the format for the name.. Can someone tell me what I am doing wrong ? Thanks. Dhanvi

2 1

Re: [Boost-users] [Regex] : Unknown character sequence.
by Sohail Somani 08 Feb '06

08 Feb '06

________________________________ From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Sandeep Bantia Sent: Tuesday, February 07, 2006 7:23 PM To: boost-users(a)lists.boost.org Subject: [Boost-users] [Regex] : Unknown character sequence. Hi, I'm new to regex. This is basically what I was trying. #include <iostream> #include <boost/regex.hpp> int main() { std::string input = "sand._12"; boost::regex reg(".*\._[0-9]+"); If you're trying to escape the "." you probably also need to escape the "\" by using "\\". It is a C string after all.

1 0

Re: [Boost-users] Getting started with Boost: what did you need to know?
by Drumheller, Michael 07 Feb '06

07 Feb '06

>>> Date: Tue, 07 Feb 2006 09:12:31 -0500 >>> From: Deane Yang <deane_yang(a)yahoo.com> >>> Subject: Re: [Boost-users] Getting started with Boost: what did you need to know? >>> bjam *is* difficult to learn at first. For me one reason >>> was simply how alien it looks compared to an IDE or make. >>> And until recently the documentation was simply incomprehensible. >>> >>> Although I'm really looking forward to the new >>> documentation Dave is going to write, the existing >>> documentation of Boost Build v2 is really not too bad for >>> learning how to compile and link libraries and executables. I have shied away from boost.build v2 because it says on the opening page "an *onging project* to rewrite Boost.Build" at <<http://www.boost.org/tools/build/v2/>>--I always assumed it was not ready fro prime time. Sounds like you're saying that, from a doc point of view, it is a good resource for learning v1. I'll check it out--thanks. >>> And once you've got the hang of it, building software on >>> different platforms (I've done it on Windows, Linux, >>> Solaris, MacOSX) is incredibly easy and can be done using a >>> single Jamfile. Do you mean that you have adopted bjam as your day-to-day build tool? Just curious. Mike D.

3 3

[BGL] there is no in_edges() function for adjacency_matrix<>
by Dmitry Bufistov 07 Feb '06

07 Feb '06

> > > On Feb 7, 2006, at 7:14 AM, Dmitry Bufistov wrote: > > --- Dmitry Bufistov wrote: > > But what about > boost::adjacency_matrix<boost::directedS>? > > No, that too has to be > boost::adjacency_matrix<boost::bidirectionalS>. > > > I've attached an updated adjacency_matrix header (also in Boost CVS) > that implements the Bidirectional Graph requirements for > adjacency_matrix. It works with adjacency_matrix<directedS>; there is > no adjacency_matrix<bidirectionalS> because it isn't needed. > > 2 Doug, > Now I also can't use > > template <typename MutableGraph> > bool read_graphviz(std::istream& in, MutableGraph& graph, > dynamic_properties& dp, > std::string const& node_id = "node_id") ; > function for the same type. I think because of following > declaractions in "adjacency_matrix.hpp" > [unimplemented add_vertex] > Will it be also implemented or how I can read my graph? > > > Adding vertices to an adjacency_matrix is *extremely* expensive, > requiring reallocation of the entire data structure. That > functionality may be implemented at some point, but I won't be able to > do it any time soon. > > Doug > Doug, Thank you very much! Now my if_enabled() function compiles and works for boost::adjacency_matrix<>. I'm gonna define read_net() function for boost::adjacency_matrix<> just by first read graph to appropriate boost::adjacency_list<> and then construct matrix from list. Is this approach ok? Thank you once again, --dima

2 1