October 2005 - Boost-users - lists.boost.org

design question for boost::program_options and dynamic options
by boost＠nogga.de 02 Nov '05

02 Nov '05

Hello, I'm new to the boost::program_options library and tried to use it for a command pattern based software. The program is launched with the following commandline: > program command [Options] <Files> The program can use different output formatters with --formatter={XML|binary|text}. Every command and formatter specify their own set of options. I first tried to collect all options form the commands and formatters and parse the command line only once. This lead to the problem, that due to the different positional options information the commandline parser always threw an exception due to wrong command line formatting. Since I couldn't find a way to add mutliple optional option groups to the parser, I moved the parsing code into my command base class and query the derived commands for the different required information, like: class Command { public: vritual options_description GetOptionsDescription (); virtual options_description GetHiddenOptionsDescription (); virtual positional_program_options GetPositionalProgramOptions (); virtual Run (variables_map const & vm) = 0; int Execute (std::vector<std:.string> args) { options_description desc; descr.add (GetOptionsDescription ()).add (GetHiddenOptionsDescription ()); variables_map vm; store(command_line_parser(args) .options(desc). .positional(GetPositionalProgramOptions ()) .run(), vm); notify(vm); Run (vm); } } The execute function is called form the program main after the "command" has been stripped from the commandline and the concrete command was created by the factory. So far so good. This works for the different commands, but not for the formatters. How can I parse the command line dynamically? E.g. if I encounter the "--formatter=xml" argument, a new XMLFormatter is created and the options description from this formatter are added to list of valid options? I know, that there is the notify mechanism, but this only works after the commandline has been successfully parsed. And also the custom parsers do not solve my problem. Any idea how to solve this issue? Best regards Dirk

2 2

configuring a library to work with multiple versions of boost
by Ross Boylan 01 Nov '05

01 Nov '05

I have a library that uses boost and also uses a pre-release version of the ptr_container (header only) library that eventually made it into boost 1.33 (I think). I would like people to be able to compile my code with various versions of boost. I've had it working with 1.31 and 1.32, and expect to have it working with 1.33 when that reaches my distribution. I have a subdirectory with the pre-release ptr_container library and some supporting code from elsewhere in boost. This is in a "boost" subdirectory of my project, which has included the directory above that unconditionally in my compilations. With 1.33 I think I want to ignore my local stuff completely. I'm using autoconf and libtool. Are there autoconf macros available for detecting the boost libraries? Do people have any suggestions about the best way to proceed? One possibility that occurs to me is to test for the version of the library at configure time, and make the include path depend on the results. Thanks. Ross Boylan

2 2

ptr_map
by Oleg Smolsky 01 Nov '05

01 Nov '05

Hello Thorsten, thanks for the ptr_vector clarifications. I have another question about ptr_map: I can declare a map iterator in the following way: ptr_map<string, SomeClass>::iterator i = m.begin(); Now, I'd expect to be able to compile the following, as with a normal map: string s = i->first; i->second.SomeMethod(); ...assuming SomeClass::SomeMethod() exists and is declared as public. However, "first" and "second" don't seem to exist in ptr_map. Could you shed some light on how I can iterate your map please? Best regards, Oleg.

2 3

Visual C++ 2005 (8.0) support
by Ovanes Markarian 01 Nov '05

01 Nov '05

Hi all, did someone test boost 1.33 with Visual C++ 2005(8.0)? I am wondering how I can compile the library using jam. Should I use VC++ 7.1 option? Another question. I derived an iterator from the iterator_facade template. As I have recently found out VC++ 8.0 did not support friend template classes, e.g. template<class ....> class my_type { template <class> friend class my_type; ... }; This will not work in VC++ 8.0. Did you know about this issue? Should this be filed as a bug at MS? With Kind Regards, Ovanes Markarian

5 5

How to build Boost for Cygwin (asked the noob)
by Olumide 01 Nov '05

01 Nov '05

Hi - Would someone please give me some (SIMPLE) pointers on how to build Boost for Cygwin. Thanks, - Olumide

2 1

[threads] read_write_mutex and condtion
by Phil Endecott 01 Nov '05

01 Nov '05

Dear All, Is it possible to use a boost::condition in combination with a read_write_mutex? I've tried but failed. This is with 1.32.0. My test code follows. As you can see it can use either a mutex or a read_write_mutex depending on which bits are commented-out. It works as expected with a mutex, but with a read_write_mutex I get a compiler error, shown below. Perhaps I am missing some fundamental reason why conditions can't work with read_write_mutexes. Can someone explain? Thanks, Phil. #include <string> #include <iostream> #include <boost/thread/mutex.hpp> #include <boost/thread/read_write_mutex.hpp> #include <boost/thread/condition.hpp> #include <boost/thread/thread.hpp> #include <boost/bind.hpp> #include <unistd.h> using namespace std; using namespace boost; struct shared_thing { string data; //typedef mutex mx_t; //typedef mutex::scoped_lock read_lock; //typedef mutex::scoped_lock write_lock; typedef read_write_mutex mx_t; typedef read_write_mutex::scoped_read_lock read_lock; typedef read_write_mutex::scoped_write_lock write_lock; mx_t mx; condition cn; shared_thing::shared_thing(void): //mx() mx(read_write_scheduling_policy::writer_priority) {} void writer(void) { while(1) { sleep(1); { write_lock l(mx); data += "."; cn.notify_all(); } } } void reader(void) { while(1) { read_lock l(mx); cn.wait(l); cout << data << endl; } } }; int main(int argc, char* argv[]) { shared_thing t; thread wr_thread(bind(&shared_thing::writer,&t)); t.reader(); } $ g++ -pthread -O -Wall -o condvar condvar.cc -lboost_thread /usr/include/boost/thread/detail/read_write_lock.hpp: In member function `void boost::condition::wait(L&) [with L = boost::detail::thread::scoped_read_lock<boost::read_write_mutex>]': condvar.cc:51: instantiated from here /usr/include/boost/thread/detail/read_write_lock.hpp:297: error: ` boost::read_write_mutex&boost::detail::thread::scoped_read_lock<boost::read_write_mutex>::m_mutex ' is private /usr/include/boost/thread/condition.hpp:92: error: within this context

3 2

Re: [Boost-users] [Multi-Index] composite_key problems
by JOAQUIN LOPEZ MU?Z 31 Oct '05

31 Oct '05

----- Mensaje original ----- De: Andreas Volz <lists(a)brachttal.net> Fecha: Lunes, Octubre 31, 2005 7:27 pm Asunto: Re: [Boost-users] [Multi-Index] composite_key problems > The questions seems to by stupic, but how > could I > access the data looked up by B alone? What should I assign to the > left? > > ??? = es.get<1>().find(b); > > An iterator or what else? This returns an iterator, yep. Please notice that es.get<1>() returns a reference to an index, which is, in many respects, entirely similar to a regular std::set. So you have find(), equal_range() etc. just like in the STL case, with equivalent syntax and semantics. You might want to give the Boost.MultiIndex tutorial a further read --anyway, don't hesitate to come back here if things aren't still clearer. Joaquin M Lopez Muñoz Telefonica, Investigacion y Desarrollo

2 1

Re: [Boost-users] Translating an absolute path to a path relative
by Gottlob Frege 31 Oct '05

31 Oct '05

> > Date: Mon, 31 Oct 2005 09:55:57 -0300 > From: Hermann Rodrigues <hermann.rodrigues(a)gmail.com> > Subject: [Boost-users] Translating an absolute path to a path relative > to the current directory > To: boost-users(a)lists.boost.org > Message-ID: > <497dfc460510310455j1bd56615peffd2f6040fe849e(a)mail.gmail.com> > Content-Type: text/plain; charset=ISO-8859-1 > > Hi, > > I am running into the following problem these days: > > I need to translate an absolute path to a path relative to the current > directory on Windows. The translation process need to include all > required ".." directory specifiers in the path to make sure it can be > accessed from the current directory. > > Have you tried the Windows functions such as PathRelativePathTo, and its compliment PathCombine?

2 1

[BUG] missing references in docs on web site (threads)
by Ross Boylan 31 Oct '05

31 Oct '05

While reading the web pages for the threads library I noticed some missing references. I don't know if they reflect a problem with the documentation source or a problem with the processing converting the source into html (if there was any such processing). I did check the html source of teh web pages to verify that it wasn't a display problem with my browser. http://boost.org/doc/html/threads.html#id1357380 In "Potentially non-thread-safe functions" this appears: "It is possible to write thread-safe implementations of these by using thread specific storage (see boost::thread_specific_ptr), and several C++ compiler vendors do just that. The technique is well-know and is explained in ." The reference at the end is missing. I thought I saw another example like this in the threads doc, but can't find it now. Also, the quote from Brinch Hansen at the end of http://boost.org/doc/html/threads/rationale.html doesn't indicate the source (though one could at least guess the article from the bibliography, though he's co-author of another one in the bibliography).

1 0

Re: [Boost-users] [Multi-Index] composite_key problems
by Andreas Volz 31 Oct '05

31 Oct '05

Am Mon, 31 Oct 2005 18:14:00 +0100 schrieb JOAQUIN LOPEZ MU?Z: > // lookup by (A,B); > > es.find(make_tuple(a,b)); > > // lookup by A alone > > es.find(make_tuple(a)); > > // lookup by B alone > // note get<1> retrieves the second index. > > es.get<1>().find(b); > > Is the issue clearer now? If not, please tell me > so and I'll try to explain myself a bit further. A little clearer, but not at all. I'm really new to Boost, so I doesn't understand all. The questions seems to by stupic, but how could I access the data looked up by B alone? What should I assign to the left? > ??? = es.get<1>().find(b); An iterator or what else? There was a print_out_by_name function in docs that I tried to adopt, but I failed. And the compile errors are not very helpfull. As normal with template programming :-( regards Andreas

1 0