- Boost - lists.boost.org

Re: [boost] Re: 1.32 documentation issue
by JOAQUIN LOPEZ MU?Z 06 Nov '04

06 Nov '04

----- Mensaje original ----- De: Beman Dawes <bdawes(a)acm.org> Fecha: Martes, Noviembre 2, 2004 9:06 pm Asunto: Re: [boost] Re: 1.32 documentation issue > At 12:39 PM 11/2/2004, Robert Ramey wrote: > > >Actually after poking around in other pages - specifically the > spirit >documentation I find that its not uncommon for links to > point to > >www.boost.org rather than have a relative path that would keep > the > session > >in the same documentation tree. This would seem to conflict > with the > idea > >that boost should be usable as a CDROM distribution. > > > >I don't know if there's time for this but the inspect should be > enhanced > to > >check for this and all the offending global links should be > altered to > >relative ones. > > We tried to do that at one time, but gave up because it is a bit > tricky. > Some absolute www.boost.org references are correct, such as to the > LICENSE > file, so it isn't as simple as just searching for "www.boost.org/" > in .htm? > files. > > I've just taken a look (using grep) at the CVS RC (not the > tarball) and it > seems that while the problem does happen, particularly in spirit > and graph > libraries, it isn't really horrible. > > I'd say we can wait until after the release ships, then see if we > can come > up with some logic that doesn't generate a lot of false positives. > Automatic detection aside, would it be possible to fix the offending links manually before release? Maybe it's not that straightforward, if those links are generated from BoostBook source. Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 2

[1.32 release] Runaway compile process in regression tests
by Zak Kipling 06 Nov '04

06 Nov '04

I'm running the regression tests for the 1.32 draft tarballs -- no problems on x86/debian, but on an amd64/gentoo (gcc 3.3.4 prerelease) machine it still hasn't completed after nearly two days, apparently due to a runaway cc1plus process, compiling numeric/conversion/test/converter_test.cpp. This appears to be due to a bug in gcc -- see http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18300. I should have a patch to gcc tested soon; I don't think it's a problem with Boost as such, although there might be a workaround for current compilers (I don't think I understand the numeric converters library well enough to be sure). Zak.

2 1

Compile fix for Boost.Test for aCC on HP-UX
by Ferdinand Prantl 05 Nov '04

05 Nov '04

Hello, you can find a small fix for boost/test/detail/basic_cstring/basic_cstring.hpp attached and a dirty workaround for a compiler bug in aCC for boost/test/unit_test_suite.hpp. As of the latter; the compiler seems to get confused by equal names in different namespaces and classes; renaming Impl helps or making the previous structure friend helps too (this I chose to minimize the change). Ferda HP-UX B.11.11 U 9000/800 aCC: HP ANSI C++ B3910B A.03.37 <<basic_cstring.hpp.diff>>

2 1

[Config] proposal of new macro, BOOST_DECL :follow up for boost::noncopyable & vc7.1 & dll
by Francis ANDRE 05 Nov '04

05 Nov '04

Hi Developers When using the boost.thread package as a dll under vc7.1, one gets for each reference of boost::mutex the C4275 warning for the inherited noncopyable class (in fact 5 times, one for mutex, one for timed_mutex, one for try_mutex, one for thread, and one for thread_group!!) c:\boost\include\boost-1_32\boost\thread\mutex.hpp(34) : warning C4275: non - DLL-interface 'boost::noncopyable_::noncopyable' used as base for DLL-interface 'boost::mutex' In order to avoid such warning, it would be nice to redefine -- again(sic) -- the BOOST_DECL macro and put it as a decorator for all dependent classes used by the boost libraries that could be generated as a dll.(thread, regex, filesystem, and so on). With the empty string as default value in boost/config.hpp, it would the responsability of each boost libraries to setup BOOST_DECL to the correct value when using it as a dll. Cheers Francis ANDRE

5 7

GUI Library Proposal for a Proposal
by Brian Braatz 05 Nov '04

05 Nov '04

(was RE: [boost] Re: GUI library - something to think about) Both http://www.torjo.com/win32gui/save_dlg.html and http://www.litwindow.com/Library/Documentation/documentation.html are extremely interesting. What I see (just after taking a peek at both): Win32Gui Interesting design No macros Looks tied to Win32Gui LitWindow Also interesting design Does have macros Also sits on top of WxWidgets, but it looks like the WxWidgets layer could be abstracted out. I propose that BOTH authors collaborate (on this list) to achieve the following: * Combination of the best of both frameworks * Can sit on WxWidgets OR can have a win32 and xwindows underlying implementation * This would allow it to be in boost, but also be LEAN and not need another framework I believe that if you took both of your thoughts, and accomplished the above bullet points, not only would it be good for boost, but it would really mean that boost had something that was much cooler than what I see available from other (ahem) vendors. I.e. instead of making a boost library that just "provides a layer" over GUI things, this combination would be actually moving the technology forward. (they don't call it "boost" for nothing :) ) I ask that both of you gentlemen seriously consider my proposal. I would also offer to help in any way that I can as I think you both have some really groundbreaking ideas in the realm of GUI frameworks. (fingers crossed) Brian :) > -----Original Message----- > From: boost-bounces(a)lists.boost.org [mailto:boost-bounces@lists.boost.org] > On Behalf Of John Torjo > Sent: Wednesday, November 03, 2004 9:14 AM > To: boost(a)lists.boost.org > Subject: Re: [boost] Re: GUI library - something to think about > > > struct SomeSettings { > > int m_aMember; > > SomeOtherStruct m_struct; > > vector<string> m_strings; > > }; > > > > the neccessary data adapter definition which you must include in a > > source file looks like this > > > > IMPLEMENT_CONTAINER_ADAPTER(vector<string>) > > > > BEGIN_DATA_ADAPTER(SomeSettings) > > PROP(m_aMember) > > PROP(m_struct) > > PROP(m_strings) > > END_DATA_ADAPTER() > > > > > Sure thing, but how about *true* bindings? > http://www.torjo.com/win32gui/save_dlg.html > > (if you download the code, you'll find quite a few examples: > examples/smart_dialogs) > > Note that you can add correspondences directly to your data + validation > code (no macros whatsoever). > >

10 11

RE: [boost] Re: GUI Library Proposal for a Proposal
by Brian Braatz 05 Nov '04

05 Nov '04

> -----Original Message----- > From: boost-bounces(a)lists.boost.org [mailto:boost-bounces@lists.boost.org] > On Behalf Of Jared McIntyre > Sent: Friday, November 05, 2004 7:28 AM > To: boost(a)lists.boost.org > Subject: [boost] Re: GUI Library Proposal for a Proposal > > >Edward Diener wrote: > >> > >>No !!!!! WxWidgets has almost nothing to do with modern C++. Let's no go > >>there, or have anything to do with supporting any other GUI > implementation. > >> > > > >Also, wxWindows is a huge library. If we were to introduce > >something in boost, I'd very much like it to be as small and light > >weight (as win32gui is supposed to be), not a collection of > >multi-megabyte dlls required, just to wrap win32 or what ever the > >underlying library is. > > > >If people want more powerful components, they can be added > >afterwards, but the underlying library around core OS controls > >should be as small as possible. > > > >Thanks > > > >Russell > > Honestly, I don't buy into the idea of cross platform GUI at all. I > have lots of reasons born out of experience with them that I won't > get into here, but in general, I find they are nice for the > developer, but poor for the user. I would much rather write my > backend code in a cross-platform language, and hook it up to a GUI > designed for each platform I support. That is why, at first glance > anyway, I like the idea behind the Lit Window design. Most of the > backend is GUI neutral, allowing you to do more work in your > cross-platform code. In theory, this could be used to simplify the > platform specific GUI code that needs to be written. I'd much rather > see C++ go a route of making it easier to develop for different GUI > libraries than to actually develop one of its own (since I don't > think it would get much use anyway). > > JJared McIntyre > _______________________________________________ > Unsubscribe & other changes: > http://lists.boost.org/mailman/listinfo.cgi/boost [Brian Braatz] " I'd much rather > see C++ go a route of making it easier to develop for different GUI > libraries than to actually develop one of its own (since I don't > think it would get much use anyway)." Yes, and that is what my original mail that started this thread suggested :) Where it gets confusing to talk about, is that by doing that, you end up doing a "little bit" of the work that "already exists" elsewhere. That is ok.

1 0

[serialization] problem with boost/serialization/is_abstract.hpp
by Joaquín Mª López Muñoz 05 Nov '04

05 Nov '04

In MSVC 6.0, the following snippet fails: #include <boost/archive/text_iarchive.hpp> #include <boost/tuple/tuple.hpp> int main() { return 0; } complaining about is_abstract being already defined. The problem lies in the following lines inside boost/serialization/is_abstract.hpp: #ifndef BOOST_TT_IS_ABSTRACT_CLASS_HPP #if defined(__GNUC__) && (34 <= _GNUC__ * 10 + __GNU_MINOR) \ || defined(__MSVC_VER) && (1310 <= __MSVC_VER) \ || defined(__EDG_VERSION__) \ /**/ #include <boost/type_traits/is_abstract.hpp> #else // default to false if not supported namespace boost { template<class T> struct is_abstract { typedef mpl::bool_<false> type; BOOST_STATIC_CONSTANT(bool, value = is_abstract::type::value); }; } // namespace boost #endif #endif // BOOST_TT_IS_ABSTRACT_CLASS_HPP The code is *incorrectly* using __MSVC_VER where it should use _MSVC_VER (just one leading underscore). Thus boost::is_abstract is defined twice, as boost/type_traits.hpp is implicitly included by boost/tupe/tuple.hpp If nobody complains, I can fix this and commit it to the release and main branch. What worries me, though, is that the code above seems flawed to me (apart from the minor typo). What happens if the compiler is not one of those taken care of explicitly in boost/serialization/is_abstract.hpp? Could anyone please test the first snippet in some compiler other than MSVC, ICC for Windows, GCC or an EDG-based compiler? My thesis is that it will fail. Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

7 7

Serialization support for mersenne_twister
by Neal D. Becker 05 Nov '04

05 Nov '04

I know I've mentioned this before. I'd really like to see this tiny bit of code added to mersenne_twister: template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & x; ar & i; } This would make it very easy to use our new serialization library with mersenne_twister. It costs nothing if you don't use it. If you are really paranoid, it could be surrounded by an #ifdef.

3 2

[spirit] select_p problem
by Vyacheslav E. Andrejev 05 Nov '04

05 Nov '04

Hi All, boost::spirit::select_p parser can not be compiled with boost::spirit::rule as argument. The problem is that metafunction boost::spirit::impl::as_embedded_parser<ParserT> returns const reference to boost::spirit::rule, not boost::spirit::rule itself, when PaerserT = boost::spirit::rule. So trying to compile parser_result<as_embedded_parser<rule<> > > deep inside the select_p implementation generates error message that const rule<>& has no result_type. I suggest to change parser_result definition as following: template <typename ParserT, typename ScannerT> struct parser_result { typedef typename boost::remove_reference<ParserT>::type::template result<ScannerT>::type type; }; -- ______________________________ Vyacheslav E. Andrejev System Architect, Excelsior, LLC

2 1

Contacting Chuck Allison [if you don't know how to contact Chuck Allison ignore]
by Vesa Karvonen 04 Nov '04

04 Nov '04

Hi, I would like to query whether The C++ Source (http://www.artima.com/cppsource/index.jsp) would be interested in publishing an article on Lazy Template Metaprogramming. I tried e-mailing the ChuckAllison(a)mailblocks.com -address, but haven't received any response in about 24-hours. So, if you know how to contact Chuck Allison, please forward this message to him. Regards, Vesa Karvonen _________________________________________________________________ Don't just search. Find. Check out the new MSN Search! http://search.msn.com/

2 1