- Boost - lists.boost.org

[SoC][Coroutine] Coroutine design
by Giovanni P. Deretta 06 May '06

06 May '06

Hello boosters, As some of you know, I've been working on a coroutine library. As some have expressed interest in it, I've decided to participate to the google SoC, as a student, with this project (and boost as a mentor of course :)) I've not yet formally applied because I wanted first to write a detailed description of the project. After four sleepless nights I'm finally (mostly) done and ready to accept critics (let's hope no flames!). You can find the design at http://libstream.sf.net/index_c.html . Under the boost vault, in the concurrency folder there is already my first attempt at this problem (the package is continuation.tar.gz). It could be useful to evaluate the earlier attempt at the solution, that has driven me to write a more consistent design. Be warned that the code in that package does not compile (and is unix only). I will submit the project to Google later this week.. So, anyone interested? -- Giovanni P. Deretta

4 8

extending type_traits::rank ?
by Toon Knapen 06 May '06

06 May '06

I need an equivalent of type_traits::rank for a vector-matrix library I am working on so I thought about extending type_traits::rank for my own vector- and matrix-types. But is extending type_traits foreseen or even possible? And what is the best way to go about? The easiest way of course is to add a specialization for boost::type_traits::rank in my project's sources but this is not a very clean solution because this would require that I add stuff to the boost-namespace in my own project. Ideas? Thanks in advance, toon

4 4

Concept for function object with result_type?
by David Abrahams 06 May '06

06 May '06

I know we have AdaptableUnaryFunction, which covers function objects with nested result_type and argument_type. But we've long since learned that argument_type is really unnecessary, and plenty of libraries are looking at result_type alone (e.g. bind). Do we have a concept for that sort of function object? If not, what shall we call it? Should we rather invent a concept for "UnaryFunction that works with boost/tr1::result_of<>" ? Maybe DeducibleUnaryFunction? -- Dave Abrahams Boost Consulting www.boost-consulting.com

1 0

[type_traits] extent doc example bug
by Kim Barrett 06 May '06

06 May '06

In boost 1.33.1 (and in the documentation available from the web-site, if that is more recent?), the first example for the extent trait says extent<int[1]> inherits from integral_constant<std::size_t, 0>. That final "0" should be "1".

2 1

A lost Sandbox file?
by Daryle Walker 05 May '06

05 May '06

For the past few months, I always get this message: //======================================================================== cvs update: nothing known about boost-sandbox/libs/fields_visitor/doc/boostbook.css //======================================================================== whenever I update my CVS. I checked that directory and the CSS file doesn't exist (or, at least, not anymore). Is there some sort of cache that needs clearing? -- Daryle Walker Mac, Internet, and Video Game Junkie darylew AT hotmail DOT com

1 0

[Boost-bugs] [ boost-Bugs-1482703 ] time_from_string silently misreads date
by SourceForge.net 05 May '06

05 May '06

Bugs item #1482703, was opened at 2006-05-05 12:21 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1482703&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: date_time Group: None Status: Open Resolution: None Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Jeff Garland (az_sw_dude) Summary: time_from_string silently misreads date Initial Comment: The following code snippet: boost::posix_time::ptime t3 = boost::posix_time::time_from_string ("2006-03-02"); cout << " t3 = " << boost::posix_time::to_simple_string(t3) << endl; Gives: t3 = 2006-May-24 14:03:02 Admittedly, the date-time string is misformatted, as it doesn't have the HH-MM-SS fields, but silently giving the wrong dete is likely to be a very serious problem to users. For the record, let me say that date_time and boost is general are really very fine work which I greatly appreciate. -Mike Albert MAlbert(a)princeton.com ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1482703&group_… ------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

SoC idea: tool that converts existing docs to quick-book
by Thorsten Ottosen 05 May '06

05 May '06

Hi, I guess the hedline says it all. In our effort to make the documentation more modern and similar, such a tool would be great IMO. -Thorsten

1 0

Re: [boost] [thread] 1.33.1 with vc71, links against msvcrt and msvcprt
by Dave McCabe 05 May '06

05 May '06

Roland Schwarz: > Don't know what you think is wrong on the above. These two dumpbin > indicate correct behaviour. Yes, they're absolutely correct, which is what I mean... Andrew had a -s version which linked against MSVCRT, which is the DLL runtime (not the single-threaded version, which Anthony thought). David ________________________________________________________________________ This e-mail, and any attachment, is confidential. If you have received it in error, do not use or disclose the information in any way, notify me immediately, and please delete it from your system. ________________________________________________________________________

2 1

Re: [boost] [thread] 1.33.1 with vc71, links against msvcrt and msvcprt
by Dave McCabe 05 May '06

05 May '06

Anthony Williams: > Any build of the library that requires MSVCPRT and MSVCRT is just > broken, since they are the single-threaded runtime, and their use > indicates that the library was built without MT support. I'm surprised > the library manages to build at all without MT support, since the CRT > thread functions, such as beginthread, aren't available unless MT > support is enabled. MSVCRT and MSVCPRT are the DLL runtimes, not the single-threaded ones; you're thinking of LIBC and LIBCP. I have STLport versions, but: C:\Boost\lib>dumpbin /directives libboost_thread-vc71-mt-sp-1_33_1.lib | findstr DEFAULTLIB | sort | uniq /DEFAULTLIB:"LIBCMT" /DEFAULTLIB:"libcpmt" /DEFAULTLIB:"OLDNAMES" /DEFAULTLIB:"stlport_vc71_static.lib" /DEFAULTLIB:"uuid.lib" C:\Boost\lib>dumpbin /directives libboost_thread-vc71-mt-p-1_33_1.lib | findstr DEFAULTLIB | sort | uniq /DEFAULTLIB:"msvcprt" /DEFAULTLIB:"MSVCRT" /DEFAULTLIB:"OLDNAMES" /DEFAULTLIB:"stlport_vc71.lib" /DEFAULTLIB:"uuid.lib" Looks like someone has managed to build the static version against DLL runtimes, or just done some creative renaming... -- David McCabe ________________________________________________________________________ This e-mail, and any attachment, is confidential. If you have received it in error, do not use or disclose the information in any way, notify me immediately, and please delete it from your system. ________________________________________________________________________

2 1

[mpl] preprocessed sources
by Michael Fink 05 May '06

05 May '06

Hi, I'm working around some issues in MPL with the evc4 (eMbedded Visual C++ 4) compiler. The compiler behaves just as a msvc6 one and includes files from "boost\mpl\aux_\preprocessed\msvc60\". In or_impl.hpp (and the same problem in and_impl.hpp) there is a defined for a templated "struct or_impl" and a specialisation for or_impl<false>::result_ after the struct. In evc4 the specialisation must be inside the class. The ..\msvc70\or.hpp implementation does this right for evc4. Now my question is how I should port the preprocessed sourcecode for evc4? Creating preprocessed files for evc4 somehow? Using the msvc70 files, given that it would work? Not using preprocessed files? I'm not familiar with this Boost library, I'm just porting it (as it gets used by many other libraries). bye Michael

1 0