January 2009 - Boost - lists.preview.boost.org

[swap] maintainer
by Daniel James 29 Jan '09

29 Jan '09

Hi, The maintainer for swap is listed as Joseph Gauterin in 'libs/maintainers.txt', but most of the recent changes seem to be by Niels Dekker. So is Joseph still the maintainer? As well as keeping the maintainers list up to date, I need to know so that I can request a trac component for the library. thanks, Daniel

3 2

[scope_exit] documentation
by Alexander Nasonov 29 Jan '09

29 Jan '09

Hi, Daniel James generated documentation for ScopeExit and commited it to trunk: http://svn.boost.org/svn/boost/trunk/libs/scope_exit/doc/html/index.html There are some differences compared to the reviewed version: * New syntax for captures. Some reviewers suggested (ref a)(cref b)(val c) syntax but I implemented (&a)(&b)(c). This syntax is very close to C++0x lambda syntax for captures: [&a, &b, c]. I can add support for const ref/val: (const& b)(const c) but I will not be able to support this syntax after a transition to C++0x. * tutorial has been updated to demonstate pass-by-value. * I've added C++0x to the Alternatives section. Any comments, suggestions etc are welcome. Thanks, Alex

1 0

[math] Multivariate root finding
by Tom Holden 29 Jan '09

29 Jan '09

Is anyone currently working on multivariate root finding in the math toolkit or elsewhere? Anyone know if this is planned? It would be good to have an alternative to GSL. Thanks in advance, Tom

4 4

[Fwd: Re: [format] please add operator String()]
by James Fowler 28 Jan '09

28 Jan '09

Thorsten Ottosen wrote: > ... > I my mind formatting is so closely tied to strings that I don't see > any problem in the implicit conversion. IIRC, String.Format in other > langauages return a String. Is some valid reason for require explicit conversion? It sure would be convenient for format to convert implicitly to string. If there is a valid reason, what about a wrapper class? I've used something close to the following for the last few years, and it hasn't bit me so far : class sformat : public boost::format { public: sformat( const std::string &s ) : boost::format( s ) {} operator std::string () { return boost::str( *this ); } template <class T> sformat & operator%( T v ) { *((boost::format*)this) % v; return *this; } }; - James -- ______________________________________________ James Fowler, Open Sea Consulting 770-714-3124 / james(a)openseaconsulting.com 1486 Surf Court, Marietta, GA, 30066 http://www.OpenSeaConsulting.com -- ______________________________________________ James Fowler, Open Sea Consulting 770-714-3124 / james(a)openseaconsulting.com 1486 Surf Court, Marietta, GA, 30066 http://www.OpenSeaConsulting.com

1 0

problem with building Boost version 1_34_1
by Oliver Kullmann 28 Jan '09

28 Jan '09

Hello, I have some "legacy code" to compile, using Boost version 1_34_1. The compiler is gcc 4.1.2 (new versions of Boost don't work anymore, and neither newer versions of gcc). It used to go through, but now on 64-bit platforms I get the following build error: In file included from ./boost/spirit/attribute/closure.hpp:24, from ./boost/wave/grammars/cpp_expression_grammar.hpp:16, from libs/wave/src/instantiate_cpp_exprgrammar.cpp:21: ./boost/spirit/phoenix/operators.hpp:355: error: ‘INT_MAX’ was not declared in this scope Now looking into the code that' correct (INT_MAX seems not be declared somewhere). (BTW, there are many other errors like that; that's just the first one.) So, apparently, somehow INT_MAX used to "slip in" by some uncontrolled inclusion, and now this is no longer the case. But what I don't understand: I always compiled it with the same local installation of gcc (4.1.2)? So I guess it is due to a newer version of glibc now installed? So I need to install first an old version of glibc (which?), then make gcc aware of it (?), and then build boost? I would be glad for some help (hints)! Oliver

2 1

Calling for 1.38 news items / library changes and additions.
by John Maddock 28 Jan '09

28 Jan '09

Just a quick reminder to everyone that since the 1.38 release is now in progress we're looking for items to go in the news/what's new section of the 1.38 release notes. Items should be added to the quickbook file website/public_html/beta/feed/history/boost_1_38_0.qbk, then cd into website/public_html/beta and run bjam. If there are no errors then commit (should be 3 changed files) and check the result looks OK at http://beta.boost.org/users/news/version_1_38_0. If there are any problems then please post the content in reply to this message. Regards, John Maddock.

3 3

Two minor (?) legal-related questions (empty lines and "Boost sponsorship")
by Gennaro Prota 28 Jan '09

28 Jan '09

Hi guys, the "how to apply to source files" entry at <http://www.boost.org/users/license.html> asks to leave an empty line before and after the comment block which contains copyright notice and license reference. I'm wondering if that was an advice from the lawyers who wrote the license or just a stylistic preference. Also, I just saw a BoostSponsorshipAgreement.pdf in the more/ subdirectory. What is it? -- Genny

5 4

[1.38.0] beta release candidate
by Beman Dawes 28 Jan '09

28 Jan '09

[1.38.0] beta release candidate packages are available at http://boost.cowic.de/rc/ I'm doing a library build right now on Windows/VC++ 9.0. I'd appreciate it if someone could try a library build on a Linux system, and report any problems. Although we won't hold the beta for minor tweaks, I would like to be sure that docs are present and libraries build before pushing this out to SourceForge. Thanks, --Beman

7 14

[serialization] serializing tr1::unordered_map
by Kenny Riddile 28 Jan '09

28 Jan '09

Just wondering if a tr1::unordered_map can be serialized yet, whether it be via the released version of the library or via some extension. I remember some discussion about it back in 2007, but I didn't have a need for it then, so I didn't follow the discussion. Thanks, Kenny Riddile

2 1

[regex] [1.35]Problem with IPV6 validation windows XP
by Gopal_Haldar＠McAfee.com 28 Jan '09

28 Jan '09

I am using same expression for .net application and javascript it is working fine but when I used for boost regular expression the following problem is facing But it is failed to validate foramt like 2001:470:1f01:115::1 or 2001::11 or 2001:1::11 Please see my code... bool IsValidIPv6String(wstring ipv6address) { wsmatch what; wregex expression(L"^((([0-9A-Fa-f]{1,4}:){7}[0-9A-Fa-f]{1,4})|(([0-9A-Fa-f]{1, 4}:){6}:[0-9A-Fa-f]{1,4})|(([0-9A-Fa-f]{1,4}:){5}:([0-9A-Fa-f]{1,4}:)?[0 -9A-Fa-f]{1,4})|(([0-9A-Fa-f]{1,4}:){4}:([0-9A-Fa-f]{1,4}:){0,2}[0-9A-Fa -f]{1,4})|(([0-9A-Fa-f]{1,4}:){3}:([0-9A-Fa-f]{1,4}:){0,3}[0-9A-Fa-f]{1, 4})|(([0-9A-Fa-f]{1,4}:){2}:([0-9A-Fa-f]{1,4}:){0,4}[0-9A-Fa-f]{1,4})|(( [0-9A-Fa-f]{1,4}:){6}((\\b((25[0-5])|(1\\d{2})|(2[0-4]\\d)|(\\d{1,2}))\\ b)\\.){3}(\\b((25[0-5])|(1\\d{2})|(2[0-4]\\d)|(\\d{1,2}))\\b))|(([0-9A-F a-f]{1,4}:){0,5}:((\\b((25[0-5])|(1\\d{2})|(2[0-4]\\d)|(\\d{1,2}))\\b)\\ .){3}(\\b((25[0-5])|(1\\d{2})|(2[0-4]\\d)|(\\d{1,2}))\\b))|(::([0-9A-Fa- f]{1,4}:){0,5}((\\b((25[0-5])|(1\\d{2})|(2[0-4]\\d)|(\\d{1,2}))\\b)\\.){ 3}(\\b((25[0-5])|(1\\d{2})|(2[0-4]\\d)|(\\d{1,2}))\\b))|(([0-9A-Fa-f]{1, 4}:){0,5}::([0-9A-Fa-f]{1,4}:){0,5}[0-9A-Fa-f]{1,4})|(::([0-9A-Fa-f]{1,4 }:){0,6}[0-9A-Fa-f]{1,4})|(([0-9A-Fa-f]{1,4}:){1,7}:))$"); Please look into the expression. Regards Gopal

1 0