June 2010 - Boost - lists.boost.org

[xint] Sixth release, preliminary review again please
by Chad Nelson 29 Jun '10

29 Jun '10

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 The next iteration of the Extended Integer library (a arbitrary-length integer math library) is now available in the sandbox (at <https://svn.boost.org/svn/boost/sandbox/xint>) and the Vault (at <http://www.boostpro.com/vault/index.php?action=downloadfile&filename=xint.z…>). I think it's nearly ready for an official review. :-) The changes between this release and the last one are: * Refined the use of Boost.Parameter. Now only fixedlength takes a parameter; the other options are non-parameter types, which simplifies both their use and their implementation. * Minor redesign, taking advantage of proper policy-based design. nan() and operator~() are now only available on nothrow and fixedlength types, respectively, and is_nan() gets different definitions in the standard and nothrow types. * Added unsigned support, by request, with a variety of options for what to do when a calculation results in a negative number. * Added the difference function. * Added support for wide-character strings, by request. * Added the safe-bool idiom, by request. This release, like the last ones, includes a slightly older version of Ion Gaztañaga's preliminary Move library too. I haven't tested it with the current version, but it should work with that as well. The Move library is not used by default, and makes only a slight difference to the speed when thread-safety is enabled. Still on my immediate to-do list: * Add support for GCC's -fno-exceptions option. * Add more examples. - -- Chad Nelson Oak Circle Software, Inc. * * * -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkwcD6UACgkQp9x9jeZ9/wSfTwCfTMiucMP3uho9YmGJCyzMPe+s MhwAoMUe9tnFTHpQVDkCIfWiwpCjcuZi =QS39 -----END PGP SIGNATURE-----

4 15

Dynamic_bitset and access to underlying storage
by Iain K. Hanson 29 Jun '10

29 Jun '10

Hi, I need to use a bitset with contiguous storage so that I can pass it to a system call. Although I can't see it in the documentation, I know from reading the code, that dynamic bitset is implemented in terms of std::vector. Also, whilst I don't need the run time resizing that dynamic bitset provides, I do need to be able to use up to at least 64K bits, So, is there any chance of an additional member function+ that returns a pointer to the zero'eth element of the underlying vector? regards /ikh -- "I disapprove of what you say, but I will defend to the death your right to say it." Evelyn Beatrice Hall who wrote under the pseudonym S.G. Tallentyre in a biography of Voltaire. #include <std_disclaimer>

4 3

Re: [boost] [intro] Preview: Introspection library
by Andrzej Krzemienski 29 Jun '10

29 Jun '10

> I'd be interested in comments on an introspection library I've been > developing. > there has been no "Is there any interest..."-e-mail yet, as it was > started out of necessity in "namespace detail" of another library, so > you're also welcome to express interest/disinterest in a library like > this, without any specific comments. > source code: > https://svn.boost.org/trac/boost/browser/sandbox/intro/boost/intro/ > the library defines a concept that exposes the members of a class, for > introspection algorithms to work with. Hi, Is it not what Boost.Fusion does already? The macro BOOST_FUSION_ADAPT_STRUCT looks very similar to your mapping: http://www.boost.org/doc/libs/1_43_0/libs/fusion/doc/html/fusion/adapted/ad… Regards, &rzej

1 0

Re: [boost] [Boost-commit] svn:boost r63408 - trunk/boost/math
by Jürgen Hunold 29 Jun '10

29 Jun '10

On Monday, 28. June 2010 17:32:19 you wrote: > Author: johnmaddock > Date: 2010-06-28 11:32:19 EDT (Mon, 28 Jun 2010) > New Revision: 63408 > URL: http://svn.boost.org/trac/boost/changeset/63408 > > Log: > Make the compile time gcd facilities work with greatest integer types, and > be mpl compatible. This just broke gcc-4.2 and above. See gcc-4.2: http://tinyurl.com/28g49t8 http://preview.tinyurl.com/28g49t8 gcc-4.5: http://tinyurl.com/24jhn29 http://preview.tinyurl.com/24jhn29 for details. Yours, Jürgen -- * Dipl.-Math. Jürgen Hunold ! Ingenieurgesellschaft für * voice: ++49 511 262926 57 ! Verkehrs- und Eisenbahnwesen mbH * fax : ++49 511 262926 99 ! Lister Straße 15 * juergen.hunold(a)ivembh.de ! www.ivembh.de * * Geschäftsführer: ! Sitz des Unternehmens: Hannover * Prof. Dr.-Ing. Thomas Siefer ! Amtsgericht Hannover, HRB 56965 * PD Dr.-Ing. Alfons Radtke !

1 0

Any problems if I apply the patch in ticket #4377
by Marshall Clow 28 Jun '10

28 Jun '10

Looks ok to me.... https://svn.boost.org/trac/boost/ticket/4377 -- Marshall

2 1

[math] common_factor template parameters
by vicente.botet 28 Jun '10

28 Jun '10

Hi, static_gcc and static_lcm have unsigned long parameters. template < unsigned long Value1, unsigned long Value2 > struct boost::math::static_gcd; Could these be changed to uintmax_t? Is there any deep reason to don't define them for singed integers (intmax_t) also? Thanks, _____________________ Vicente Juan Botet Escribá http://viboes.blogspot.com/

4 8

[mini-review] Update of Boost.Assign (13th-19th of June)
by Thorsten Ottosen 28 Jun '10

28 Jun '10

Dear Boost Developers and Users, Erwann Rogard has implemented an update to Boost.Assign which we shall evaluate during the next week. His extensions allows one to write code like the following: (A) A list of references (to temporaries, in this example): boost::array<int,8> a = cref_csv( 1,5,3,4,2,9,0,7 ) (B) Chained ranges: vector<int> v = vector<int>( 3,1 ) && vector<int>( 3,2 ); If you have comments and suggestions about these new features, please submit a small review or voice your oppinion on the Boost developer list. If you have other feedback for Boost.Assign, now would be a good time to mention them too. If you submit a review, please follow the guidelines here: http://www.boost.org/community/reviews.html#Comments To access the new docs, please see https://svn.boost.org/svn/boost/sandbox/statistics/support/libs/assign/doc/… To access the new code, please see https://svn.boost.org/svn/boost/sandbox/statistics/support/boost/assign/ best regards Thorsten Ottosen, Review Manager

4 13

[typeof] 2 questions
by Eric Niebler 28 Jun '10

28 Jun '10

1) Some compilers don't support typeof in any form, even emulated. From looking at typeof.hpp, it looks like just #include'ing typeof.hpp on those compilers results in a hard error. I'd like a way to test whether typeof is supported and use it if so, otherwise do something else. How can I do that? 2) I can't find "decltype" by grepping the typeof code. Why doesn't typeof use the decltype keyword on compilers that support it? And I lied. I have another question. How expensive at compile-time is emulated typeof? TIA, -- Eric Niebler BoostPro Computing http://www.boostpro.com

4 7

[program_options] error reporting from custom validators
by Monty Taylor 28 Jun '10

28 Jun '10

Hi! I've been working on using notifier() to register a custom validator for an option. If the validator doesn't pass, it throws invalid_option_value. The issue I'm running in to though is that I do not have an context to report which option had a problem- I'm actually also really just abusing the notifier system as well, as I throw invalid_option_value with a descriptive message instead of just the option value. If I pass in an already bogus value, such as passing in a string to an int option, I get something like this: Error processing options: in option 'connect_timeout': invalid option value 'eggs' So it seems there are places where the code is validating options (seemingly just from any_cast) with context of which option is being validated. Is there some place I'm missing to hook a validator in where throwing an exception would properly pick up option name context? I saw this comment: /* We call validate so that if user provided a validator for class T, we use it even when parsing vector<T>. */ What I'd really love would be if I could call something like validator(some_validator).validator(some_other_validator), and that those would be passed some sort of context. Specifically, in the code we're working on replacing, it's quite easy to specify a min and max value for an integer option, and I'm trying to figure out a good way to keep this without requiring folks to write a completely independent notifier function for each option that emits its own error message. Any thoughts on where I might want to go with this?

1 0

[typeof] message.hpp and BOOST_TYPEOF_SILENT
by vicente.botet 27 Jun '10

27 Jun '10

Hi, I'm unable to find where message.hpp and BOOST_TYPEOF_SILENT are documented. Could this be added to the documentation is missing, and otherwise, could you point me where it is? Thanks, _____________________ Vicente Juan Botet Escribá http://viboes.blogspot.com/

2 1