- Boost - lists.boost.org

[iostreams] Bugs in iostreams
by Chad Walters 05 Jun '07

05 Jun '07

Hey folks, Not really sure what the right procedure is here, but I posted two bugs with patches in iostreams to the ticket system last week: http://svn.boost.org/trac/boost/ticket/1002 http://svn.boost.org/trac/boost/ticket/1003 One of them is a simple typo fix in some typedefs and the other is a clear issue where the close() code does not perform the documented behavior. I understand that the regular maintainer of iostreams is otherwise occupied. Is there anyone else who can look over the bugs and either commit them or give me feedback? Thanks for any assistance or info (eg: pointers on proper procedure, expected turnaround, etc). Chad Walters

1 0

[interprocess] container requirements for B.IP allocators
by Joaquín Mª López Muñoz 05 Jun '07

05 Jun '07

In the section entitled "Container requirements for Boost.Interprocess allocators" from Boost.Interprocess documentation (http://tinyurl.com/2bruhb ), it is stated that containers compatible with B.IP allocators must honor the following conditions: 1. STL containers may not assume that memory allocated with an allocator can be deallocated with other allocators of the same type [...] 2. Containers' internal pointers should be of the type allocator::pointer and containers may not assume allocator::pointer is a raw pointer. 3. All objects must be constructed-destroyed via allocator::construct and allocator::destroy functions. I see two problems with requirement 3: a. Chances are that allocator member functions construct, destroy and address be deprecated in C++0x, if the N2257 paper by Matt Austern (http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2257.html ) is taken into consideration. b. I've casually inspected the code of B.IP and seems like the allocator construct/destroy interface is *not* actually used, see for instance list_alloc in boost/interprocess/containers/list.hpp So, if condition 3 is not used, I wonder what the actual requirements are with respect to allocator::pointer. I guess the main assumption B.IP containers work on is that allocator::pointer is implicitly convertible to/from T*. Also, what the requirements are when handling (nonraw) pointers to types other than T? Consider the following: struct T { X x; }; If I've got an allocator::pointer pt to an object t of type T and want to store a nonraw pointer to x, how should I do? I guess something like: allocator::rebind<X>::other::pointer px=pt->x; but this is just a guess from my part, as it's not documented. Some clarifications on these issues, and maybe a rewriting of the relevant doc sections, would be welcome for those container writers willing to become B.IP allocator compatible. Thank you! Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

4 6

[function] allocator template parameter question
by Emil Dotchevski 05 Jun '07

05 Jun '07

One important advantage of using boost::function is that it acts like a function pointer, reducing physical coupling between compilation units. This is very much like shared_ptr. A nice feature of shared_ptr is that it has a single template parameter, T, even though diferent instances of shared_ptr<T> can use different allocators. When using boost::function, one can also provide an allocator, but unlike shared_ptr, the allocator is a default parameter of the boost::function class template. Is there a reason why this is necessary? Can't boost::function use similar technique to the one employed by shared_ptr to avoid the need for this second template parameter? Emil Dotchevski

4 4

[bimap] New Version
by Matias Capeletto 05 Jun '07

05 Jun '07

Hello boost, I have released a new version of the Boost.Bimap. The main attraction of this new release is the possibility to add information about each relation. I have been working very hard to ensure that Bimap works well with the others boost libraries. I have added new sections to the "Bimap and Boost" docs to show how to combine them. For example, from Bimap and Boost.Foreach: ------------------------------------------------------------------------------------------------------ typedef bimap< int, string > bm_type; bm_type bm; assign::insert( bm ) (1,"one") (2,"two") (3,"three"); BOOST_FOREACH( bm_type::right_const_reference p, bm.right ) { std::cout << p.first << " ---> " << p.second << std::endl; } ------------------------------------------------------------------------------------------------------ prints: ------------------------------------------------------------------------------------------------------ one ---> 1 three ---> 3 two ---> 2 ------------------------------------------------------------------------------------------------------ ================================================== I really need feedback! ================================================== If you are using this library please send me your comments (if you want by private email: matias <dot> capeletto <at> gmail <dot> com ) Library sources are in the vault: ---------------------------------------------------------------------------------------------------- http://tinyurl.com/2t8c87 Documentation can be viewed online here: ---------------------------------------------------------------------------------------------------- http://tinyurl.com/22sja5 Or can be downloaded from: ---------------------------------------------------------------------------------------------------- http://tinyurl.com/22jfdf Main changes ---------------------------------------------------------------------------------------------------- Lib: * Added support for information about the relations * Added range_type and const_range_type typedef to indexed views * Change the way you tag the collections. * hard implementation decision * Some small bug fixes Docs: * New relation and pair layout diagram * New section: Additional information * More examples in Bimap and Boost * New section: Bimap and Boost.Xpressive * New section: Bimap and Boost.Typeof * New section: Bimap and Boost.Foreach * Update the reference Comments are welcome Best regards Matias

1 1

[soc][regex] multiple existing practices for named sub-expressions
by Hugh Wimberly 04 Jun '07

04 Jun '07

Part of my GSoC project (and probably the easiest part) is to implement named sub-expressions in Boost.Regex. Right now, most regex engines have this capability, but there are two camps for how the named sub-expressions match up with the numbers used for backreferences. A very useful overview of the two possibilities can be found at http://www.regular-expressions.info/named.html Essentially, .NET numbers all the named captures after all the unnamed captures, and everybody else simply numbers them all in order from left to right. My personal opinion is that the Boost.Regex implementation should follow Python's and PCRE's practice instead of the .NET syntax or an amalgam of the two, but this is a decision for the boost community, not for me. Thoughts? -Hugh

3 2

[uuid] uuid::create() is extremely slow
by Dave Jenkins 04 Jun '07

04 Jun '07

Andy, You did a nice job speeding up the uuid::create(engine) function to 0.4 seconds for 1 million UUIDs. But now the uuid::create() function is extremely slow - 131 seconds for 1 million UUIDs (source code below). Can you change it to use a static mt19937 engine? Thanks, Dave Jenkins #include <boost/uuid.hpp> #include <boost/progress.hpp> #include <string> int main(int, char*[]) { using namespace boost; int const N = 1000000; // mt19937 engine; timer t1; for (int i=0; i<N; i++) uuid::create(); // uuid::create(engine); std::cout << "elapsed time: " << t1.elapsed() << '\n'; return 0; }

6 12

GSoC: "Boost.Extension and Reflection" library [plugin]
by Janek Kozicki 04 Jun '07

04 Jun '07

The link from http://code.google.com/soc/boost/about.html states: Title: Boost.Extension and Reflection Student: Mariano Gabriel Consoni Mentor: Hartmut Kaiser "The project that I propose is centered in adding to Boost a library that allows the programmer to include extensions/plugins. Additionally this will be the start point to implement reflection concepts. [...]" I want to ask what is the progress with this library? I remember from GSoC 2006 conclusions that we all agreed that the progress during SoC should be advertised on boost mailing list to encourage participation from other people (especially looking at code during its evolution). I'm not even sure if SoC has already started, or will it start next month? Anyway I want to point out that I'm particularly curious about the progress in this area. PS: why this wiki webpage for GSoC 2007 is wiped out? http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl?Google_Summe… -- Janek Kozicki |

6 13

[soc 07] SoC 2007 Projects
by Jeff Garland 04 Jun '07

04 Jun '07

Hi All - I'm happy to announce that Boost will be mentoring 9 Google SoC projects this year. They are: Title Student Mentor ----------------------------------------------------------------------------- User-friendly graphs and their measures Andrew Sutton Jeremy Siek FastCGi and SCGI-compatible CGI library Darren Garvey Christopher M Kohlhoff Boost 'Big Integer' proposal Arseny Kapoulkine Jeff Garland Signal Network library Stjepan Rajko Douglas Gregor Evolutionary Computation Library Aaron Garrett Jaakko Järvi Application to extend Boost regex with recursive matching, named sub-expressions, and automata-based matching Hugh Wimberly John Maddock Visualization Of Arrays And STL Containers Jacob R>Voytko Joaquín L. Muñoz Boost.Extension and Reflection Mariano G. Consoni Hartmut Kaiser A Generic Geometric Library Huseyin Akcan Herve Bronnimann You can find the abstracts of these projects online here: http://code.google.com/soc/boost/about.html As with last year, there were many worthy students and projects that ultimately were not selected. So, for all the students that were not selected realize that the competition for a very few slots was immense. My main advice for students that might want to try again next year is to start participating now -- it really helps to hang out on the list, submit patches, and work on ideas well in advance of SoC. I'd like to thank the mentoring team that evaluated the ~220 applications we received this year. It takes a huge effort to go thru these, read, rank, clarify and give feedback. Note that in addition to the mentors above Andreas Huber, Rene Rivera, and Daniel Wallin all helped with evaluations. For those coming to BoostCon I'll be giving a presentation on Friday about Boost and SoC: http://www.boostcon.com/program/sessions#garland-summer-of-code And, if you aren't coming to BoostCon, well, you should ;-) http://www.boostcon.com/registration Jeff

5 5

[ [bimap] CodeWarrior 9.4
by Matias Capeletto 04 Jun '07

04 Jun '07

I have solve all the problems with Boost.Bimap in CW 9.4 except for one. I do not know how to solve this last one. Regression are here: http://engineering.meta-comm.com/boost-regression/CVS-HEAD/developer/bimap.… Can a CW friend give me a hint? Thanks! Best Regards Matias

1 0

[bimap] typo in test_bimap_project.cpp
by Boris Gubenko 04 Jun '07

04 Jun '07

bimap library test test_bimap_project.cpp fails to compile because of typo: list_of< tagged< lstd::string, right_tag > ^ Thanks, Boris

2 1