- Boost - lists.boost.org

Re: [boost] [LibrariesUnderConstruction] update of LibrariesUnderConstruction wiki page
by vicente.botet 14 Jan '09

14 Jan '09

Hi, The wiki page Libraries Under Construction contains now much more entries. Good trip, Vicente ----- Original Message ----- From: "vicente.botet" <vicente.botet(a)wanadoo.fr> To: <boost(a)lists.boost.org>; <boost-announce(a)lists.boost.org>; <boost-users(a)lists.boost.org> Sent: Sunday, December 07, 2008 5:50 PM Subject: [LibrariesUnderConstruction] new LibrariesUnderConstruction wiki page Hi, I have created a new wiki page containing some of the libraries under construction for Boost. You can access this page from the Boost Trac wiki https://svn.boost.org/trac/boost/wiki > Projects > Libraries Under Construction or directly https://svn.boost.org/trac/boost/wiki/LibrariesUnderConstruction The contents of this page is by nature incomplete and outdated. Please help to maintain this page updated by modifying it directly (if you have the rights) or posting on the Boost mailing lists boost(a)lists.boost.org or boost-users(a)lists.boost.org with the prefix [LibrariesUnderConstruction]. If you're the author of one of these libraries and desire that all or part of the information related to your library be removed or modified, please send a mail or you can do it directly if you prefer. If you're a library author and plan on submitting a library but don't know when it will be ready for review, and you want to include your library in this list, please add it directly (if you have the rights) or send a mail. If you're a library author and plan on submitting a library for review in the next 3-6 months, email Ron Garcia or John Phillips ("garcia at osl dot iu dot edu" and "phillips at mps dot ohio-state dot edu" respectively) a short description of your library and they will add it to the Review Wizard Report Libraries Under Construction section. HTH, Vicente

1 0

[interthreads] Library under development v0.2
by vicente.botet 14 Jan '09

14 Jan '09

Hi, I have added the threader/joiner classes following the design of Kevlin Henney N1833 - Preliminary Threading Library Proposal for TR2 (http:/www.open-std.org/jtc1/sc22/wg21/docs/papers/2005/n1883.pdf). I use the Boost.Futures library from Anthony (review ongoing) In addition and as a generalization of these classes I have started a framework working with asynchronous executors -AE (threader) and asynchronous completion tokens -ACT (joiner). The library include also an launcher (AE) associated to the unique_future (ACT) which is the equivalent of the thread_pool in N2276 - Thread Pools and Futures (http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2276.html) With some minor changes in the ThreadPool library (review scheduled), the tp:pool can be seen as a AE and the tp::task as a ACT. The library allows also to fork several functions at the same time. The result of the fork_all operation is a fusion tuple of asynchronous completion token handles. The user can apply any fusion algorithm on this tuple as for example. To make common tasks easier the library provide some functors and some functions that are the result of applying these functors to the tuple using a fusion algorithm. Last the library provides also some functions like wait_for_all that forks all the functions and wait for the result of all of them and wait_for_any which works with functions that return the same type, and forks all the functions and wait for the first one. For the moment the AE/ACT framework is included it in Bosst.InterThreads library, but this is independent and can be included in an independent library (any suggestion for a name and a namespace?) The library is available at Boost Sandbox: https://svn.boost.org/svn/boost/sandbox/interthreads and Boost Vault: http://www.boostpro.com/vault/index.php?action=downloadfile&filename=intert… Includes every thing on the Sandbox + Html documentation + Boost.Futures + Boost.ThreadPool adaptations I'd appreciate it if people could take a look, test it on their compiler and give me some feedback. It's not quite ready for formal review yet. Best regrads, Vicente _____________________ Vicente Juan Botet Escribá

1 0

[mpl] sequence_tag
by Eric Niebler 14 Jan '09

14 Jan '09

A type is an mpl sequence if mpl::sequence_tag<T>::type is something other than mpl::non_sequence_tag. There are a few ways of hooking mpl::sequence_tag, but the easiest is for T to have a nested typedef named "tag". I find this unfortunate. Consider proto expression types, which, in addition to being nodes in expression trees, are also both fusion sequences and mpl sequences. A proto expression type has three tag types associated with it: 1) The Proto tag (e.g., proto::tag::plus), a nested typedef named "proto_tag" 2) The Fusion tag, a nested typedef named "fusion_tag" 3) The MPL tag I can't in good conscience use a nested typedef named "tag" for the MPL tag because that would lead to confusion. It would be nice if MPL had instead chosen the name "mpl_tag" for its tag type. It's not too late to make the change ... MPL could first look for "mpl_tag" before falling back on "tag". For now, I have chosen to specialize mpl::sequence_tag for the proto::expr type. This is an incomplete solution because types that extend proto::expr are not themselves automatically MPL sequences. Another solution might be to allow SFINAE on sequence_tag, like ... template<typename T, typename EnableIf = void> struct sequence_tag; That way I could partially specialize for proto expression types, like: template<typename T, typename enable_if<proto::is_expr<T> >::type> struct sequence_tag { ... Thoughts? FWIW, I'd prefer MPL to rename the nested tag typedef to "mpl_tag". -- Eric Niebler BoostPro Computing http://www.boostpro.com

4 5

[linux drivers] Is it possible to use boost in a Linux driver?
by Sid Sacek 14 Jan '09

14 Jan '09

Does anybody know if it's possible to use boost within a Linux driver?

8 13

[tr1] Help needed with Darwin port
by John Maddock 14 Jan '09

14 Jan '09

Can someone with access to various flavours of gcc-4.x on Darwin let me know what directory the glibc++ std lib headers are in? Currently Boost.TR1 is using #include <../c++/header_name> which is taken from STLPort's configuration, but is apparently causing regressions in the boost code, Thanks, John.

4 10

[scope_exit] Doesn't included in Boost yet?
by Dmitriy Maksimov 14 Jan '09

14 Jan '09

Hi, I didn't found boost::scope_exit library in Boost 1.37 althougth the library approved in May 2008. Why? Kind regards, Dmitriy

6 15

Re: [boost] [lexical_cast] A suggestion
by Andrew Troschinetz 14 Jan '09

14 Jan '09

Alexander Nasonov <alnsn <at> yandex.ru> writes: > This looks interesting but boost::optional<T> is already > InputStreamable: > #include <boost/optional/optional_io.hpp> This sounded really cool to me so I attempted a cursory hack to get this working. The only way I could figure out how to provide an override for optional<T> that never throws was to change detail::lexical_cast from a function to a struct to get partial template specialization. Maybe there is a better way, I'm new at this. It seems like there must be because I wound up with lots of almost- duplicated code. Here's the old familiar detail::lexical_cast, structified: And here's the optional<T> specialization: Initially this didn't work however. I traced it to what might be a bug in optional_io.hpp's insertion operator: if ( in.good() ) { int d = in.get(); if ( d == ' ' ) /* assuming a space in stream, what if there's not? */ { T x ; in >> x; v = x ; } else v = optional<T>() ; } return in; With the above insertion operator code, this would fail: istringstream si ("test"); optional<string> o; si >> o; assert (o); /* fails, would work if si were constructed with " test" instead */ That seems wrong to my thinking, but maybe there's a reason for the if (d == ' ') test? If the intent was to simply eat up spare white space, how about this instead: if (in) { T x; if (in >> std::ws >> x) v = x; else v = optional<T>(); } return in; Using that code in optional_io.hpp, and with the above changes, things work out just fine: int invalid_as_uint = -1; int valid_as_uint = 5; unsigned default_uint = 6; // Normal form, throws. try { unsigned i = lexical_cast<unsigned> (invalid_as_uint); cout << i << endl; } catch (const bad_lexical_cast&) { cout << "bad lexical cast error" << endl; } // Overload for lexical_cast<optional<T>>, does not throw. optional<unsigned> i = lexical_cast<optional<unsigned> > (invalid_as_uint); if (i) cout << *i << endl; else cout << "invalid return value" << endl; i = lexical_cast<optional<unsigned> > (valid_as_uint); if (i) cout << *i << endl; else cout << "invalid return value" << endl; Also, I would love to see a fallback style lexical_cast for the same reasons Neil already mentioned. -- Andrew Troschinetz

1 0

MSVC9 SFINAE quirks?
by Simonson, Lucanus J 14 Jan '09

14 Jan '09

Is there some gotcha about metaprogramming with MSVC9? I recently finished my geometry library rewrite and found that I couldn't port it to windows because the SFINAE overloading of generic functions is viewed as ambiguous in the more complex cases, but was readily accepted by gcc and icc in linux. The following code compiles in recent versions of gcc, but fails in MSVC9 //simple SFINAE overload of foo works in both template <typename T> typename gtl_if<typename is_mutable_polygon_90_type<T>::type>::type foo(const T& t) { return gtl_yes(); } template <typename T> typename gtl_if<typename is_mutable_polygon_45_type<T>::type>::type foo(const T& t) { return gtl_yes(); } //metafunction wrapper around SFINAE to supply return type works in both template <typename T> typename requires_1<typename gtl_if<typename is_mutable_polygon_90_type<T>::type>::type, void>::type foo2(const T& t) { } template <typename T> typename requires_1<typename gtl_if<typename is_mutable_polygon_45_type<T>::type>::type, void>::type foo2(const T& t) { } //meta function wrapper around two checks works in gcc but not msvc9 // error C2668: 'gtl::foo4' : ambiguous call to overloaded function template <typename polygon_type, typename polygon_type_2> typename requires_1< typename gtl_if< typename gtl_and< typename is_mutable_polygon_90_type<polygon_type>::type, typename is_mutable_polygon_90_type<polygon_type_2>::type >::type>::type, polygon_type>::type foo4(polygon_type& polygon, const polygon_type_2& point) {return point;} template <typename polygon_type, typename polygon_type_2> typename requires_1< typename gtl_if< typename gtl_and< typename is_mutable_polygon_45_type<polygon_type>::type, typename is_mutable_polygon_45_type<polygon_type_2>::type >::type>::type, polygon_type>::type foo4(polygon_type& polygon, const polygon_type_2& point) {return point;} int main() { using namespace gtl; polygon_45_data<int> p45; foo(p45); foo2(p45); foo4(p45, p45); //error here //this works as expected and has no syntax error in msvc9 gtl_if<gtl_and< is_mutable_polygon_45_type<polygon_45_data<int> >::type, is_mutable_polygon_45_type<polygon_45_data<int> >::type >::type>::type a; //this works as expected, generating syntax error in msvc9 // error C2039: 'type' : is not a member of 'gtl::gtl_if<gtl::gtl_no> gtl_if<gtl_and< is_mutable_polygon_90_type<polygon_45_data<int> >::type, is_mutable_polygon_90_type<polygon_45_data<int> >::type >::type>::type b; Does someone like Steve Wanatabe with more experience with MS compilers know what I have inadvertently run afoul of and how to get myself out of this jam? Thanks, Luke

2 11

[mp_int] Incorrect comparison with native types
by Nathan Kitchen 13 Jan '09

13 Jan '09

When I create an negative mp_int outside the range of a native type and compare it to a number of that type, I get incorrect results. Here is an example to reproduce the behavior: #include <iostream> #include <limits> #include <boost/cstdint.hpp> #include <boost/mp_math/mp_int.hpp> using namespace boost::mp_math; using namespace std; int main() { int32_t MIN = numeric_limits<int32_t>::min(); mp_int<> a = MIN; cout << a << endl; a -= 1; // Should print "-2147483649 1" but prints "-2147483649 0" instead cout << a << ' ' << (a < MIN) << endl; return 0; }

2 7

1.38 release schedule
by John Maddock 13 Jan '09

13 Jan '09

Since there haven't been any announcements, do I correctly assume that the 1.38 schedule is likely to slip by a couple of weeks? Cheers, John.

2 1