- Boost - lists.preview.boost.org

[website] SSL certificate for *.boost.org has expired
by Lars Viklund 17 Feb '13

17 Feb '13

It seems like the certificate for *.boost.org expired on the 16th, I'd reckon it's time to renew it and review the routines for renewing it in time. -- Lars Viklund | zao(a)acc.umu.se

2 1

[serialization] Limited Forward Compatibility
by Dmitry Shesterkin 17 Feb '13

17 Feb '13

Hello, We use xml serialization for saving and loading several types of metadata and we have to support possibility of reading new versions of files with old versions of soft. Boost::serialization is very good at backward compatibility but does not provide anything with opposite problem. So, This patch introduces new flag boost::archive::skip_unknown_tail_tags. When this flag is set and basic_xml_iarchive is faced with unknown tags at the end of element it will skip them (include nested). E.g., the first version of some file was: <Test class_id="0" tracking_level="0" version="0"> <a>31</a> <t2 class_id="1" tracking_level="0" version="0"> <a>11</a> 12 </t2> 32 </Test> After a while, the second version could be: <Test class_id="0" tracking_level="0" version="0"> <a>31</a> <t2 class_id="1" tracking_level="0" version="1"> <a>11</a> 12 <c>13</c> <d>14</d> <t1 class_id="2" tracking_level="0" version="0"> <a>1</a> 2 </t1> </t2> 32 </Test> Therefore, if the first version software set skip_unknown_tail_tags it can read second version of file (if does not set then exception will be thrown). See patch in attachments for details. Could it be useful, and should it be merged in trunk? Thanks in advance, Dmitry

2 3

[geometry] Documentation building
by Alexey Kutumov 17 Feb '13

17 Feb '13

I'm working on CMake'fication of boost libraries, and i 've got issue when building documentation of Boost.Geometry. This library uses doxygen_xml2qbk utility for convertiong doxygen XML to quickbook, and this utility uses RapidXML for xml processing. I found rapidxml in Boost.PropertyTree library, but it is in detail subdirectory and also have different namespaces (`boost::property_tree::detail::rapidxml` vs `rapidxml`) Can i use rapidxml from PropertyTree library, or i must write CMake script for searching rapidxml as thirdparty library?

2 1

[atomic] errors since the introduction of constexpr noexcept.
by Vicente J. Botet Escriba 17 Feb '13

17 Feb '13

Hi, There are some errors since the introduction of constexpr noexcept. Could you apply this patch? Best, Vicente svn diff boost/atomic Index: boost/atomic/detail/cas64strong.hpp =================================================================== --- boost/atomic/detail/cas64strong.hpp (revision 82919) +++ boost/atomic/detail/cas64strong.hpp (working copy) @@ -13,6 +13,7 @@ // atomic, so they are funnelled through cmpxchg as well. #include <cstddef> +#include <boost/config.hpp> #include <boost/cstdint.hpp> #include <boost/memory_order.hpp> #include <boost/atomic/detail/config.hpp> @@ -34,7 +35,8 @@ typedef T value_type; typedef T difference_type; public: - BOOST_CONSTEXPR base_atomic(atomic_type v) BOOST_NOEXCEPT: v_(v) {} + int xxx; + BOOST_CONSTEXPR base_atomic(value_type v) BOOST_NOEXCEPT: v_(v) {} base_atomic(void) {} void @@ -158,7 +160,7 @@ typedef void * value_type; typedef ptrdiff_t difference_type; public: - BOOST_CONSTEXPR base_atomic(atomic_type v) BOOST_NOEXCEPT: v_(v) {} + BOOST_CONSTEXPR base_atomic(value_type v) BOOST_NOEXCEPT: v_(v) {} base_atomic(void) {} void @@ -253,7 +255,7 @@ typedef T * value_type; typedef ptrdiff_t difference_type; public: - BOOST_CONSTEXPR base_atomic(atomic_type v) BOOST_NOEXCEPT: v_(v) {} + BOOST_CONSTEXPR base_atomic(value_type v) BOOST_NOEXCEPT: v_(v) {} base_atomic(void) {} void Index: boost/atomic/detail/gcc-x86.hpp =================================================================== --- boost/atomic/detail/gcc-x86.hpp (revision 82919) +++ boost/atomic/detail/gcc-x86.hpp (working copy) @@ -1621,7 +1621,7 @@ template<typename T> inline T -platform_load64(const volatile BOOST_NOEXCEPT T * ptr) +platform_load64(const volatile T * ptr) BOOST_NOEXCEPT { T expected = *ptr; for (; !platform_cmpxchg64_strong(expected, expected, const_cast<volatile T*>(ptr));)

2 3

[mixin] Introducing a new library.
by Borislav Stanimirov 17 Feb '13

17 Feb '13

Hello, Boost.Mixin is a library I'm developing with the idea to someday submit it to Boost. It is an entity-component system, with full abstraction between interface and implementation. In short, users can define objects by combining components (often referred to from here on out as mixins). A mixin is a regular C++ class for which two macros are defined. The main thing however are the messages. Instead of getting the component with some function like `get_component` (which is possible but discouraged in the library), users deal with objects only. They can call messages (inspired by Smalltalk's messages) which are dynamically dispatched to the appropriate mixin. The messages (also defined with macros) are the only thing needed for a user to call methods from the object's mixins. When defining a mixin, the user tells which messages it's going to implement, and when adding this mixin to the object a virtual table of sorts is filled and the object starts responding to those messages by calling the mixin's actual C++ methods. Some of the key features are: * Multicast messages. A message that is implemented and handled by more than one mixin. See the message `trace` in the example. (As opposed to unicast messages, that are handled by a single mixin) * Object access from within the mixin. The macro `bm_this` (Boost.Mixin this). See the renderer do in the example. * "Live" object mutation by adding and removing mixins. As demonstrated in `change_rendering_system` in the example. This is done via the `object_transformer` class and is currently the only way of adding mixins to objects. * Mixin (and message) domains. The domain is a collection of mixins and messages that work together. If you create a set of mixins that will never conceivably have to exist in the same object as another set of mixins, you should use a different domain. Here is a github link to the library: http://github.com/iboB/boost.mixin A working example can be found here: http://github.com/iboB/boost.mixin/tree/master/libs/mixin/example/basic The example is nothing more than some objects and mixins working together as they would hypothetically in a simple game. All "game" code has been shortcut to text outputs like: "right now an object is being rendered by the d3d renderer". The example demonstrates most of the currently supported library features. There is no jamfile yet. I develop it with Qt Creator and Code::Blocks in Linux and MS Visual Studio Express 2008 and 2012 in Windows (the Windows projects might be somewhat outdated), so the are project files for these four in libs/mixin/build. The same goes for the example. I'd qualify the current state of the library as pre-alpha, so it might have some bugs. I also create a small performance testing project (much more is to be added). It calls a regular function, a virtual function, `std::function` and a unicast mixin message a hundred million times. So far I'm generally happy with the performance, which is not to say that I won't be working on improving it. Still my computer is the only one I've tested it on and I'd be happy if somebody else tries the perf project. Here are my results: 100,000,000 x | Debug t | Release t ------------------------------------------ regular method | 781 ms | 211 ms virtual method | 639 ms | 253 ms `std::function` | 10695 ms | 272 ms Mixin message | 3298 ms | 865 ms It should also be noted that currently the library depends on a piece of code from Don Clugston's fastdelegate (found here: http://www.codeproject.com/Articles/7150/Member-Function-Pointers-and-the-F…) The piece in question has been copy-pasted verbatim (wrapped around in the boost::mixin namespace) in the third_party subfolder. Don Clugston has done a wonderful job of creating a platform-independent representation of C++ methods as a pair of pointers. Boost.Mixin heavily relies on such a thing for its virtual tables. For now I am the only developer of the library. Zahary Karadjov (credited in the license) hasn't written anything in it yet, but he is responsible for most of the library's public interface design, the idea to reuse fastdelegate, and the ideas and implementation suggestions for many of the roadmap features listed below. The programming languages Smalltalk, Ruby, and, to a degree, Objective C have also influenced the design. So, here is a project roadmap of sorts: 1. Soon: * Overloaded messages - messages that call overloaded functions. The message's name is different for each overload but the calling function remains the same * Work on more sensible (and helpful) runtime and compile-time errors. Missing messages, ambiguous message priority, etc. * Add multicast result collectors. Like `and` or `or` for boolean return values. Like `sum` for numeric values. Sample usage some_multicast_message(obj, p1, p2, boost::mixin::sum). User-defined result collectors will be possible. * Add a new mixin feature `fact`. Facts are static const variables in a mixin. No multicasts. No virtual tables. * Add object type templates. Instead of manually transforming each object, you would be able to call `type_template::new_object()` * Add more graceful runtime error handling: default message implementations (if none is provided), more specific non-breaking exceptions * Rigorous unit tests * Performance tests * Tutorials 2. Later: * Boost.Jam build files * Full reference documentation * Formal submission to Boost?? * Call mixin constructors when mutating object or constructing them via object type templates. Sample usage: `mutate(obj).add<mixin>("foo", param2).add<other_mixin>(42)` * Reduce library memory footprint * Optional external allocators per domain. Every memory allocation should pass through them. * Make use of the fact that a BIG execution time improvement can be made if a user has no mixins or messages defined in a dynamic library (and has no plans to add such) * Integration with Boost.Serialization. As long as the mixins of an object are serializable, the object will be, too. * "Static" mixins. As long as they're not mutated in or out, it provides a faster message execution. * Reflection. Create objects with mixins by string. Call messages by string. * Object mutation injections. Automatically add some mixins when others are present. Substitute a mentioned mixin with another. Ban a mixin from being added to an object. 3. When humankind becomes a Kardashev type two civilization: * Integration with Boost.Thread. Call messages for a specific thread. * User-defined message hooks. Sample implementation with messages called on objects on another device through Boost.Asio. * A way to mutate static mixins in and out. Slight mutation execution penalty. * Some magic to make it work even faster. * Integration with Boost.Python and luabind. * Remove C++03 backward compatibility???? * Library being accepted in Boost???? I think I will be done with the items form list one by April (some of them are even semi-implemented) and with the items of list two by the end of the year (and if I have more free time maybe even sooner). As for the items in the this list... well... they are nice to think about. I do have at least some vague ideas for each of them, but then again, this is list three. Any questions, comments, and suggestions are welcome. ------ Borislav Stanimirov ------ PS: Whoa. I'm sorry this post turned out so long. I didn't expect it to.

6 22

[Lockfree] ringbuffer
by Rich E 17 Feb '13

17 Feb '13

First off, all Boost.Lockfree tests in trunk are passing for me on clang 4.0 on OS X, hurrah! Thank you Tim for working on this for so long... I'm running through the examples now, only with the outdated documentation as a guide (more than a couple years old). I cannot find the ringbuffer container - was it renamed or removed altogether during the review period? I notice spsc_queue acts awefully like a ringbuffer, and it's implementation contains many

2 6

Track system Connection failing
by Vicente Botet 16 Feb '13

16 Feb '13

While connecting to the Trac system https://svn.boost.org/trac/boost/newticket I'm getting svn.boost.org uses an invalid security certificate. The certificate expired on 16/02/2013 17:02. The current time is 16/02/2013 17:25. (Error code: sec_error_expired_certificate) Could others confirm this behavior? Best, Vicente -- View this message in context: http://boost.2283326.n4.nabble.com/Track-system-Connection-failing-tp464291… Sent from the Boost - Dev mailing list archive at Nabble.com.

2 1

[array] failures in c++11 mode
by Tim Blechmann 15 Feb '13

15 Feb '13

hi, r82834 seems to break compilation of boost.array when compiled in c++11 mode: ../../../boost/array.hpp:131:9: error: body of constexpr function ‘constexpr const T& boost::array<T, N>::operator[](boost::array<T, N>::size_type) const [with T = boost::heap::detail::skew_heap_node<q_tester, false>*; long unsigned int N = 2ul; boost::array<T, N>::const_reference = boost::heap::detail::skew_heap_node<q_tester, false>* const&; boost::array<T, N>::size_type = long unsigned int]’ not a return-statement the boost.heap testsuite exposes this failure: http://www.boost.org/development/tests/trunk/developer/output/Sandia-gcc-4-… thnx, tim

5 9

setting <cxxflags> and <linkflags> in Jamfile?
by Oliver Kowalke 14 Feb '13

14 Feb '13

Hello, if the define BOOST_USE_XYZ is given at b2/bjam command-line how can I set special compiler- and linker-flags in the Jamfile? something like this: Jamfile.v2: *project boost/abc : requirements <toolset>gcc-4.7,<defined>BOOST_USE_XYZ:<cxxflags>-fxyz <toolset>gcc-4.7,<defined>BOOST_USE_XYZ:<linkflags>"-static-libxyz" <link>static <threading>multi : source-location ../src : usage-requirements <link>shared:<define>BOOST_ABC_DYN_LINK=1 ; * if the lib is build with b2 toolset=gcc cxxflags="-DBOOST_USE_XYZ" then -fxyz static-libxyz is applied to the compiler otherwise not. How can I achieve this behaviour? best, Oliver

2 2

Re: [boost] [smart ptr] Any interest in copy-on-write pointer for C++11?
by Andreas Pfaffenbichler 14 Feb '13

14 Feb '13

hi ralph! a heavy argument for a 'COW' pattern is made here: http://cppnow.org/session/value-semantics-and-concepts-based-polymorphism ( 'copy_on_write' class template included). maybe it could be interesting to compare your 'cow_ptr' to this 'copy_on_write'. but no doubt: the direct support of 'COW' in boost would fill e need thank you for your efforts! best regards andreas pfaffenbichler

7 20

[gil] ToolBox uploaded
by Christian Henning 14 Feb '13

14 Feb '13

Hi all, I finally have managed to upload the "toolbox" extension for boost::gil. The toolbox provides some helpful extension for gil that are too small to be an extension themselves. There are: * color spaces: cmyka, gray_alpha, hsl, hsv, lab, xyz * metafunctions: channel_type_to_index, is_bit_aligned, is_similar, is_homogeneous, get_num_bits, channel_type, get_pixel_type, pixel_bit_size * image types: indexed_image * color converters: gray -> rgba I would need some help hooking up the tests and doc generation. Let me know if you any questions or improvements. Regards, Christian

3 9

Relational database library (was: [GSOC] Are we going to participate?)
by Mateusz Loskot 14 Feb '13

14 Feb '13

On 14 February 2013 09:40, Roland Bock <rbock(a)eudoxos.de> wrote: > On 2013-02-14 06:04, Michael Marcin wrote: >> - relational database library (what ever happened with boost and soci?) > FYI: > I am currently working on a C++11 SQL library. Once you declare your > table structures in C++, it will allow you to express queries very close > to SQL: For example: > > TabSample t; > auto results = sql::select(sql::distinct, t.b, t.c) > .from(t) > .where(t.a > 7 and t.b != "cheesecake") > .order_by(t.c).run(db); > > for (auto row: results) > { > int a = row.a; > std::string c = row.c; > } > > The library will take care of escaping (to prevent code injection), the > compiler will enforce type safety (e.g. disallow comparison of integral > and string values). There will also be a lot of sanity checks at compile > time, for instance you cannot call .from() twice. It will even be able > to tell you, if you are referencing tables, which are missing in > .from(), etc. > > No more string concatenation and hoping to get the syntax right. The > compiler will tell you, if you do it wrong. > No more indexed access to string results. Result rows will allow you to > access individual columns by name (and they will have a decent type). > [...] Roland, It all sounds exciting. I remember the discussion long time ago we had, with similar ideas brainstormed [1], where you also asked this question, later discussed with Maciej: "Maybe it would make sense to add such a layer as an option to SOCI?" FYI, in SOCI, we're planning [2] to move SOCI towards C++11 It likely may happen during SOCI 4.0.0 release. So, I'd be interested in taking another look at this and having such layer adopted in SOCI too. Where I could learn about your solution more? [1] http://lists.boost.org/Archives/boost/2010/09/170983.php [2] https://github.com/SOCI/soci/wiki/Roadmap Best regards, -- Mateusz Loskot, http://mateusz.loskot.net

1 0

[Git] Documentation update
by Beman Dawes 14 Feb '13

14 Feb '13

The "Git and Modular Boost" documentation has been updated: * The GitHub mothership with be github.com/boostorg * Full history for the current trunk and branches/release will be available. * There are now active milestone and issue trackers for the conversion. See https://svn.boost.org/trac/boost/wiki/WikiStart#GitandModularBoost --Beman

10 31

adding comment to boost_1_54_0.qbk?
by Oliver Kowalke 14 Feb '13

14 Feb '13

Is it allowed to add an entry for updated libraries in public_html/live/feed/history/boost_1_54_0.qbk? Oliver

2 1

[smart ptr] Any interest in copy-on-write pointer for C++11?
by Ralph Tandetzky 14 Feb '13

14 Feb '13

Hi, is there any interest in a copy-on-write pointer implementation? I wrote a cow_ptr<T> template class for C++11 which has the following use cases: 1. They are well suited as pimpl pointers for classes with value semantics. With most other smart pointers it is always necessary to reimplement the copy constructor, and the copy assignment operator, sometimes also the destructor. This is not necessary here and you'll get the right semantics and even the additional performance gain due to lazy copying. A futher advantage is that constness of cow pointer member functions propagates to the Impl object naturally making it easier to write const-correct code. // central_class_with_value_semantics.h class CentralClassWithValueSemantics { public: // ... public interface goes here ... // private: struct Impl; // forward declaration cow_ptr<Impl> m; }; // central_class_with_value_semantics.cpp struct CentralClassWithValueSemantics::Impl { // ... definition of hidden members goes here ... // } This is called the pimpl idiom, or private implementation idiom, handle body idiom, or compiler firewall idiom. 2. For classes with members whose copy-operations are expensive and/or which take a lot of space in memory, these members can be wrapped in a cow pointer. An example are matrix or image classes whose data might be stored in a std::vector. The matrix header information may be changed without deep copy. This boosts performance and optimizes memory usage, but at the same time retains the value semantics one feels comfortable with. class Matrix { public: // ... public interface goes here ... // private: size_t nRows, nCols; // can be touched without deep copy. cow_ptr<std::vector<float>> data; // copy-on-write } In such classes the default version of copy and move construction will usually work just fine. 3. You can add cloning to a class hierarchy from the outside. With cow_ptr<Base> a( new Derived1 ); cow_ptr<Base> b( new Derived2 ); cow_ptr<Base> c; c = a; // performs a shallow copy. c->doSomething(); // makes a deep copy of a as a Derived1 class. // There is no slicing involved. you copy Base objects polymorphically. The class Base can even be abstract here. It is only required that Derived1 and Derived2 be CopyConstructible. 4. You can create arrays with elements that retain polymorphic behaviour and have genuine value sematics at the same time with std::vector<cow_ptr<Base>> polymorphic_array_with_value_semantics; -- There are probably other use cases. I published the code on github ( https://github.com/ralphtandetzky/cow_ptr) Any suggestions for improvements would be greatly appreciated. Regards, Ralph Tandetzky.

10 42

[container] varray aka static_vector
by Andrew Hundt 13 Feb '13

13 Feb '13

As Adam mentioned in an earlier thread we have renamed static_vector to varray to better represent how it works like an array with size without requiring the values to be default constructible, and to better distinguish it from hybrid_vector. varray location: https://svn.boost.org/svn/boost/sandbox/varray/ Documentation: https://svn.boost.org/svn/boost/sandbox/varray/doc/html/index.html One open question is if the 3rd strategy parameter should be public or in the detail namespace. Based on some of the feedback we moved the strategy from container to container_detail, and made the public version varray<typename T, std::size_t N>. However, we wanted to get others' thoughts on this change. Additionally, should primitives be default initialized with the contents of memory like an array or value initialized to 0 like a standard container in the constructor, resize(), and emplace_back()? Generally, built-in types are initialized when the constructor is called explicitly. For example: int a; // a has the value of underlying memory int b = int(); // b has a value of 0 Right now the default behavior of our container is more like a vector than an array when creating values using the default ctor. Default constructable values are explicitly initialized in the varray ctor, resize(), emplace() and emplace_back() functions, even when they are trivially constructable. We find this behavior more intuitive and therefore we have made it the default. If the performance needs to be improved slightly the behavior may be changed or disabled in container_detail::varray_traits<V, C, S>. Here is an example demonstrating the current default behavior: vector<int> vec(3); varray<int, 3> varr(3); array<int, 3> arr; vec[0] == varr[0] and vec[0] != arr[0] Any feedback is very much appreciated. Cheers! Andrew Hundt Adam Wulkiewicz

5 19

[bind] Boost.Bind compile error with msvc-11 and msvc-10
by Ian Emmons 13 Feb '13

13 Feb '13

I have attached a short, self-contained test case that uses Boost.Bind. this test case compiles and runs correctly with the following tool sets: darwin (gcc 4.2.1 on MacOS 10.8) clang (clang on MacOS 10.8) gcc (gcc 4.7 on Ubuntu 12.04) msvc-9 (Visual Studio 2008 on Windows 7) It fails to compile on these: msvc-11 (Visual Studio 2012 on Windows 7) msvc-10 (Visual Studio 2010 on Windows 7) I have also attached the error messages from msvc-11, which seem to indicate that Boost.Bind is the problem. Can anyone shed any light on this issue? Is this an error in my usage of Boost.Bind? A Boost.Bind bug? A Visual Studio bug? Something else? Thanks, Ian

6 7

[Contract] Eagerly waiting...
by Robert Jones 13 Feb '13

13 Feb '13

Hi Lorenzo (or anyone who knows) AFAICS your contract library seems to have been approved for a while, but not made it into a release. Do you have an ETA for this? Thx, Rob.

2 1

[container] boost::container::vector, hybrid_vector, and performance
by Andrew Hundt 13 Feb '13

13 Feb '13

After some of the varray discussion and interest in hybrid_vector I modified an existing synthetic benchmark from the C++Next site (cpp-next.com) to test varray, boost::container::vector, boost::container::vector<T, stack_allocator>, and std::vector. Essentially for each container it times constructing a container of containers of primitives, std::sort, std::rotate, and destruction. Note that the test calls reserve() when setting up each container so that each gets a fair shot and isn't unfairly bogged down by a simple allocation optimization. The short results summary: varray was generally at or near the top for performance due to its simplicity, and interestingly on almost all platforms std::vector significantly outperformed boost::container::vector, which edged out boost::container::vector<T, stack_allocator>. Of course these are synthetic tests of only a subset of these containers’ functionality using only primitives so it should not be construed to necessarily represent the performance of these containers for other uses. Nonetheless I thought it may be interesting for discussion. There is also an additional benchmark of Adam Wulkiewicz’s rtree using linux clang comparing varray, container::vector, and std::vector with similar results. One question this leads me to ask is if someone might be able to illustrate a significant advantage hybrid_vector could have over boost::container::vector<T, stack_allocator>? The test can be found at: https://svn.boost.org/svn/boost/sandbox/varray/example/bench_varray.cpp Results are below. Cheers! Andrew Hundt --------------------------------------------------------------------------------------------------------------------------- Apple LLVM version 4.2 (clang-425.0.24) (based on LLVM 3.2svn) Target: x86_64-apple-darwin12.2.0 Thread model: posix Processor 2.2 GHz Intel Core i7 Memory 16 GB 1333 MHz DDR3 Software OS X 10.8.2 (12C60) --------------------------------------------------------------------------------------------------------------------------- set stack size limit to: 17000000 bytes N = 1000 varray benchmark: construction took 10.576854s wall, 10.580000s user + 0.020000s system = 10.600000s CPU (100.2%) sort took 4.323139s wall, 4.290000s user + 0.010000s system = 4.300000s CPU (99.5%) rotate took 0.909033s wall, 0.910000s user + 0.000000s system = 0.910000s CPU (100.1%) destruction took 0.000725s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) done Total time = 15.824697s wall, 15.780000s user + 0.040000s system = 15.820000s CPU (100.0%) boost::container::vector benchmark construction took 11.656541s wall, 11.180000s user + 0.320000s system = 11.500000s CPU (98.7%) sort took 5.213782s wall, 4.940000s user + 0.000000s system = 4.940000s CPU (94.7%) rotate took 0.003201s wall, 0.010000s user + 0.000000s system = 0.010000s CPU (312.4%) destruction took 2.482177s wall, 0.640000s user + 2.000000s system = 2.640000s CPU (106.4%) done Total time = 19.367074s wall, 16.770000s user + 2.330000s system = 19.100000s CPU (98.6%) boost::container::vector benchmark with stack_allocator construction took 11.698265s wall, 11.110000s user + 0.430000s system = 11.540000s CPU (98.6%) sort took 5.227070s wall, 5.180000s user + 0.000000s system = 5.180000s CPU (99.1%) rotate took 0.003778s wall, 0.010000s user + 0.000000s system = 0.010000s CPU (264.7%) destruction took 2.493007s wall, 0.510000s user + 1.900000s system = 2.410000s CPU (96.7%) done Total time = 19.433571s wall, 16.810000s user + 2.340000s system = 19.150000s CPU (98.5%) std::vector benchmark construction took 11.605458s wall, 11.400000s user + 0.300000s system = 11.700000s CPU (100.8%) sort took 2.256151s wall, 2.250000s user + 0.000000s system = 2.250000s CPU (99.7%) rotate took 0.003200s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 2.498035s wall, 0.370000s user + 2.040000s system = 2.410000s CPU (96.5%) done Total time = 16.374640s wall, 14.020000s user + 2.350000s system = 16.370000s CPU (100.0%) varray/boost::container::vector total time comparison: wall = 0.817093 user = 0.940966 system = 0.0171674 (user+system) = 0.828272 varray/(boost::container::vector + stack_allocator) total time comparison: wall = 0.814297 user = 0.938727 system = 0.017094 (user+system) = 0.82611 varray/std::vector total time comparison: wall = 0.966415 user = 1.12553 system = 0.0170213 (user+system) = 0.966402 --------------------------------------------------------------------------------------------------------------------------- Windows VS2012 Core2Duo P7350 2GHz, 3GB Ram --------------------------------------------------------------------------------------------------------------------------- N = 1000 varray benchmark: construction took 18.901735s wall, 18.798121s user + 0.000000s system = 18.798121s CPU (99.5%) sort took 3.093771s wall, 3.260421s user + 0.000000s system = 3.260421s CPU (105.4%) rotate took 0.336938s wall, 0.280802s user + 0.000000s system = 0.280802s CPU (83.3%) destruction took 0.000270s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) done Total time = 22.339054s wall, 22.339343s user + 0.000000s system = 22.339343s CPU (100.0%) boost::container::vector benchmark construction took 18.757410s wall, 18.595319s user + 0.202801s system = 18.798121s CPU (100.2%) sort took 0.120896s wall, 0.140401s user + 0.000000s system = 0.140401s CPU (116.1%) rotate took 0.004482s wall, 0.015600s user + 0.000000s system = 0.015600s CPU (348.0%) destruction took 0.282280s wall, 0.046800s user + 0.171601s system = 0.218401s CPU (77.4%) done Total time = 19.167624s wall, 18.798121s user + 0.374402s system = 19.172523s CPU (100.0%) boost::container::vector benchmark with stack_allocator construction took 18.811076s wall, 18.361318s user + 0.312002s system = 18.673320s CPU (99.3%) sort took 0.134590s wall, 0.218401s user + 0.000000s system = 0.218401s CPU (162.3%) rotate took 0.004111s wall, 0.015600s user + 0.000000s system = 0.015600s CPU (379.5%) destruction took 0.279619s wall, 0.031200s user + 0.280802s system = 0.312002s CPU (111.6%) done Total time = 19.231975s wall, 18.626519s user + 0.592804s system = 19.219323s CPU (99.9%) std::vector benchmark construction took 19.587929s wall, 19.359724s user + 0.234002s system = 19.593726s CPU (100.0%) sort took 0.112920s wall, 0.124801s user + 0.000000s system = 0.124801s CPU (110.5%) rotate took 0.002486s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 0.262937s wall, 0.031200s user + 0.218401s system = 0.249602s CPU (94.9%) done Total time = 19.968799s wall, 19.515725s user + 0.452403s system = 19.968128s CPU (100.0%) varray/boost::container::vector total time comparison: wall = 1.16546 user = 1.18838 system = 0 (user+system) = 1.16517 varray/(boost::container::vector + stack_allocator) total time comparison: wall = 1.16156 user = 1.19933 system = 0 (user+system) = 1.16234 varray/std::vector total time comparison: wall = 1.1187 user = 1.14468 system = 0 (user+system) = 1.11875 --------------------------------------------------------------------------------------------------------------------------- Windows mingw_4_7 -O2 Core2Duo P7350 2GHz, 3GB Ram --------------------------------------------------------------------------------------------------------------------------- N = 1000 varray benchmark: construction took 13.752772s wall, 13.447286s user + 0.000000s system = 13.447286s CPU (97.8%) sort took 2.207612s wall, 2.449216s user + 0.000000s system = 2.449216s CPU (110.9%) rotate took 0.381684s wall, 0.452403s user + 0.000000s system = 0.452403s CPU (118.5%) destruction took 0.000265s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) done Total time = 16.348433s wall, 16.348905s user + 0.000000s system = 16.348905s CPU (100.0%) boost::container::vector benchmark construction took 16.736949s wall, 15.631300s user + 1.060807s system = 16.692107s CPU (99.7%) sort took 5.170353s wall, 5.226034s user + 0.000000s system = 5.226034s CPU (101.1%) rotate took 0.001871s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 0.537667s wall, 0.124801s user + 0.374402s system = 0.499203s CPU (92.8%) done Total time = 22.453639s wall, 20.982135s user + 1.435209s system = 22.417344s CPU (99.8%) boost::container::vector benchmark with stack_allocator construction took 16.227005s wall, 15.412899s user + 0.624004s system = 16.036903s CPU (98.8%) sort took 4.376050s wall, 4.477229s user + 0.000000s system = 4.477229s CPU (102.3%) rotate took 0.001871s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 0.523920s wall, 0.109201s user + 0.514803s system = 0.624004s CPU (119.1%) done Total time = 21.134513s wall, 19.999328s user + 1.138807s system = 21.138135s CPU (100.0%) std::vector benchmark construction took 15.182662s wall, 14.149291s user + 0.904806s system = 15.054096s CPU (99.2%) sort took 1.408413s wall, 1.528810s user + 0.000000s system = 1.528810s CPU (108.5%) rotate took 0.001668s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 0.451160s wall, 0.046800s user + 0.421203s system = 0.468003s CPU (103.7%) done Total time = 17.049415s wall, 15.724901s user + 1.326008s system = 17.050909s CPU (100.0%) varray/boost::container::vector total time comparison: wall = 0.728097 user = 0.779182 system = 0 (user+system) = 0.729297 varray/(boost::container::vector + stack_allocator) total time comparison: wall = 0.773542 user = 0.817473 system = 0 (user+system) = 0.773432 varray/std::vector total time comparison: wall = 0.958885 user = 1.03968 system = 0 (user+system) = 0.958829 --------------------------------------------------------------------------------------------------------------------------- Linux clang_3_0_6 -O2 Core2Duo P7350 2GHz, 3GB Ram --------------------------------------------------------------------------------------------------------------------------- set stack size limit to: 9000000 bytes N = 1000 varray benchmark: construction took 27.455347s wall, 27.360000s user + 0.050000s system = 27.410000s CPU (99.8%) sort took 5.683607s wall, 5.750000s user + 0.010000s system = 5.760000s CPU (101.3%) rotate took 1.651716s wall, 1.560000s user + 0.000000s system = 1.560000s CPU (94.4%) destruction took 0.002050s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) done Total time = 34.810039s wall, 34.700000s user + 0.060000s system = 34.760000s CPU (99.9%) boost::container::vector benchmark construction took 39.073566s wall, 33.070000s user + 5.960000s system = 39.030000s CPU (99.9%) sort took 13.405286s wall, 13.440000s user + 0.020000s system = 13.460000s CPU (100.4%) rotate took 0.011300s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 1.005573s wall, 0.810000s user + 0.150000s system = 0.960000s CPU (95.5%) done Total time = 53.509749s wall, 47.330000s user + 6.130000s system = 53.460000s CPU (99.9%) boost::container::vector benchmark with stack_allocator construction took 39.457245s wall, 33.660000s user + 5.790000s system = 39.450000s CPU (100.0%) sort took 13.454378s wall, 13.190000s user + 0.050000s system = 13.240000s CPU (98.4%) rotate took 0.013398s wall, 0.040000s user + 0.000000s system = 0.040000s CPU (298.5%) destruction took 1.007952s wall, 0.860000s user + 0.300000s system = 1.160000s CPU (115.1%) done Total time = 53.947026s wall, 47.750000s user + 6.140000s system = 53.890000s CPU (99.9%) std::vector benchmark construction took 31.662683s wall, 25.550000s user + 6.240000s system = 31.790000s CPU (100.4%) sort took 3.102917s wall, 2.870000s user + 0.080000s system = 2.950000s CPU (95.1%) rotate took 0.010094s wall, 0.020000s user + 0.000000s system = 0.020000s CPU (198.1%) destruction took 1.002616s wall, 0.770000s user + 0.210000s system = 0.980000s CPU (97.7%) done Total time = 35.793019s wall, 29.220000s user + 6.530000s system = 35.750000s CPU (99.9%) varray/boost::container::vector total time comparison: wall = 0.650536 user = 0.73315 system = 0.00978793 (user+system) = 0.650206 varray/(boost::container::vector + stack_allocator) total time comparison: wall = 0.645263 user = 0.726702 system = 0.00977199 (user+system) = 0.645018 varray/std::vector total time comparison: wall = 0.972537 user = 1.18754 system = 0.00918836 (user+system) = 0.972308 --------------------------------------------------------------------------------------------------------------------------- Linux gcc_4_7_2 -O2 Core2Duo P7350 2GHz, 3GB Ram --------------------------------------------------------------------------------------------------------------------------- set stack size limit to: 9000000 bytes N = 1000 varray benchmark: construction took 27.677216s wall, 27.560000s user + 0.050000s system = 27.610000s CPU (99.8%) sort took 5.934591s wall, 6.080000s user + 0.010000s system = 6.090000s CPU (102.6%) rotate took 1.720024s wall, 1.590000s user + 0.000000s system = 1.590000s CPU (92.4%) destruction took 0.002077s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) done Total time = 35.350653s wall, 35.240000s user + 0.060000s system = 35.300000s CPU (99.9%) boost::container::vector benchmark construction took 38.196097s wall, 32.530000s user + 5.830000s system = 38.360000s CPU (100.4%) sort took 12.107587s wall, 12.070000s user + 0.010000s system = 12.080000s CPU (99.8%) rotate took 0.011146s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 1.054511s wall, 0.680000s user + 0.190000s system = 0.870000s CPU (82.5%) done Total time = 51.383338s wall, 45.290000s user + 6.030000s system = 51.320000s CPU (99.9%) boost::container::vector benchmark with stack_allocator construction took 37.421275s wall, 31.590000s user + 5.840000s system = 37.430000s CPU (100.0%) sort took 12.009866s wall, 11.930000s user + 0.020000s system = 11.950000s CPU (99.5%) rotate took 0.013459s wall, 0.010000s user + 0.000000s system = 0.010000s CPU (74.3%) destruction took 1.053435s wall, 0.770000s user + 0.300000s system = 1.070000s CPU (101.6%) done Total time = 50.511806s wall, 44.300000s user + 6.170000s system = 50.470000s CPU (99.9%) std::vector benchmark construction took 31.849724s wall, 25.920000s user + 5.760000s system = 31.680000s CPU (99.5%) sort took 3.244223s wall, 3.240000s user + 0.110000s system = 3.350000s CPU (103.3%) rotate took 0.010454s wall, 0.000000s user + 0.000000s system = 0.000000s CPU (n/a%) destruction took 1.054326s wall, 0.780000s user + 0.280000s system = 1.060000s CPU (100.5%) done Total time = 36.173664s wall, 29.950000s user + 6.150000s system = 36.100000s CPU (99.8%) varray/boost::container::vector total time comparison: wall = 0.687979 user = 0.778097 system = 0.00995025 (user+system) = 0.687841 varray/(boost::container::vector + stack_allocator) total time comparison: wall = 0.699849 user = 0.795485 system = 0.00972447 (user+system) = 0.699425 varray/std::vector total time comparison: wall = 0.977248 user = 1.17663 system = 0.0097561 (user+system) = 0.977839 --------------------------------------------------------------------------------------------------------------------------- Linux clang_3_0_6 -O2 Core2Duo P7350 2GHz, 3GB Ram Additional benchmark of Adam Wulkiewicz’s rtree --------------------------------------------------------------------------------------------------------------------------- ins[s] qB[s] qiwo[s] qn[s] qnP5[s] qnB[s] rem[s] mem[MB] - container::vector 10.8 14.4 6.9 3.1 4.8 6.0 5.7 184.1 - std::vector 10.9 13.5 6.9 3.4 5.2 5.9 5.4 274,8 - varray 10.2 12.6 6.3 2.8 4.5 5.8 4.9 207,1 So in this case varray is slightly faster from both and consumes less memory than std::vector.

6 10

[iostreams] Check zlib/bzip2 availability on configuration checks
by Kohei Takahashi 13 Feb '13

13 Feb '13

Hi all, I attached a patch for preventing build error with the environment which provides no zlib and/or bzip2. This patch detects zlib.h and bzlib.h availability on configuration checks timing. It also should work with any environment variable (e.g. ZLIB_SOURCE,BZIP2_SOURCE, ...). Suggestions and comments welcome. Thanks, Kohei Takahashi

1 0

[mpl] is_lambda_expression bug?
by James Hirschorn 12 Feb '13

12 Feb '13

The compile-time assertion fails in the following program. I have tried with more than one version of boost including 1.53. Is it a bug? Shall I report it? -------------------------------------------------- #include <boost/mpl/assert.hpp> #include <boost/mpl/lambda.hpp> #include <boost/mpl/plus.hpp> namespace mpl = boost::mpl; struct MetafunctionClass { template<typename T1, typename T2> struct apply { typedef typename mpl::plus<T1,T2>::type type; }; }; BOOST_MPL_ASSERT((mpl::is_lambda_expression<MetafunctionClass>)); int main(int argc, char * argv[]) { } --------------------------------------------------

4 12

[test] missing include files
by Vicente J. Botet Escriba 12 Feb '13

12 Feb '13

Hi, I have needed to apply this patch to be able to compile with Boost.Test. Could I commit it? Best, Vicente svn diff boost/test Index: boost/test/tools/detail/print_helper.hpp =================================================================== --- boost/test/tools/detail/print_helper.hpp (revision 82823) +++ boost/test/tools/detail/print_helper.hpp (working copy) @@ -25,6 +25,7 @@ #include <boost/type_traits/is_array.hpp> #include <boost/type_traits/is_function.hpp> #include <boost/type_traits/is_abstract.hpp> +#include <limits> #include <boost/test/detail/suppress_warnings.hpp> Index: boost/test/utils/is_forward_iterable.hpp =================================================================== --- boost/test/utils/is_forward_iterable.hpp (revision 82823) +++ boost/test/utils/is_forward_iterable.hpp (working copy) @@ -33,6 +33,7 @@ // STL #include <utility> +#include <type_traits> #endif //____________________________________________________________________________//

2 2

[odeint] Docs don't build in 1.53.0
by John Maddock 12 Feb '13

12 Feb '13

I've finally got around to building the PDF docs for 1.53, and the the docs for Odeint don't build in 1.53: xslt-xsltproc.windows ..\..\..\..\bin.v2\libs\numeric\odeint\doc\msvc-10.0\debug \auto-index-internal-on\auto-index-verbose-on\auto-index-on\threading-multi\refe rence-xml.doxygen file:///D:/data/boost/branches/1.53.0/bin.v2/libs/numeric/odeint/doc/msvc-10.0/d ebug/auto-index-internal-on/auto-index-verbose-on/auto-index-on/threading-multi/ reference-xml/index.xml:1327: parser error : Opening and ending tag mismatch: co mpound line 334 and doxygenindex </doxygenindex> ^ file:///D:/data/boost/branches/1.53.0/bin.v2/libs/numeric/odeint/doc/msvc-10.0/debug/auto-index-internal-on/auto-index-verbose-on/auto-index-on/threading-multi/reference-xml/index.xml:1328: parser error : Premature end of data in tag compound line 333 ^ file:///D:/data/boost/branches/1.53.0/bin.v2/libs/numeric/odeint/doc/msvc-10.0/debug/auto-index-internal-on/auto-index-verbose-on/auto-index-on/threading-multi/reference-xml/index.xml:1328: parser error : Premature end of data in tag compound line 266 Any ideas? Thanks, John.

3 2

[1.53] PDF doc build
by John Maddock 12 Feb '13

12 Feb '13

There's a preliminary build of the PDF docs for 1.53 here: https://sourceforge.net/projects/boost/files/boost-docs/1.53.0/ Currently missing Odeint, but should otherwise be complete and OK. Regards, John.

1 0

[interprocess] Platform-specific flags for mapped_region
by Andrey Semashev 12 Feb '13

12 Feb '13

Hi, I wanted to try using Linux huge pages in my application, which uses Boost.Interprocess currently for interprocess communication. However there seems to be no way to pass MAP_HUGETLB flag to mmap in the interface of mapped_region. I understand that this is a platform-specific option, so probably it would be best to just add an opaque unsigned int flags parameter to the mapped_region constructor with the default value of 0. Or is there a better way of doing this? I'll create a ticket for this feature, but I wanted to know if the general idea of this extension is welcome.

2 3