November 2008 - Boost - lists.preview.boost.org

[utility/swap] boost::swap release update
by Niels Dekker - mail address until 2010-10-10 27 Nov '08

27 Nov '08

As far as I'm concerned, the boost::swap utility is now ready to be used within other Boost libraries, at the release branch. Joseph Gauterin has merged the utility from the trunk to the release branch two weeks ago, and its release regression page is now entirely green: http://www.boost.org/development/tests/release/developer/utility-swap_.html Just to remind you: The boost::swap utility makes it easier to swap two objects of the same type in a generic and efficient way, because it tries to find the most appropriate swap overload by argument-dependent lookup (ADL), whenever applicable. Thereby it works around the issues some compilers appear to have regarding ADL. The boost::swap utility also supports swapping built-in arrays. The utility was developed with help from David Abrahams and Steven Watanabe. See also: http://svn.boost.org/svn/boost/branches/release/libs/utility/swap.html http://svn.boost.org/svn/boost/branches/release/boost/utility/swap.hpp BTW, I've just added a boost::swap overload to the release version of Fernando Cacciola's value_initialized (utility/value_init.hpp), which is actually implemented by calling the boost::swap utility: https://svn.boost.org/trac/boost/changeset/49967 According to the following ticket: https://svn.boost.org/trac/boost/ticket/2243 Unswappably yours, Niels -- Niels Dekker http://www.xs4all.nl/~nd/dekkerware Scientific programmer at LKEB, Leiden University Medical Center

1 0

Boost.Future & review of C++0x accepted libraries implementations
by vicente.botet 27 Nov '08

27 Nov '08

Hi, The Anthony Future + Packaged task proposals have been adopted by the C++ Standard. Is there a sense to review both Futures libraries together? Can we accept the Braddock library if it not conforms to the standard? As the interface and the design rationale for libraries already adopted by the C++ Standard such Chrono seem to be out of the scope of a review, which is the contents expected for such libraries to be adopted? Do we need a different review process? BTW, the threadpool library uses Braddock Future library. Thanks, Vicente

5 14

[regex] Matching space - help needed
by Roman Yakovenko 27 Nov '08

27 Nov '08

Hello. I have some strange problem and have no idea how to solve it or where to look father. I have two regular expressions: (\s) ( ) //<== this one contains single space between brackets. I use match algorithm and as a argument I pass string with single space. In the product, the first regular expression (\s) doesn't work, while the second does. I created stand-alone tester - both expressions work as expected. Using proc file system, I compared environment variables, specifically LANG is set to "en_US.UTF-8". Both executables load same Boost.Regex shared library. I am using GCC 3.3.3, on RedHat. The boost version is 1.34. Searching the web didn't help too. Any help is appreciated. Thank you. -- Roman Yakovenko C++ Python language binding http://www.language-binding.net/

3 8

boost::signals2 performance
by Hansi 27 Nov '08

27 Nov '08

I have made now a lot of tests with the new signals2 library. The only thing because I don't want to use it now, is because it is really slow... I have made now performance tests with the dummy_mutex which is included. I have found out the about 50% of the time that it takes to call 10000000 is inside in slot_call_iterator---->lock_next_callable function. Here the follwing lines takes the 50% of the executing time: lock_type lock(**iter); tracked_ptrs = (*iter)->nolock_grab_tracked_objects(); I hope that one has an idea how it can be made it faster... Best regards Hansjörg

2 2

[interthreads] Library under development
by vicente.botet 26 Nov '08

26 Nov '08

Hi, If the Boost community is interested in the InterThreads library I could work on to complete for submission. InterThreads ------------------- :Author: Vicente J. Botet Escribá :Download: Boost Vault: <http://www.boostpro.com/vault/index.php?action=downloadfile&filename=intert…> Boost Sandbox: https://svn.boost.org/svn/boost/sandbox/interthreads Html doc included only on the Vault :Description: Boost.InterThreads extends Boost.Threads adding some features: * thread decorator: thread_decorator allows to define setup/cleanup functions which will be called only once by thread: setup before the thread function and cleanup at thread exit. * thread specific shared pointer: this is an extension of the thread_specific_ptr providing access to this thread specific context from other threads. As it is shared the stored pointer is a shared_ptr instead of a raw one. * thread keep alive mechanism: this mechanism allows to detect threads that do not prove that they are alive by calling to the keep_alive_point regularly. When a thread is declared dead a user provided function is called, which by default will abort the program. * thread tuple: defines a thread groupe where the number of threads is know statically and the threads are created at construction time. * set_once: a synchonizer that allows to set a variable only once, notifying to the variable value to whatever is waiting for that. * thread_tuple_once: an extension of the boost::thread_tuple which allows to join the thread finishing the first, using for that the set_once synchronizer. * thread_group_once: an extension of the boost::thread_group which allows to join the thread finishing the first, using for that the set_once synchronizer. (thread_decorator and thread_specific_shared_ptr) are based on the original implementation of threadalert written by Roland Schwarz. Best regards, Vicente

1 0

[preprocessor] half-open or closed ranges?
by Eric Niebler 26 Nov '08

26 Nov '08

I've noticed an inconsistency with how the preprocessor library handles integer ranges. Consider this: #define M0(Z,N,DATA) N BOOST_PP_REPEAT_FROM_TO(1,3,M0,~) This treats the range 1,3 as half-open. It generates the sequence "1 2". But consider this: #define BOOST_PP_LOCAL_MACRO(N) N #define BOOST_PP_LOCAL_LIMITS (1,3) #include BOOST_PP_LOCAL_ITERATE() This treats the range 1,3 as closed. It generates the sequence "1 2 3". Why the difference? -- Eric Niebler BoostPro Computing http://www.boostpro.com

2 1

[Circular Buffer] Partial thread-safety
by Lasse Kärkkäinen 26 Nov '08

26 Nov '08

It would be very helpful if this container provided lock-free thread-safe implementation of concurrent reads and writes by two threads (producer-consumer), as such function is often needed where ring buffers are used. A prime example of that being receiving audio from a callback (possibly a signal handler), where the callback must not be blocked in any circumstances (meaning that it cannot wait for a lock). As this is a rather common use case, the realtime audio processing framework JACK actually implements such lock-free ring buffer: http://jackit.sourceforge.net/cgi-bin/lxr/http/source/libjack/ringbuffer.c Is there any reason why Boost.circular_buffer should not also implement this kind of thread-safety?

5 5

accepted libraries for boost
by Hansi 26 Nov '08

26 Nov '08

Hello, I have one question regarding to the accepted libraries for boost. At the moment are a lot of libraries accepted which don't get the way to the released version of boost... is it not possible at least that there is one and only one place where the last version of the library is saved? At the moment it is really hard to find this last version.. Best regards Hansjörg

1 0

[intrusive] constant/non-constant time size lists
by Kevin Sopp 26 Nov '08

26 Nov '08

Hi Ion, have you thought about instantiating two different interfaces for constant and non-constant time size() lists respectively? Instead of instantiating one interface that has differing complexity guarantees and is restricted by the lowest common denominator of what is possible with constant-time size(). Non-constant time size() lists would allow to implement higher level functionality as static member functions or as non-member functions, and some functionality like splicing needs no additional list& parameter. If I had such a list I could make my functions work on iterator ranges without the need to pass the list object along. So what I need is access to the nodes so that I can call circular_list_algorithms directly on these. I don't see an easy way to achieve that goal right now. As far as I can tell I need to use the undocumented list_iterator::pointed_node() function - which is okay, I'm only hesitant because it is undocumented. I'm willing to change my code if that interface ever changes though. Kevin

3 4

[UUID] seed_rng on Windows
by Kaj Björklund 26 Nov '08

26 Nov '08

Hi all, It would be nice if seed_rng used CryptGenRandom (or something similar) on Windows instead of trying to open /dev/urandom. The patch for seed_rng.hpp at: http://www.dynaroad.com/win.diff tries to use CryptGenRandom when BOOST_WINDOWS and _WINDOWS_ (i.e. windows.h has been included before uuid.hpp) are defined. This seems to work on Windows XP SP2 and Visual Studio 2005. Cheers, -kaj

2 1