- Boost-users - lists.boost.org

VMD library accepted
by Steven Watanabe 27 Feb '15

27 Feb '15

AMDG The VMD library submitted by Edward Diener is ACCEPTED into Boost. I apologize for the long delay for this announcement. Thanks to everyone who participated in the formal review * Nial Douglass * Bjorn Reese * Rodrigo Madera * Zhou Zhenghui * Matt Calabrese * Paul Mensonides Reviews for the library were generally positive. The library was considered to be a useful set of tools for macro programmers. Paul expressed concern about the way the the library handles v-sequences and v-identifiers. After some discussion, Edward has agreed that the interface could be significantly improved. Work Items: * Add examples and improved motivation in the documentation * Update the v-identifier/v-sequence interface based on Paul's comments * Fix inaccurate statements about Boost.Preprocessor per Paul In Christ, Steven Watanabe

2 1

[Regex] Cannot build Regex without ICU
by Rob Desbois 27 Feb '15

27 Feb '15

I want to build the Boost libs without ICU support in Regex, but have been unable to prevent it from being linked in. I've run the following commands on a freshly extracted distribution of 1.57: ./bootstrap.sh --without-icu --prefix=/home/rob/workspace/deps/boost/1.57.0/linux/x64/ ./b2 install The bootstrap stage output includes this, which seems to indicate it's behaving correctly: Unicode/ICU support for Boost.Regex?... disabled. The resulting project-config.jam makes no mention of ICU. The build log includes: - has_icu builds : yes - icu : yes The shared library dependencies for the resulting libboost_regex.so.1.57.0 then include: libicuuc.so.50 => /lib64/libicuuc.so.50 (0x00007f74e5e3b000) libicui18n.so.50 => /lib64/libicui18n.so.50 (0x00007f74e5a3b000) libicudata.so.50 => /lib64/libicudata.so.50 (0x00007f74e4467000) I know the docs for Boost.Regex specify the bootstrap argument as --disable-icu, however this is rejected as invalid, unlike the argument I provide. Have I messed up somewhere or is this a bug? TIA -- rob -- Rob Desbois http://theotherbranch.wordpress.com/

2 2

Re: [Boost-users] [ASIO] io_service::service::shutdown_service
by chmhou＠sina.com 26 Feb '15

26 Feb '15

>Hi, >Let us consider following execution sequence: >1. "io_service" destructor calls "io_service::service::shutdown_service"; >2. documentation of "io_service::service::shutdown_service" says that it >should destroy all user defined handler objects. >Is it true that user defined handler objects may access to (e.g. post a >handler) that "io_service" being destroyed and therefore has not exist >already? >Could You please also clarify me what does "user defined handler object" >really mean? >Thanks, >Andrew. Hi, Andrew For your 2nd question "user defined handler object", I supposed that the parameter for the io_service::post() which is kinda "user defined handler object". 1st, it could be a handler like "bind(&fun, this, _1, _2)". Martin

1 0

Why does read_xml call fail if my property tree contains char16_t?
by ashish g 25 Feb '15

25 Feb '15

Hi I have a property tree which has strings of char16_t as both key and value datatypes. When I try to call read_xml() to load an xml file into this tree it fails. The same code however works with wchar_t. Seems like its failing on parse_bom() function. The code and error message can be seen here: http://coliru.stacked-crooked.com/a/3f922b43838fd7e8 Any tips on how to fix this issue. -- *Ashish Gudla*

2 1

[fusion] Why is the std::tuple adapter undocumented?
by Rob Desbois 24 Feb '15

24 Feb '15

What it says on the tin really: Why is boost/fusion/adapted/std_tuple.hpp not documented at all? I'd like to use it but if it's undocumented because it's considered experimental then I'll use fusion's vector instead... TIA -- rob -- Rob Desbois http://theotherbranch.wordpress.com/

3 2

Re: [Boost-users] Performance optimization in Boost using std::vector<>
by chmhou＠sina.com 24 Feb '15

24 Feb '15

Hi I got it, it supposed that I made a mistake on the duration. I will follow this question. Martin

1 0

Re: [Boost-users] Performance optimization in Boost using std::vector<>
by chmhou＠sina.com 23 Feb '15

23 Feb '15

>Hi, >not sure if the OP needs std::vector but... I'd recommend >boost::container::vector which has a dedicated constructor [1] and >resize() [2] method tagged with default_init_t argument, both of which >default initialize the values in the vector. For primitives it basically >means it leaves them uninitialized, hence there's no overhead when the >vector is to be filled with real data soon. >WBR, >Adam Romanek Hi Adam As following your point, the benefit of boost::container::vector supported the "default_init_t" method so that it 's quite convince and safety comparing with std::vector<> is that right ? are there additional advantages ? Hi Ilja You said "Boost::mpi version is always at least 10 times slower. " make me confused, I guess it's should be a typo, isn't ? Martin

3 2

Signals2 benchmark
by Joren Heit 23 Feb '15

23 Feb '15

Hi all, New to this list, so I hope I'm at the right place, and not asking too much of you guys. Please inform me if either of those is true! ;-) I have been working on a little toy project to implement a template-based Signal/Slot library and now that I have a usable end-product, I want to test it against boost::signals2 to get an idea of how well it performs. Therefore my first question is whether boost::signals2 is generally considered a high performance library. Second, I'm a little reluctant to build a benchmark suite myself, because I want to be fair to both myself and the other party. I don't know the signals2 library at all, let alone what its virtues and shortcomings are. Some tips on what to test exactly would therefore be great, although I'm afraid this request might qualify for "asking too much". Initial tests show that my own library can be up to 10 times faster than the boost-implementation, which makes me even more suspicious that I might not be testing the right things (but at the same time prematurely proud). Thanks in advance and cheers! Joren

13 42

[ASIO] io_service::service::shutdown_service
by Andrew C 23 Feb '15

23 Feb '15

Hi, Let us consider following execution sequence: 1. "io_service" destructor calls "io_service::service::shutdown_service"; 2. documentation of "io_service::service::shutdown_service" says that it should destroy all user defined handler objects. Is it true that user defined handler objects may access to (e.g. post a handler) that "io_service" being destroyed and therefore has not exist already? Could You please also clarify me what does "user defined handler object" really mean? Thanks, Andrew.

1 0

Re: [Boost-users] Strange boost::threads behaviour
by Gavin Lambert 20 Feb '15

20 Feb '15

On 19/02/2015 21:11, Lorenzo Trivelli wrote: >> >> Typically the way things work is that as long as a lock (or >> whatever it is you're using -- you didn't actually say) goes >> uncontended, or minimally contended such that a spin resolves it, >> then it will allocate no kernel handles. A kernel handle is only >> required when the lock is contended for sufficiently long that it >> needs to put the thread to sleep. > > thank you ver much for your response, > I'll try to detail the issue better. > The original software is an conccurrent transactional server which > preallocates a certain number of ftherads (say 100 ~ 130) and then > accepts and serves TCP/IP connections in a request-reply fashion. While thread-per-connection is a popular model for older or small-scale software (because it's easy to program), performance suffers quite significantly when a large number of connections are in use, especially when each connection needs to contend on the same locks. Especially if the workload is mostly I/O bound rather than CPU bound, you may want to consider a "properly" async design (eg. using Boost.Asio) where #threads == #cores (or thereabouts) and connection processing can shift between threads as needed. > Obviuosly, it heavily uses boost::shared_ptr to handle these connections, > but in the end I managed to use other shared pointers (for testing purposes only) > and the boost::thread object alone and a single shared static mutex object > > so, the sample program starrts, creates some threads which simply does some memory > allocations loops and s. In the main function I enclosed all the code between brackets, > and I print the handle count before and after, ans it's not the same, > something like > int main() { > GetProcessHandleCount().... > > { > [creates threads and waits for their termination through ->join] > > } > GetProcessHandleCount().... > > } > > The difference between the first and the last count is not always the same, it > depends on how much concccurent threads you create and how much iterations of the > same loop each thread does. Well, as I said that is not an error if some synchronization objects (eg. mutexes) still exist.

1 0