- Boost-users - lists.boost.org

[Review] Phoenix review results
by Hartmut Kaiser 05 Oct '08

05 Oct '08

Hi all, the Phoenix review is over and I counted 15 votes, 11 of which vote for acceptance and 4 vote for conditional acceptance. Let me explain. All reviewers stated that the outstanding quality of the library and its documentation fully merit immediate acceptance. There is no direct concern with regard to the library itself. Phoenix V2 is already in the Boost distribution as a Spirit sub-library and has matured for a long time, proving its stability and usability. On the other hand, Phoenix provides functionality already covered by Boost.Bind and Boost.Lambda. It is the general intention to use Phoenix as the development ground for a new unified Boost library in this area. The review discussions revealed quite some details and certain problems which have to be resolved for this merger to happen. Joel has a full list of these details and promised to address all of them before adding Phoenix to the Boost SVN. The library authors made clear from the beginning that the review of the current Version 2 of the library is just a stepping stone towards this goal. At the same time Eric Niebler presented a first version of a Proto based Phoenix rewrite (referenced as Phoenix V3). It is the general consensus to use Phoenix V3 as the basis for further development. At the same time Phoenix V3 exposes the same interface as Phoenix V2 and passes all related tests. This review formally was about Phoenix V2, but largely turned out to be a discussion of the future of Bind/Lambda and how Boost should go forward in this direction. Again, the general consensus here is that a) we need a unified approach to bind, lambda, et.al., and b) Boost users need interface stability and minimal disruption. So here is the review result: The quality of Phoenix warrants immediate acceptance into Boost. OTOH, adding Phoenix to Boost as is introduces yet another library dealing with functional idioms. This possibly increases the confusion around the fact which of the libraries to use, etc. The goal should be to have a clear integration of, and migration path from the existing Boost.Bind and Boost.Lambda libraries before final addition of Phoenix to Boost. We will have a mini review before Phoenix gets merged to SVN to make sure whether the feedback from the v2 review was accommodated. Additionally this review will have to discuss: - the breaking interface changes from v2 - the migration path from boost::bind and lambda to Phoenix - how the interoperability with std::bind is solved (result_of semantics) - C++0x features such as rvalue references and variadic templates - the new extensibility mechanism - unified placeholders and interoperability issues with other Proto-based DSELs (such as Spirit.Qi, Spirit.Karma, and Xpressive) - compile times In general, the review was valuable for what it accomplished. It moved us closer towards a unified Boost library in the domain of functional programming idioms. Thanks to Joel and Dan for the high quality submission and to all participating the discussion for their valuable insights! Regards Hartmut Review Manager

3 2

Re: [Boost-users] [statechart] handling unexpected events
by Robert Mecklenburg 05 Oct '08

05 Oct '08

On 2008-09-18 20:54:11 GMT, Robert wrote: > The only thing I can see is that the unconsumed_event() function is > called with the unexpected event object. What I'd like to do in this > function is: > > 1. Abort my transaction and force the state machine back to the Ready > state. I'm not sure how to do this. I've implemented a hack with an internal event, ResetEvent, that I post. This sort of works, but it executes the exit() member function of the aborted state. I'd rather not, I'm trying to abort the behavior of the current state, not complete it. If there was a way from unconsumed_event() to get the current state object and set a flag in it, I could avoid completing the state behavior with a guard. I can use state_downcast<const T *>() to get a const pointer to my state object, but because it is const I can't set a flag. I cannot seem to get a non-const pointer. > 2. Reprocess the unexpected event if it is a hello or get, discard it > otherwise. I'm not sure how to do this, either. I've discovered that post_event() will enqueue any event for processing after the current react() is finished. Thanks for any insight you can provide, -- Robert

2 1

[statechart] handling unexpected events
by Robert Mecklenburg 05 Oct '08

05 Oct '08

I'm attempting to use StateChart to model a simple socket protocol and I'm having difficulty understanding how to handle protocol errors. I've read the sections on Error handling and two phase exit without gaining much insight. Those sections seem to be concerned with user-written code that cannot complete and thus must throw an exception. The type of error I'm currently concerned with is an unexpected event. For example, in this simple protocol we must first receive a hello before being ready to handle requests. A request returns data and must be followed by a commit, which completes the transaction and returns us to the ready state. (start state) (hello event) -> (ready state) (ready state) (get event) -> (data delivered) (data delivered) (commit event) -> (ready state) My problem is this: how do I detect and recover when I'm in the (data delivered) state and the next event is not a (commit event)? The only thing I can see is that the unconsumed_event() function is called with the unexpected event object. What I'd like to do in this function is: 1. Abort my transaction and force the state machine back to the Ready state. I'm not sure how to do this. 2. Reprocess the unexpected event if it is a hello or get, discard it otherwise. I'm not sure how to do this, either. At a more abstract level, it seems that unconsumed_event() is the only way to deal with an unexpected event and to try to reset the state machine. Is this true? Thanks, -- Robert

2 1

[program_options] Decorating output
by Boris 05 Oct '08

05 Oct '08

When I send options_description to cout I would like to put an empty line between the caption and the options - something like: Allowed options: --help produce help message --compression arg set compression level As : is automatically appended to the caption it's not possible to add \n. Is this what I'm trying to do supported? Another question: Can the argument arg behind the --compression option above be changed to <arg>? I read at http://article.gmane.org/gmane.comp.lib.boost.user/38506 that it's not possible to replace arg with level but maybe putting <> around arg works somehow? Boris

2 1

[BGL] Mapping a source Graph to a target Graph
by Ioannis Nousias 05 Oct '08

05 Oct '08

Hello BGL developers, I'm looking for an algorithm that can map a source graph to a target graph. The special requirement is that the graphs have a set of vertices which are differentiated by type(colour if you like). Thus 'red' source vertices should only map to 'red' target vertices, etc. The mapping should be 1-to-1 (a 'blue' target vertex can accept one 'blue' source vertex). I assume the target graph has enough vertices of each type(colour) to map all vertices of the source graph. The target graph represents a topology, while the source graph is the data-graph to map onto that topology. I've briefly looked at the 'layout' specific algorithms provided by BGL. I thought the 'gursoy_atun_layout'could be used, but then I realised it was not meeting the requirements. The main thing that's missing is the way to differentiate vertices by type. Also I don't understand why the 'topology' is not a BGL graph as well. To put it in other words, the layout algorithms in BGL operate on a homogeneous environment. The source vertices can freely occupy any 'point' or 'position' on the target medium. I'm looking for an algorithm that operates on a heterogeneous environment, where the target 'slots' can only accept source vertices of a given type. Is there any current BGL algorithm that fits that description, or can be easily modified to cover this case? If not, can somebody point me to an efficient mapping algorithm that I could base an implementation on. thanks in advance, Ioannis -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.

1 1

using boost in x64 app
by LeRetha Childress 04 Oct '08

04 Oct '08

I am using Windows Vista and Visual Studio 2005 to create my 64 bit application I used bjam to build my boost libraries with the following command bjam --toolset=msvc-8.0_64 stagedir=. --build-type=complete But when I link it to my library I get many warnings like the following 1>..\..\3rdparty\boost\boost/config/abi_prefix.hpp(19) : warning C4103: 'd:\brokaw.x64\3rdparty\boost\boost\config\abi_prefix.hpp' : alignment changed after including header, may be due to missing #pragma pack(pop) 1>..\..\3rdparty\boost\boost/archive/detail/abi_prefix.hpp(11) : warning C4103: 'd:\brokaw.x64\3rdparty\boost\boost\archive\detail\abi_prefix.hpp' : alignment changed after including header, may be due to missing #pragma pack(pop) 1>..\..\3rdparty\boost\boost/archive/basic_binary_oprimitive.hpp(52) : warning C4103: 'd:\brokaw.x64\3rdparty\boost\boost\archive\basic_binary_oprimitive.hpp' : alignment changed after including header, may be due to missing #pragma pack(pop) The library builds but with 892 of these boost warnings Then when I build the driver for the app I get 219 linker errors like the following 2>overlaylibd.lib(bsVolumeLineStrip.obj) : error LNK2001: unresolved external symbol "public: void __cdecl boost::archive::detail::basic_iarchive::load_object(void *,class boost::archive::detail::basic_iserializer const &)" (?load_object@basic_iarchive@detail@archive@boost@ @QEAAXPEAXAEBVbasic_iserializer@234@@Z) 2>overlaylibd.lib(bsVolumePolygons.obj) : error LNK2019: unresolved external symbol "public: void __cdecl boost::archive::detail::basic_iarchive::load_object(void *,class boost::archive::detail::basic_iserializer const &)" (?load_object@basic_iarchive@detail@archive@boost@ @QEAAXPEAXAEBVbasic_iserializer@234@@Z) referenced in function "private: void __cdecl bs::Volume::Serialize::load<class boost::archive::binary_iarchive>(class boost::archive::binary_iarchive &,unsigned int)" ( ??$load@Vbinary_iarchive@archive@boost@@@Serialize@Volume@bs@@AEAAXAEAVbinary_iarchive@archive@boost@@I@Z ) Most of them are for boost::archive , boost::serilization and boost::signal Can anyone tell me what I'm doing wrong or how to fix these issues Thank you LeRetha Childress

2 1

[Proto] Using custom function tag in the simplified calalble context
by Joel Falcou 04 Oct '08

04 Oct '08

I've defined some custom tag for user specific function like : struct conj_tag {}; I have made the following expression builder using make_expr : template<class X> typename proto::result_of::make_expr< conj_tag, X const &>::type conj( X const& x ) { return proto::make_expr<conj_tag>(boost::cref(x)); } And I match this function using the following grammar rule : struct my_grammar : or_< // some terminal , bp::function< bp::terminal<conj_tag>, my_grammar> > {}; Now, how should I overload operator() of a callable_context to match this special tag ? Thanks in advance -- ___________________________________________ Joel Falcou - Assistant Professor PARALL Team - LRI - Universite Paris Sud XI Tel : (+33)1 69 15 66 35

4 4

[thread] win32 mutex optimization
by Alexander Gutenev 04 Oct '08

04 Oct '08

I'd like a small change to: 1. Increase mutex::lock() performance. 2. Remove auto-link to date_time. a patch attached (For some reason yesterday posting failed, trying to post again) begin 666 mutex_optimize_lock.patch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` end

1 0

How to build only these binaries?
by Sachin Garg 04 Oct '08

04 Oct '08

I want to only build both debug and release statically linked binaries. What are the build options to build just these on windows, linux and mac? To clarify, binaries which would link directly inside my executable (the libs that won't need any dll or so). On windows I am using /MT in my projects, so the libs should also statically link to msvc runtime. When using --build-type=complete it generates 5GB of binaries, which for 32-bit and 64-bit on windows, mac and linux together ends up taking a lot more space than I can afford.

4 6

[Phoenix] Review ended September 30th
by Hartmut Kaiser 03 Oct '08

03 Oct '08

Hi all, just a (last) reminder for all of you still wanting to submit a review. The official review period is over, but since I'm going prepare the review summary this weekend only anyway - you still have a day or two. Regards Hartmut Review Manager

3 2