September 2010 - Boost-users

[any] Is it possible to support polymorphism in boost::any?
by Tang Jiang Jun 07 Sep '10

07 Sep '10

Hello, It seems that boost::any does not support polymorphism in the following sample. class A {}; class B : public A {}; void foo( boost::any& _any ) { try { A* p = any_cast< A* >( _any ); <-- Only succeed when _any contain a pointer to a object that is exactly class A, but not any object that is descendant of class A, like B. do_some_thing( p ); } catch( const bad_any_cast& ) { } } void goo() { A* pa = new B; boost::any a = pa; foo( a ); } Is it possible to make it support this polymorphism usage? Because boost::any is a type-safe replacement of void*, I hope that boost::any has the same polymorphic behavior as void*. Regards, Tang Jiang Jun

5 9

How to store on exception object ?
by Timothy Madden 07 Sep '10

07 Sep '10

Hello In a callback function I need to store and save an exception object for until the callback is finished and the control is returned to my application code. The callback is invoked by the C language code in minizip library from zlib-1.2.3 contrib directory, and I need to throw the exception when control returns to my application invoking minizip functions. This is true for any exception that might get thrown and caught in my callback, so I would like to do that for any exception type. However how can I store a thrown object for later re-throw without the actual exception type ? Is there a way to do that ? Thank you, Timothy Madden

6 14

Re: [Boost-users] [MSM] stack overflow happens when there is a false-guard anonymous transition
by Christophe Henry 06 Sep '10

06 Sep '10

Hi Michael, It's always nice to hear from you and have an interesting UML discussion with an expert! > I hope you are well. You know how I like to dig into this stuff... so > I'll give my thoughts. They are tainted by the the UML predecessor; > however, I think there is good logic and possibly even a specification > argument for my point of view. > > UML 2.2 Superstructure section 15.3.14 [Transition] provides the > associations for: > > trigger : Trigger[0..*] -- I am assuming that no trigger is what > you are referring to as > an anonymous transition. I don't see > that term used in current > specifications anymore. > > guard: Constraint[0..1] -- "A guard is a constraint that provides > a fine-grained control > over the firing of the transition. The > guard is evaluated > when an event occurrence is dispatched > by the state machine. > If the guard is true at that time, the > transition may be > enabled; otherwise, it is disabled. > Guards should be > pure expressions without side effects. > Guard expressions > with side effects are ill formed." > > > I would argue that the definition of the guard constraint requires that > the transition is evaluated when an event is dispatched and is only > evaluated once for that event. Because the transition is disabled if the > guard evaluates false, then the transition cannot be evaluated again > until another event is dispatched. I think your interpretation makes sense. Agreed. > I would also argue that the system is considered stable once evaluation > of an event begins. Guards are not allowed to cause side effects; > therefore, the 'state' of the system when evaluation begins should be > the same as when evaluation has completed if no transition is taken. Agreed again. > Personally, the least surprising thing (imnsho) would be that: > > 1. event is dispatched > 2. internal transitions are evaluated for current state > 3. external transitions are evaluated for current state > 4. hierarchy is followed up and 2/3 repeated > 5. if top is reached without satisfying event ... an error has > occurred and is acted upon as normal This is what is implemented for all events except for what the standard calls "Completion Events" (also called anonymous transitions in an UML book from the Booch/Rumbaugh collection). > I do not think whether a trigger is defined or not for a transition will > affect this behavior. Transitions without a trigger are simply > candidates for any event. The steps outlined above are those used for > transitions with triggers, no? I agree on the first part and disagree on the second. So, I took from the standard: "A completion transition is a transition originating from a state or an exit point but which does not have an explicit trigger, although it may have a guard defined. A completion transition is implicitly triggered by a completion event. In case of a leaf state, a completion event is generated once the entry actions and the internal activities (“do” activities) have been completed. If no actions or activities exist, the completion event is generated upon entering the state. If the state is a composite state or a submachine state, a completion event is generated if either the submachine or the contained region has reached a final state and the state’s internal activities have been completed. This event is the implicit trigger for a completion transition. The completion event is dispatched before any other events in the pool and has no associated parameters. For instance, a completion transition emanating from an orthogonal composite state will be taken automatically as soon as all the orthogonal regions have reached their final state. If multiple completion transitions are defined for a state, then they should have mutually exclusive guard conditions." So you are right and I should handle them like any other event. OTOH, the standard does not see them as a candidate for any event but as an implicit event which I have to trigger after the target state entry behavior. If a transition without an explicit trigger (event) is defined, it fires. And I should generate it only once, thus avoiding the overflow. What you suggest is a kind of an "any" take-all event, which I also find a good idea (I do not remember seeing it in the standard, but a good idea should be implemented if possible). The part of the standard which I find unpracticable is, if the guard rejects the completion event, how do we move out of this case? If I follow the standard, we would need to reenter a state, which means we would need a dummy event (not nice). This is why I suggest a process_completion() which would simply resend a completion event. Ok? Thanks for your (as usual) helpful insights, Christophe PS: reading again, I see that there is another implementation error in the current implementation, completion events have a lower priority than the event queue, I will change this on the way too.

2 1

[asio] Close file descriptors on exec.
by Rok Breulj 06 Sep '10

06 Sep '10

Why does boost.asio not set its file descriptors to close on exec? (in 1.42, but didn't see any changelogs or tickets mentioning it).

1 0

[MSM] stack overflow happens when there is afalse-guard anonymous transition
by Christophe Henry 06 Sep '10

06 Sep '10

>When start() got called in main(), it will enter a endless reentering between state_machine::process_event() and >state_machine::handle_eventless_transitions_helper::process_completion_event(). Finally it caused a stack overflow. > > >Could somebody please help me to fix this problem? I just committed a tentative fix into the trunk: - completion events are only tried if the previous transition was successful. This should prevent the overflow and some border cases. - calling fsm.process_completion() will retry the completion event if the guard returned false. The event is sent only once for every call of process_completion() - the completion events are, standard conforming, tried before the event queues If it works and we all agree this is the right solution, I will keep it. Thanks, Christophe

1 0

A question about "prim_minimum_spanning_tree.hpp"
by 王伟 06 Sep '10

06 Sep '10

Hi, I'm studying MST algorithms in GPU recently.I need BGL implementation as coparison,But I have a problem in using prim_minimum_spanning_tree.hpp.The problem is:when I use large input(e.g. >1M),there's a "segmnetation error" problem.But when the input(e.g. 1K) is small,the result is noproblem. I haven't found know the reason yet.My complier is gcc 4.1.2,OS is RHEL5. Help would be appreciated. Thanks in advance. Wang wei The complete code is in the attachment.

3 4

[MSM] stack overflow happens when there is afalse-guard anonymous transition
by Christophe Henry 06 Sep '10

06 Sep '10

<snip> >When start() got called in main(), it will enter a endless reentering between state_machine::process_event() and >state_machine::handle_eventless_transitions_helper::process_completion_event(). Finally it caused a stack overflow. > > >Could somebody please help me to fix this problem? Hi, The problem is that UML does not specify how often an anonymous transition should be "tried" and I probably chose the wrong solution (to try again and again). So, as your guard always returns false, msm keeps trying, which means calls to process_event() until the stack overflows. The solution probably is to try just once but it will force you to manually try to reprocess by calling, say a process() function without event, but it arguably simply means passing the buck to you. Please let me think about it a bit more, I'd like to think this out a bit more. For the moment, I suggest you to either write a useful guard or revert to a named event, which you will process when it makes sense in your application. Regards, Christophe

2 1

How to work with large files in boost ?
by Timothy Madden 06 Sep '10

06 Sep '10

Hello Can I open and work with 64bit files (over 2Gb) in boost ? I see iostreams library has a stream<file_descriptor>off_type of long, which is 32 bit ... Thank you, Timothy Madden

1 0

Can I create and open a file only if it does not exist ?
by Timothy Madden 05 Sep '10

05 Sep '10

Hello I see file_descriptor_sink uses the openmode flags from std::ios_base, and also file class actually calls the open function standard C++ basic_filebuf. I would like to open a temporary file (generated with unique_path) only if it does not exist, as if by POSIX open() with O_CREAT | E_EXCL, and no combination of ios_base::openmode flags allows for that. Is there a way in boost to create and open a file only if it does not exist ? Thank you, Timothy Madden

1 0

combining mpl sequences
by Karsten Ahnert 05 Sep '10

05 Sep '10

Hi, I have a templatized mpl-vector template< class T > class containers : public mpl::vector< std::vector< T > , std::list< T > , std::deque< T > > { }; and a mpl-vector holding some types: typedef mpl::vector< float , double , long double >::type my_types; Is there a simple way to build a mpl vector which contains the elements of the first vector with template arguments of the second vector, hence mpl:vector< std::vector< float > , std::vector< double > , std::vector< long double > , std::list< float > , std::list< double > , std::list< long double > , std::deque< float > , std::deque< double > , std::deque< long double > > I can do this manually with mpl::insert_range typedef mpl::insert_range< mpl::vector0<> , mpl::end< mpl::vector0<> >::type , containers< mpl::at< my_types , mpl::int_<0> >::type > >::type first; typedef mpl::insert_range< first , mpl::end< first >::type , containers< mpl::at< my_types , mpl::int_<1> >::type > >::type second; typedef mpl::insert_range< second , mpl::end< second >::type , containers< mpl::at< my_types , mpl::int_<2> >::type > >::type all_containers; But this way is not very elegant. Does some has any hints how to do this automatically? Best regards, Karsten

3 4