- Boost-users - lists.preview.boost.org

RE: [Boost-Users] Re: Why scoped_ptr dost provide custom dealloca tion?
by bjorn.karlsson＠readsoft.com 14 Feb '03

14 Feb '03

> From: Peter Dimov [mailto:pdimov@mmltd.net] [snip] > (An example of a convincing case: "In my project I have > encountered the > following situation: <description with real(istic sounding) > identifiers> and > I believe that scoped_ptr<X, D> would be a perfect fit." Some > non-examples: > "Someone might find it useful." "Wouldn't it be cool if...") In my project I have encountered the following situation: A lot of the code needs to handle (no pun intended) resources that are represented by indiscriminate types, such as void* or long. Different resources (and different functions for allocation/deallocation) use this handle type, so specialization is not an option. It's possible to use shared_ptr to handle lifetime management of the resource: HANDLE hPrinter=0; OpenPrinter(_T("Some printer"),&hPrinter,0); if (boost::shared_ptr<void> p(hPrinter,&ClosePrinter)) { } While this works, it doesn't convey the intent of the code (nor preserve it): the resource should typically not be shared, only scoped, and I believe that scoped_ptr<X,D> would be a perfect fit. There are other tools in Boost (outside too, of course) for solving this type of problem, but I think it would be more than cool for scoped_ptr to have this feature; I'd definitely use it in real code. Bjorn Karlsson

2 1

[BGL] Strange result with internal graph properties
by yg-boost-users＠m.gmane.org 13 Feb '03

13 Feb '03

I have the following code in one of my header files: struct vertex_type_t { typedef boost::vertex_property_tag kind; }; enum vtype { ROOT, INNER, LEAF }; typedef boost::property<vertex_type_t, enum vtype> vertex_type; typedef boost::property<boost::vertex_index_t, int, vertex_type> vertex_index; typedef boost::property<boost::vertex_name_t, std::string, vertex_type> vertex_property; typedef boost::adjacency_list<boost::vecS, boost::vecS, boost::directedS, vertex_property, edge_property> tree; Notice the error: The vertex properties are not chained together. Rather, the actual vertex property that is added to the graph points to vertex_type, and the vertex_index /also/ points to the vertex_type! Now the strange thing: I have been running this and using all of the vertex properties, including vertex_index, without any problems, but when I noticed this, and tried setting the last template parameter of vertex_property to vertex_index instead I get segmentation faults. I have no idea why this is. Björn

1 1

regex_match() Division By Zero
by Glover Barker 13 Feb '03

13 Feb '03

I am using boost::regex (1.29.0) under MSVC++ 6.0. Whenever I call regex_match(), I get: "Unhandled exception in filename.exe: 0xC0000094: Integer Divide by Zero." I have tried various ways to make the call (including the snippets posted on www.boost.org) Even this simple example fails: regex expression("abc"); cmatch what; if ( regex_match("xyz", what, expression) ) { // Whatever ... } Breakpoints reveal that after expression is set, the value of _expression_len is still 0. So when regex_match() is called, it in turn sends the value of expression.size() (which is just _expression_len) to re_detail::_priv_match_data(), which sends this value to estimate_max_state_count(). Then estimate_max_state_counts(), which calls this value "states", does the following: states *= states; // and later ... if(dist > (difference_type)(lim / states)) Since states is 0, this last expression causes division by 0, and the resulting exception. In case my environment matters, I am building my program under MSVC++ 6.0, running with boost_regex.dll. I am defining BOOST_REGEX_NO_LIB and BOOST_REGEX_USE_CPP_LOCALE, and linking in boost_regex.lib. What am I doing wrong? Glover Barker

2 1

Compile problem with regex.hpp and VS6SP5
by soulofsealion <dworthem＠symyx.com> 13 Feb '03

13 Feb '03

When I included regex.hpp into my project: #include <boost\regex.hpp> I received the following errors c:\cvsprojectsroot\renaissance\3rdpartyfiles\boostlibrary\boost\regex. hpp(860) : error C2899: typename cannot be used outside a template declaration c:\cvsprojectsroot\renaissance\3rdpartyfiles\boostlibrary\boost\regex. hpp(440) : error C2899: typename cannot be used outside a template declaration So at lines 440-2 template <class iterator, class Allocator = #if !(defined(BOOST_MSVC) && (BOOST_MSVC <= 1300)) BOOST_DEFAULT_ALLOCATOR(typename re_detail::def_alloc_param_traits<iterator>::type) > I Tried the following: //This section was altered to handle the problem of the VS6 compiler not handling of a typename in a macro for a template parameter-DWW 02/06/2003 template<class iterator> class def_alloc_param_traits_fixclass { public: typedef typename re_detail::def_alloc_param_traits<iterator>::type def_alloc_param_traits_fix; }; template <class iterator, class Allocator = #if !(defined(BOOST_MSVC) && (BOOST_MSVC <= 1300)) //BOOST_DEFAULT_ALLOCATOR(typename re_detail::def_alloc_param_traits<iterator>::type) > BOOST_DEFAULT_ALLOCATOR (def_alloc_param_traits_fixclass<iterator>::def_alloc_param_traits_fix ) > //End of section altered to handle the problem of the VS6 compiler not handling of a typename in a macro for a template parameter-DWW 02/06/2003 And at lines 858-862 namespace re_detail{ template <class iterator, class Allocator = BOOST_DEFAULT_ALLOCATOR (typename def_alloc_param_traits<iterator>::type) > class match_results_base { I tried: namespace re_detail{ //This section was altered to handle the problem of the VS6 compiler not handling of a typename in a macro for a template parameter-DWW 02/06/2003 template<class iterator> class def_alloc_fixclass { public: typedef typename def_alloc_param_traits<iterator>::type def_alloc_fix; }; //template <class iterator, class Allocator = BOOST_DEFAULT_ALLOCATOR (typename def_alloc_param_traits<iterator>::type) > template <class iterator, class Allocator = BOOST_DEFAULT_ALLOCATOR(def_alloc_fixclass<iterator>::def_alloc_fix) > //End of section altered to handle the problem of the VS6 compiler not handling of a typename in a macro for a template parameter-DWW 02/06/2003 class match_results_base { This compiled OK, has any one else had this problem and any comments on this fix or is there anything else that I missed so I do nto have to change the regex.hpp file? it appeared to me that the VS6 compile has a problem with a typename as a parameter in a macro that is in turn a template parameter. Thank you for any help.

3 2

Ah.......Boost Socket.....
by hjlee＠hurim.net 12 Feb '03

12 Feb '03

Anyone who knows when boost socket library is comming and where I can get beta version..? [Non-text portions of this message have been removed]

2 1

boost::bind/function and C function pointers
by Thomas Pollinger 12 Feb '03

12 Feb '03

Apologies if this topic has already been discussed. I have looked around on various mailing lists, in addition I have tried to make it work with the current C++ STL standard with std::mem_fun and std::bind1st but with no success. The problem is basically to bind a member function pointer with an instance of its declaring class and to assign the result to an existing C library function that expects a C style function pointer. This certainly works for ordinary C functions and static C++ functions, however fails for bound member function pointers. Here is what I am trying to achieve: #include <boost/bind.hpp> // C function declared in some library struct E; extern void setCallBack(void (*callback_fct)(E*)); class A { public: void init(); void callBack(E *e); .... }; void A::init() {setCallBack(boost::bind(&A::callBack, this));} void A::callBack(E *e) {...} I tried also using the STL calls, but without success either: ... setCallBack(std::bind1st(std::mem_fun(&A::callBack), this)); I like the elegant approach of boost::bind and would like to use it for this case. The workaround would be to declare a static function and to store the instance somewhere to access the other members of the A instance - but I'd prefer the other way if there is a solution to make it work. BTW, I am using gcc v 3.2 on Linux/i386 - but I guess that this is a more general problem than a specific compiler issue. The error I've gotten is /opt/src/boost/boost/bind.hpp: In instantiation of `boost::_bi::result_traits<boost::_bi::unspecified, void (A::*)(E*)>': /opt/src/boost/boost/bind/bind_template.hpp:16: instantiated from `boost::_bi::bind_t<boost::_bi::unspecified, void (A::*)(E*), boost::_bi::list1<boost::_bi::value<A*> > >' controller.cpp:28: instantiated from here /opt/src/boost/boost/bind.hpp:59: `void (A::*)(E*)' is not a class, struct, or union type Chaning boost::bind(...) to boost::bind<void>(...) to explicitly delclare the return type yields: controller.cpp:28: cannot convert `boost::_bi::bind_t<void, void (A::*)(E*), boost::_bi::list1<boost::_bi::value<A*> > >' to `void (*)(E*)' for argument `1' to `void setCallBack(void (*)(E*))' which probably reveals the root of the problem, that of not being able to cast a boost function pointer type to a regular C function type. I tried applying some cast operators, but that failed already in the syntactical stage of the parser, e.g. static_cast<void (*)(E*)>. -Thomas

2 5

up-to-date documentation dor boost.python
by Pierre Barbier de Reuille 12 Feb '03

12 Feb '03

Hi ! I'm totally new with boost.python and I need to use some tool to wrap C++ code into python interpreter. The problem I have is that, apparently, the tutorial given with boost.python does not reflect the current implementation. Is there any way to have an up to date tutorial or an organized list of the modifications I can refer to when looking at the tutorial ? Thanks, -- Pierre Barbier de Reuille INRA - UMR Cirad/Inra/Cnrs/Univ.MontpellierII AMAP Botanique et Bio-informatique de l'Architecture des Plantes TA40/PSII, Boulevard de la Lironde 34398 MONTPELLIER CEDEX 5, France tel : (33) 4 67 61 65 77 fax : (33) 4 67 61 56 68

1 0

RE: Re: [Boost-Users] Re: building threads lib with latest from C VS
by William E. Kempf 11 Feb '03

11 Feb '03

> From: Mark Storer <mstorer(a)cardiff.com> > > >The MS APIs do not give you any means with which to cleanup data > >allocated to TLS, and they recommend you add this in DllMain. > > What can you do in DllMain that you can't do with atexit, or some Loki-like > priority singleton system? Handle destruction of TLS data _at thread termination_. The atexit approach would only clean up the data at program termination, which isn't good enough. From what I recall of Loki, it won't apply here either. William E. Kempf wekempf(a)cox.net

1 0

RE: Re: [Boost-Users] Re: building threads lib with latest from C VS
by Mark Storer 11 Feb '03

11 Feb '03

>The MS APIs do not give you any means with which to cleanup data >allocated to TLS, and they recommend you add this in DllMain. What can you do in DllMain that you can't do with atexit, or some Loki-like priority singleton system? --Mark Storer Software Engineer Cardiff Software #include <disclaimer> typedef std::disclaimer<Cardiff> Discard;

2 1

Re: Re: [Boost-Users] Re: building threads lib with latest from CVS
by William E. Kempf 11 Feb '03

11 Feb '03

> From: Michael Hunley <mhunley(a)pocketpurchase.com> > At 09:56 AM 2/11/2003 -0500, you wrote: > >Sorry, I forget that some acronyms aren't known by everyone. TLS stands > >for "thread local storage", also known as TSS or "thread specific storage". > > > > Thanks for clearing that up. > > >Testing and development. I compile against 5 compilers, and with both > >static and dynamic libraries this means compiling 10 variants at a minimum > >(through in both debug and release and you have 20 variants), which is > >time consuming. But the real kicker is that boost::thread itself is going > >to rely on TLS shortly, making it so you'll always need a DLL on the > >Windows platforms. > > Understand & sympathize. What about people who want to bundle that part > into their own DLL (which is largely what I'm doing)? Does it hurt to > leave the build in and doc it as untested? Short term, probably not. Long term even this won't work. > Also, why would inclusion of > TLS require it to be a DLL? By itself, TLS doesn't require this. However, TLS cleanup *does* require this. (We're still trying to find an alternative, because even this approach to TLS cleanup is fraught with peril.) The MS APIs do not give you any means with which to cleanup data allocated to TLS, and they recommend you add this in DllMain. This is why you have incompatible thread creation routines such as CreateThread, _beginthread(ex) and AfxBeginThread. > It would seem like including TLS could also be > done in a static lib or am I being completely obtuse? TLS, yes, TLS cleanup, no. > On an almost related note, I saw mention of a cooperative cancel policy > from another post of yours. Do you have any ETA on that and any info on > what it will support vs what it won't support? Is that info on the > developer mailing list? It's partially implemented in CVS on a branch named "thread_dev" (though this isn't stable). ETA is hopefully very soon, as the C++ standards meeting looms. William E. Kempf wekempf(a)cox.net

1 0