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

RE: [Boost-Users] shared_ptr and lazy copy
by Michael Davis 30 May '02

30 May '02

There is an implementation of the method you described (const and non-const versions of operator-> and operator*) at the following website: http://ootips.org/yonat/4dev/ Perhaps this guy might be talked into submitting it. -----Original Message----- From: Mark Storer To: 'Boost-Users(a)yahoogroups.com' Sent: 5/30/2002 1:35 PM Subject: [Boost-Users] shared_ptr and lazy copy Is there an easier way to use shared_ptr as a lazy copy method? It seems to me that the only way to do it would be by testing unique() (or use_count()) every time the pointed-to object was to be changed: boost::shared_ptr<foo> bar(new foo); boost::shared_ptr<foo> baz = bar; if(!baz->unique()) ... // make a deep copy baz->ChangeAFoo(); I can't think of a good way to alter shared_ptr either. The only thing I came up with would be const and non-const versions of operator-> and operator* where the non-const versions checked unique() and made a deep copy. This would seem to fall into that last question in the FAQ anyway... "Why doesn't shared_ptr provide (your pet feature here)". I'd think this could easily lead to unnecessary deep copies unless the programmer took explicit steps (lots of const_cast's) to avoid it. Am I missing anything? --Mark Storer Software Engineer Cardiff Software #include <disclaimer> typedef std::disclaimer<Cardiff> Discard; Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

1 0

shared_ptr and lazy copy
by Mark Storer 30 May '02

30 May '02

Is there an easier way to use shared_ptr as a lazy copy method? It seems to me that the only way to do it would be by testing unique() (or use_count()) every time the pointed-to object was to be changed: boost::shared_ptr<foo> bar(new foo); boost::shared_ptr<foo> baz = bar; if(!baz->unique()) ... // make a deep copy baz->ChangeAFoo(); I can't think of a good way to alter shared_ptr either. The only thing I came up with would be const and non-const versions of operator-> and operator* where the non-const versions checked unique() and made a deep copy. This would seem to fall into that last question in the FAQ anyway... "Why doesn't shared_ptr provide (your pet feature here)". I'd think this could easily lead to unnecessary deep copies unless the programmer took explicit steps (lots of const_cast's) to avoid it. Am I missing anything? --Mark Storer Software Engineer Cardiff Software #include <disclaimer> typedef std::disclaimer<Cardiff> Discard;

1 0

Problem with Boost.Thread and HP aCC
by raulh39 29 May '02

29 May '02

We are trying to compile Boost.Thread with HP aCC (with -AA option). In boost/function/function_base.hpp the compiler give us this error: Error 20: "function_base.hpp", line 37 # '<type name>' expected before '>'. template<bool> struct truth {}; We are using Boost 1.28.0. The line 37 in function_base.hpp has: template<bool> struct truth {}; We have found two ways to make it work: 1.- Comment out the line !! 2.- Replace with: template<bool a> struct truth {}; We think that the compiler is wrong but, is ok to comment this line? In earlier versions of this file this line doesn't exists, and it is never used, so it is necesary? Regards, Raúl.

2 1

RE: [Boost-Users] Problem with Boost.Thread and HP aCC
by Tom Matelich 29 May '02

29 May '02

When I first started trying to support aCC with Boost that was a common issue. Most of them were ferreted out. I don't think anybody is actively looking at that compiler, perhaps Jens Maurer still is. Anyway, the proper solution is to name the template parameter, your option 2. I've cc'ed Bill Kempf in case he doesn't follow this list. Tom Matelich -----Original Message----- From: raulh39 [mailto:rhuertas@tcpsi.es] Sent: Wednesday, May 29, 2002 10:15 AM To: Boost-Users(a)yahoogroups.com Subject: [Boost-Users] Problem with Boost.Thread and HP aCC We are trying to compile Boost.Thread with HP aCC (with -AA option). In boost/function/function_base.hpp the compiler give us this error: Error 20: "function_base.hpp", line 37 # '<type name>' expected before '>'. template<bool> struct truth {}; We are using Boost 1.28.0. The line 37 in function_base.hpp has: template<bool> struct truth {}; We have found two ways to make it work: 1.- Comment out the line !! 2.- Replace with: template<bool a> struct truth {}; We think that the compiler is wrong but, is ok to comment this line? In earlier versions of this file this line doesn't exists, and it is never used, so it is necesary? Regards, Raúl. Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

1 0

VC7, threads, and threads starting other threads
by Bill Eidson 28 May '02

28 May '02

Hi all, I am using boost 1.28 and the threads library on windows with VC7.Net. I am creating threads in a threadgroup use function objects. Inside of these threads, I am creating other threads again using function objects. The problem is at run time the debug environment is informing me of a call type collision indicating the return address/stack is set up incorrectly. The debug message: "The value of ESP was not properly saved across afunction call. This is usually a result of calling a function declared with one calling convention with a function pointer declared with a different calling convention." this is in the declaration template<typename Functor> void assign_to(Functor f) { this->assign_to(f,tag()); } The error is at the end of the function First, is it legal in boost::thread to start one thread in another Second, If it is, what is going on with the function_template.hpp code here Thanks bille

1 0

problem with c_regex_traits.cpp
by René Gutschmidt 28 May '02

28 May '02

Hi, i have problems with the file c_regex_traits.cpp. I work under CodeWarrior 6/7 with a Macintosh. These problem ist with boost 1.25.1,1.26. template <class charT> unsigned int BOOST_REGEX_CALL re_get_message(charT* buf, unsigned int len, unsigned int id) { unsigned int size = _re_get_message((char*)0, 0, id); if(len < size) return size; boost::scoped_array<char> cb(new char[size]); _re_get_message(cb.get(), size, id); size = boost::c_regex_traits<wchar_t>::strwiden(buf, len, cb.get()); return size; } inline unsigned int BOOST_REGEX_CALL re_get_message(char* buf, unsigned int len, unsigned int id) { return _re_get_message(buf, len, id); } Is that a known problem? Do you know a workaround? (Renaming one of the 're_get_message()' functions does the job, but I'm not sure about the implications.) Regards, Rene

2 2

any_ptr
by cl_corba_at 27 May '02

27 May '02

Hi all! any_ptr was a typo in an older documentation, but this leads me to this question: any stores the type values in the template class holder, which is derived from placeholder to provide an uniform interface. This approach works fine for new classes where the data representation is done inside the any class, but for legacy code I would also like to have an any_ptr, which uses a holder that POINTS to a variable rather than containing it. With this I could extend existing classes with algorithms depending on any. Has any_ptr considered in the past and if yes, what have been the reasons to decide against it? Thanks in advance, CL

1 0

BGL, 6 degrees of Kevin Bacon and MSVC 6 compile error
by a_schweitzer 25 May '02

25 May '02

Hello, I am trying to build the "6 degrees of kevin bacon" example of the BGL in MSVC 6. When I do I get an "internal compile error". Does anyone know a fix/work-around? The compilation gets to this line (about #73): typedef adjacency_list < vecS, vecS, undirectedS, property<vertex_name_t,std::string >, property < edge_name_t,std::string > > Graph; Graph g; If Graph g is declared I get the following ugly error: d:\program files\microsoft visual studio\vc98\include\xmemory(59) : fatal error C1001: INTERNAL COMPILER ERROR (compiler file 'msc1.cpp', line 1794) Please choose the Technical Support command on the Visual C++ Help menu, or open the Technical Support help file for more information Note that if I change the std::string in vertex_name property to an int the compilation proceeds. Thanks for any help Andy Schweitzer

3 3

RE: [Boost-Users] boost::lexical_cast string to string problem - New proposition available
by Terje Slettebø 25 May '02

25 May '02

>--- In Boost-Users@y..., Tom Matelich <tmatelich@z...> wrote: > I and many others have requested fixes for this for a very long time. > Basically, it involves creating class templates to do the actual conversion. > It's not the most complicated system, just nobody has done it yet. If you > search the boost list archives, you'll see a lot of mail about this. There > is no quick solution to your problem though. I've searched the archive, now, and see that there's been quite a few propositions for changes, but apparently none of them have ended up in the lexical_cast implementation. There's also a couple of propositions, here (http://groups.yahoo.com/group/boost/files/lexical_cast_propositions/) Some of the proposed changes are: - wlexical_cast - A version that can use wide characters - basic_lexical_cast - A version that takes the interpreter (stringstream) as a parameter, to support things like changing locale - Being able to set number base (dec, oct or hex). This would be implicitly possible using basic_lexical_cast - Being able to set precision for numerical conversions. This may also be done with basic_lexical_cast - Checking for change of sign, when using unsigned numbers. This is addressed using an integer adapter, in the above lexical_cast_propositions One problem with doing these changes, after having talked with Kevlin Henney a while ago (without knowing it had been already proposed, I proposed an extra parameter, for setting the number base). is that he prefers to not have extra parameters in the interface, as it then will no longer look like a cast, and it may be less generic, depending on the way it's done. For example, a base-setting argument wouldn't make much sense if you converted from floating point to string. In any case, I have actually just now made a version of lexical_cast, that fixes the problem pointed out by OP, here, and I've made it available in the Boost files section (http://groups.yahoo.com/group/boost/files/lexical_cast_proposition/) I've also put a unit test for it, there. Feedback is welcome. It doesn't change the interface in any way, so it should be ok. The new version supports conversion between string and char, including char containing whitespace, and when the string is empty, or containing only whitespace. If the source or destination (or both) is basic_string or char/wchar_t, it will set the interpreter to use the char type used, so it implicitely supports wide characters. Wide character support requires partial specialisation if one of the operands is not either basic_string or char/wchar_t. In the case where the source is basic_string, it also uses pass by const reference, for efficiency. It's tested on Intel C++ 6.0, MSVC 6.0 and BCC 5.6. Regards, Terje

2 4

boost::shared_ptr and atomic_count_linux.hpp
by Jerry Napoli 24 May '02

24 May '02

Hi, I'm using boost version 1.27.0 on an x86 linux platform and have just upgraded my kernel to 2.4.18. After the upgrade, boost::shared_ptr fails to compile because functions such as "atomic_inc_and_test", "atomic_dec_and_test", etc.. which were previously defined in "atomic.h" are no longer defined there. I can get around the problem by using an older "atomic.h" or by using pthreads, but I was wondering why those functions were removed from the header. Any ideas? Thanks, Jerry jnapoli(a)iteamsolutions.com [Non-text portions of this message have been removed]

3 2