March 2004 - Boost - lists.preview.boost.org

[OT] AspectC++ 0.8 (Stable)
by Neal Becker 19 Mar '04

19 Mar '04

This may interest some of you: http://freshmeat.net/projects/aspectc/?branch_id=41657&release_id=154775

1 0

Trying to do something like boost::python for Netscapes SpiderMon key JavaScript engine
by Warrick Buchanan 18 Mar '04

18 Mar '04

Hi, As the subject says I'm trying to a boost::python like equivalent for SpiderMonkey (Netscape's JavaScript engine) and I've been trying(!) to look at boost::python for inspiration. I've got to the stage where I can quite happily expose a class like so: // D3D's Vector class just for example Declare<D3DXVECTOR>::Property<0>("x", &D3DXVECTOR3::x); // 0 is 'tinyid' Declare<D3DXVECTOR>::Property<1>("y", &D3DXVECTOR3::y); // 1 is 'tinyid' Declare<D3DXVECTOR>::Property<2>("z", &D3DXVECTOR3::z); // 2 is 'tinyid' Declare<D3DXVECTOR>::Definition("Vector3", jsContext, jsObject, jsPrototype); Just from those four lines you can then use the C++ class in JavaScript without issue ie: var v = new Vector3(); v.x = 4.0f; ...etc... Now I want to expose methods and constructors. Originally I was planning to have a load of templates for each number of arguments a function could possibly take (which obviously isn't ideal), but looking at boost::python it seems it has a more elegant but currently illegible to me way of doing this? Has anyone got any pointers to how this works or comments that would help my understanding? Thanks, Warrick

2 1

mem_fn warnings
by David Abrahams 18 Mar '04

18 Mar '04

I keep getting lots of these: c:\boost\boost\mem_fn.hpp(264) : warning C4180: qualifier applied to function type has no meaning; ignored Would it be possible for the mem_fn header to #pragma push a warning disabler for MSVC? Thanks, -- Dave Abrahams Boost Consulting www.boost-consulting.com

2 4

RE: [boost] Trying to do something like boost::python for Netscap es SpiderMon key JavaScript engine
by Warrick Buchanan 18 Mar '04

18 Mar '04

Aaah and I've just noticed that boost::python uses the boost preprocessor stuff that I think does effectively through macros generate a load of templates for up to a constant N number of templated function arguments - and am I right in that this mad preprocessor stuff it recursively includes itself? eeek(!) -----Original Message----- From: Warrick Buchanan [mailto:Warrick@argonaut.com] Sent: 18 March 2004 10:44 To: 'boost(a)lists.boost.org' Subject: [boost] Trying to do something like boost::python for Netscapes SpiderMon key JavaScript engine Hi, As the subject says I'm trying to a boost::python like equivalent for SpiderMonkey (Netscape's JavaScript engine) and I've been trying(!) to look at boost::python for inspiration. I've got to the stage where I can quite happily expose a class like so: // D3D's Vector class just for example Declare<D3DXVECTOR>::Property<0>("x", &D3DXVECTOR3::x); // 0 is 'tinyid' Declare<D3DXVECTOR>::Property<1>("y", &D3DXVECTOR3::y); // 1 is 'tinyid' Declare<D3DXVECTOR>::Property<2>("z", &D3DXVECTOR3::z); // 2 is 'tinyid' Declare<D3DXVECTOR>::Definition("Vector3", jsContext, jsObject, jsPrototype); Just from those four lines you can then use the C++ class in JavaScript without issue ie: var v = new Vector3(); v.x = 4.0f; ...etc... Now I want to expose methods and constructors. Originally I was planning to have a load of templates for each number of arguments a function could possibly take (which obviously isn't ideal), but looking at boost::python it seems it has a more elegant but currently illegible to me way of doing this? Has anyone got any pointers to how this works or comments that would help my understanding? Thanks, Warrick _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

2 1

Is there any interest in this container?
by Peter Palotas 18 Mar '04

18 Mar '04

I was wondering if there is any interest in a container that is a model of sequence (just like std::vector or std::list) and with an interface matching that of std::vector (maybe with minor differences due to effiency) but with different time complexities? Basically this is a Random Access Container but with the following time complexity guarantees instead: * insert/erase anywhere in the list: O(lg n) * element Access (i.e. [] or at()): O(lg n) * front()/back(): O(1) This would make it better than a list for random accessing elements and better than a vector for inserting/deleting elements in the middle of the list. I am currently working of an implementation of this container and thought I would see if there would be any interest in such a container from the boost community. Sincerely, Peter Palotas

7 18

64-bit IRIX build problems with Boost 1.31.0
by Patrick Hartling 18 Mar '04

18 Mar '04

I am trying to make a 64-bit build of Boost 1.31.0 on IRIX using the MIPSpro Compilers 7.3.1.3m, but I cannot seem to get any of the dynamic libraries to link. The bjam command line I am using is this: bjam -sTOOLS=mipspro "-sBUILD=debug release <runtime-link>static/dynamic <address-model>64" (I have $PYTHON_ROOT pointing to a 64-bit build of Python 2.2.3.) All link actions for dynamic libraries fail thusly: ld64: FATAL 12 : Expecting n64 objects: /usr/lib32//c++init.o is n32. Looking at the output after the failed action, I see -L/usr/lib32 in the link line, and I am pretty sure it should not be there. The complete output of the first failed action is this: "/usr/bin/CC" \ -o "bin/boost/libs/date_time/build/libboost_date_time.so/mipspro/debug/address-model-64/libboost_date_time-mp-d-1_31.so" \ "bin/boost/libs/date_time/build/libboost_date_time.so/mipspro/debug/address-model-64/greg_month.o" "bin/boost/libs/date_time/build/libboost_date_time.so/mipspro/debug/address-model-64/greg_weekday.o" "bin/boost/libs/date_time/build/libboost_date_time.so/mipspro/debug/address-model-64/date_generators.o" \ \ "-DBOOST_DATE_TIME_POSIX_TIME_STD_CONFIG" "-DBOOST_ALL_DYN_LINK=1" \ "-nostdinc" "-LANG:std" "-OPT:Olimit=0" "-OPT:IEEE_NaN_inf=ON" "-no_auto_include" "-64" "-mips3" "-O0" "-g" "-INLINE:none" \ \ \ "-I/usr/include/CC" "-I//usr/include/CC" "-I/mnt/rigby/home13/users/patrick/src/Boost/blah/boost_1_31_0/boost/compatibility/cpp_c_headers" \ "-I/usr/include" "-I//usr/include" "-I/mnt/rigby/home13/users/patrick/src/Boost/blah/boost_1_31_0" \ "-nostdlib" "-shared" \ \ \ \ "-L/usr/lib32/mips3" "-L/usr/lib32/" "-L//usr/lib32/mips3" "-L//usr/lib32/" "-L/usr/lib32/mips3" "-L/usr/lib32/" "-L//usr/lib32/mips3" "-L//usr/lib32/" "-L/usr/lib64/mips3" "-L/usr/lib64/" "-L//usr/lib64/mips3" "-L//usr/lib64/" \ "-lm" All the object files being linked in this case are 64-bit, so I think that the compile action is behaving correctly. Am I forgetting something in my bjam command or otherwise misusing bjam? -Patrick -- Patrick L. Hartling | Research Assistant, VRAC patrick(a)vrac.iastate.edu | 2274 Howe Hall Room 2624 http://www.vrac.iastate.edu/~patrick/ | http://www.vrac.iastate.edu/ PGP: http://wwwkeys.gpg.cz:11371/pks/lookup?op=get&search=0xEBF86398

1 0

VC++ boost.Thread library usage ?
by raja 18 Mar '04

18 Mar '04

Hi, I am getting the following error. I see that it is trying to link to the static library, but I still get this error. Would appreciate if someone can correct my usage. Linking to lib file: lib"boost_thread"-vc71-mt-sgd-1_31.lib c:\Boost\include\boost-1_31\boost\config\auto_link.hpp(269) : fatal error C1189: #error : "Mixing a dll boost library with a static runtime is a really bad idea..." Here is my sample: #define _MSC_VER 1310 #define BOOST_LIB_NAME "boost_thread" #define BOOST_LIB_DIAGNOSTIC "on"// Show library file details. #include <boost/config/auto_link.hpp> #include <boost/thread/thread.hpp> #include <iostream> void hello() { std::cout << "Hello world, I'm a thread!" << std::endl; } int main(int argc, char* argv[]) { boost::thread thrd(&hello); thrd.join(); return 0; }

6 6

compiling error using boost::fsm by Andreas Huber VC7.0
by Ahmed Ali 18 Mar '04

18 Mar '04

The solution was for vc7.1 but i have vc7.0 and i got these errors from contructor.h C:\Boost\include\boost-1_31\boost\fsm\detail\constructor.hpp(42): error C2039: 'context_ptr_type' : is not a member of 'operator``global namespace''' C:\Boost\include\boost-1_31\boost\fsm\detail\constructor.hpp(42): error C2825: 'boost::fsm::detail::outer_constructor<ContextList,OutermostContext>::to_construct::context_ptr_type': cannot form a qualified name C:\Boost\include\boost-1_31\boost\fsm\detail\constructor.hpp(42): fatal error C1507: previous user errors and subsequent error recovery halt further compilation _________________________________________________________________ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail

2 1

compiling error using boost::fsm by Andreas Huber VC7.0
by Ahmed Ali 18 Mar '04

18 Mar '04

The solution was for vc7.1 but i have vc7.0 and i got these errors from contructor.h C:\Boost\include\boost-1_31\boost\fsm\detail\constructor.hpp(42): error C2039: 'context_ptr_type' : is not a member of 'operator``global namespace''' C:\Boost\include\boost-1_31\boost\fsm\detail\constructor.hpp(42): error C2825: 'boost::fsm::detail::outer_constructor<ContextList,OutermostContext>::to_construct::context_ptr_type': cannot form a qualified name C:\Boost\include\boost-1_31\boost\fsm\detail\constructor.hpp(42): fatal error C1507: previous user errors and subsequent error recovery halt further compilation _________________________________________________________________ Protect your PC - get McAfee.com VirusScan Online http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963

1 0

filesystem::path assigment operator question
by Janusz Piwowarski 18 Mar '04

18 Mar '04

Hello There isn't defined path assigment operator, is it intentional? Implementation uses default one. Regards, Janusz

2 2