Mailman 3 September 2006 - Cplusplus-sig

Re: [C++-sig] Stuck on accessing numpy.ndarray in c extension
by Qinfeng(Javen) Shi Sept. 25, 2006

Sept. 25, 2006

No, I did not use raw_function for the export.I tried hello.testArray((c,)), it crashed. Then I used PyArray_Check(args) to check the input type, it crashed even when I passed a numpy.ndarray to it. void testArray(PyObject * args) { if(!(PyArray_Check(args))) { printf("args is not list type!\n"); return; } } >>> a = numpy.arange(25).reshape((5,5)); >>> rarray2 = hello.testArray(a,) Segmentation fault >>> a = numpy.arange(25).reshape((5,5)); >>> rarray2 = hello.testArray(a) Segmentation fault A *segmentation* *fault* occurs when a program attempts to access a memory location that it is not allowed to access, or attempts to access a memory location in a way that is not allowed. It seems my PyObject * args can not neither be acessed by PyArray_Check() nor by PyArg_ParseTuple(). I don't get it. -- Qinfeng(Javen) Shi Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

1 0

Py++ - problem with split_module and return_opaque_pointer-policy
by Ganssauge, Gottfried Sept. 25, 2006

Sept. 25, 2006

The return_value_policy<return_opaque_pointer> requires an additional macro call (BOOST_PYTHON_OPAQUE_SPECIALIZED_TYPE_ID) in the declaration section of a module. This can be added e.g. using module_builder_t.add_declaration_code(), and it works as long as you're generating the code using write_module(). In the moment you're using "split_module" to split your output, this scheme breaks, because BOOST_PYTHON_OPAQUE_SPECIALIZED_TYPE_ID is required in every translation unit using the opaque pointer type whose type_id is defined with that macro. Instead it is only generated into the main module. An example: test.hpp: ========= typedef struct opaque_ *opaque_pointer; opaque_pointer get_opaque(); test.py: ======== import os from pyplusplus import module_builder from pyplusplus.decl_wrappers import \ return_value_policy, \ return_opaque_pointer mb = module_builder.module_builder_t( ["test.hpp"] ) mb.add_declaration_code ("BOOST_PYTHON_OPAQUE_SPECIALIZED_TYPE_ID(opaque_)") mb.free_function (name = "get_opaque").call_policies = \ return_value_policy (return_opaque_pointer) mb.build_code_creator( module_name='test_opaque' ) mb.split_module( '.' ) When compiled with g++ -c -I/usr/include/python2.4 test_opaque*cpp this produces the following sermon: boost/python/type_id.hpp: In function `boost::python::type_info boost::python::type_id() [with T = opaque_]': boost/python/lvalue_from_pytype.hpp:23: instantiated from `boost::python::type_info boost::python::detail::extractor_type_id(T&(*)(U)) [with T = opaque_, U = boost::python::opaque_pointer_converter<opaque_*>::instance&]' boost/python/lvalue_from_pytype.hpp:88: instantiated from `boost::python::lvalue_from_pytype<Extractor, python_type>::lvalue_from_pytype() [with Extractor = boost::python::opaque_pointer_converter<opaque_*>, const PyTypeObject*python_type = ((const PyTypeObject*)(&boost::python::opaque_pointer_converter<opaque_*>::type_ object))]' boost/python/opaque_pointer_converter.hpp:116: instantiated from `void boost::python::opaque_pointer_converter<Pointer>::register_self() [with Pointer = opaque_*]' boost/python/opaque_pointer_converter.hpp:70: instantiated from `boost::python::opaque_pointer_converter<Pointer>::opaque_pointer_conver ter(const char*) [with Pointer = opaque_*]' boost/python/return_opaque_pointer.hpp:25: instantiated from `PyObject* boost::python::detail::opaque_conversion_holder<Pointer>::operator()(Poi nter) const [with Pointer = opaque_*]' boost/python/detail/invoke.hpp:75: instantiated from `PyObject* boost::python::detail::invoke(boost::python::detail::invoke_tag_< false, false>, const RC&, F&) [with RC = boost::python::detail::opaque_conversion_holder<opaque_*>, F = opaque_*(*)()]' boost/python/detail/caller.hpp:199: instantiated from `PyObject* boost::python::detail::caller_arity<0u>::impl<F, Policies, Sig>::operator()(PyObject*, PyObject*) [with F = opaque_*(*)(), Policies = boost::python::return_value_policy<boost::python::return_opaque_pointer, boost::python::default_call_policies>, Sig = boost::mpl::vector1<opaque_*>]' boost/python/object/py_function.hpp:38: instantiated from `PyObject* boost::python::objects::caller_py_function_impl<Caller>::operator()(PyOb ject*, PyObject*) [with Caller = boost::python::detail::caller<opaque_*(*)(), boost::python::return_value_policy<boost::python::return_opaque_pointer, boost::python::default_call_policies>, boost::mpl::vector1<opaque_*> >]' test_opaque_free_functions.pypp.cpp:21: instantiated from here boost/python/type_id.hpp:71: error: invalid use of undefined type `struct opaque_' test.hpp:1: error: forward declaration of `struct opaque_' Cheers, Gottfried

2 7

Stuck on accessing numpy.ndarray in c extension
by Qinfeng(Javen) Shi Sept. 25, 2006

Sept. 25, 2006

Dear *All *I got stuck on it on wrapping c++ code which need accept numpy.ndarray as argument. The numpy.version I am using is '1.0rc1'.* *(My previous question is too massy, so I revised it. Sorry if I borthered you!) I used two ways to access numpy.ndarray, but they all reported error. ----------------------------------------------------------------------------------- 1st way: void testArray(PyArrayObject * args) { } >>> hello.testArray(b) Traceback (most recent call last): File "<stdin>", line 1, in ? Boost.Python.ArgumentError: Python argument types in hello.testArray(numpy.ndarray) did not match C++ signature: testArray(PyArrayObject*) >>> ------------------------------------------------------------------------------------ 2nd way: void testArray(PyObject * args) { PyObject *input; PyArrayObject *array; if (!PyArg_ParseTuple(args, "O", &input)) printf("getting input failed!\n"); else printf("getting input sucess!\n"); } >>> import hello >>> import numpy >>> c = numpy.arange(25).reshape((5,5)) >>> d = hello.testArray(c); getting input failed! >>> d SystemError: new style getargs format but argument is not a tuple >>> Any suggestion will be appreciated. Thanks! -- Qinfeng(Javen) Shi Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

2 1

Pass the numpy.ndarray to c extension
by Qinfeng(Javen) Shi Sept. 25, 2006

Sept. 25, 2006

Thank for Dave Abrahams's suggestion to "numpy.ndarray did not match PyArrayObject *" problem. I didn't use the lvalue converter before. I will look up how to do it. Since what I really want to do is to pass the numpy.ndarray to c extension, so I started a new title. I find a way to pass the argument from the webpage: http://numpy.scipy.org/numpydoc/numpy-13.html . It uses PyArg_ParseTuple to transfer PyObject into PyArrayObject. I implemented two functions with PyArg_ParseTuple(), but they all reported error. /***************c extension*********/ void testArray(PyObject * args) { PyObject *input; PyArrayObject *array; if (!PyArg_ParseTuple(args, "O", &input)) printf("getting input failed!\n"); else printf("getting input sucess!\n"); array = (PyArrayObject *) PyArray_ContiguousFromObject(input, PyArray_INT, 0, 3); if (array == NULL) printf("getting array failed!\n"); else printf("getting array sucess!\n"); } static PyObject * trace(PyObject *self, PyObject *args) { PyObject *input; PyArrayObject *array; double sum; int i, n; if (!PyArg_ParseTuple(args, "O", &input)) return NULL; array = (PyArrayObject *) PyArray_ContiguousFromObject(input, PyArray_DOUBLE, 2, 2); if (array == NULL) return NULL; n = array->dimensions[0]; if (n > array->dimensions[1]) n = array->dimensions[1]; sum = 0.; for (i = 0; i < n; i++) sum += *(double *)(array->data + i*array->strides[0] + i*array->strides[1]); Py_DECREF(array); return PyFloat_FromDouble(sum); } /************error report************/ >>> import hello >>> import numpy >>> c = numpy.arange(25).reshape((5,5)) >>> print c [[ 0 1 2 3 4] [ 5 6 7 8 9] [10 11 12 13 14] [15 16 17 18 19] [20 21 22 23 24]] >>> b = hello.trace(c,c) >>> b SystemError: new style getargs format but argument is not a tuple >>> d = hello.testArray(c); getting input failed! >>> d SystemError: new style getargs format but argument is not a tuple >>> -- Qinfeng(Javen) Shi Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

1 0

Opportunities at Naughty Dog...
by CP Sept. 25, 2006

Sept. 25, 2006

Dear Mr. Summers, I hope to become one of Naughty Dogs employees one day sadly I am not old enought yet, but before you delete this message like your company did when I e-mailed them 1-2 months back just listen to my question and get back to me. I want to know if there is going to be another Jak game coming because the game in development "Big" just doesn't look like it is for me. So is there goning to be another Jak game? That is all I want to know. When I reach college I hope to become a content manager like Sony's own Gene Strocco. I want to study at mabe USC and go on to work at Nauhty Dog after I get my degree. Thank you for listening, Chris

1 0

Opportunities at Naughty Dog...
by CP Sept. 25, 2006

Sept. 25, 2006

Dear Mr. Summers, I hope to become one of Naughty Dogs employees one day sadly I am not old enought yet, but before you delete this message like your company did when I e-mailed them 1-2 months back just listen to my question and get back to me. I want to know if there is going to be another Jak game coming because the game in development "Big" just doesn't look like it is for me. So is there goning to be another Jak game? That is all I want to know. When I reach college I hope to become a content manager like Sony's own Gene Strocco. I want to study at mabe USC and go on to work at Nauhty Dog after I get my degree. Thank you for listening, Chris

1 0

numpy.ndarray did not match PyArrayObject *?
by Qinfeng(Javen) Shi Sept. 23, 2006

Sept. 23, 2006

Dear All, I wrapped a c++ code which need accept numpy.ndarray as argument. Here numpy.version='1.0rc1'. And to the best of my knowledge, numpy.ndarray's c type is PyArrayObject. But why when I run it, it reports type did not match? running result as following: >>> b = numpy.array([1]) >>> b array([1]) >>> hello.testArray(b) Traceback (most recent call last): File "<stdin>", line 1, in ? Boost.Python.ArgumentError: Python argument types in hello.testArray(numpy.ndarray) did not match C++ signature: testArray(PyArrayObject*) >>> Any suggestion? Thanks! -- Qinfeng(Javen) Shi Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

2 1

(no subject)
by Eilif Mueller Sept. 21, 2006

Sept. 21, 2006

Hi All, maybe this is obvious to most, but... the following code ***************** BOOST_PYTHON_MODULE(atest) { using namespace numeric; using namespace boost::python; import_array(); array::set_module_and_type("numpy", "ArrayType"); ... ******************** works with numpy-0.9.5 but not with numpy-1.0b5 or up since it seems the 'ArrayType' attribute has been removed from the namespace. While the module will build and import, the following messages occur when using functions that return arrays or have arrays as arguements: *************** ImportError: No module named 'numpy' or its type 'ArrayType' did not follow the NumPy protocol *************** It is sufficient to replace the above line with array::set_module_and_type("numpy", "ndarray"); cheers, Eilif -- NEU: GMX DSL Sofort-Start-Set - blitzschnell ins Internet! Echte DSL-Flatrate ab 0,- Euro* http://www.gmx.net/de/go/dsl

1 0

memory leaks using the "manage_new_object" return policy...
by Kevin Jones Sept. 21, 2006

Sept. 21, 2006

Hi, I am using the manage_new_object return policy to wrap "newed" C++ object in an effort to conserve memory. Even though I am deleting the returned object in Python and invoking the Python garbage collector I still have a leak. This is a long-running systems adminstration app the read the contents of a very large database. Any suggestions on how to make manage_new_object behave correctly? Thanks in advance, Kevin __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

4 11

Complete confustion about using Boost.Python
by Noah Schwartz Sept. 20, 2006

Sept. 20, 2006

Hi, I've recently built Boost based on the instructions at http://boost.org/more/getting_started.html. It found my python installation OK and so as far as I know that means Boost.Python has been installed on my system. The issue I am having is that I can't seem to figure out how to build any modules that are not within the directory tree of the source I extracted from sourceforge. If I copy the exact tutorial out of libs/python/example/tutorial and try to build it elsewhere, it complains about not being able to find a project-root.jam file. Also, I have noticed that if I set the BOOST_ROOT path in say, for example, the Jamrules file in the that tutorial directory to where boost is now actually installed (/usr/local/include/boost-1_33_1), I can no longer build. What am I missing here? Right now it seems like I can only build projects in the directory tree of the source I extracted from the website. Thanks, Noah

4 14

[solved]custom smart pointers
by Roman Yakovenko Sept. 20, 2006

Sept. 20, 2006

Hi. Few weeks ago I asked for help to solve problem with custom smart pointer. David Abrahams helped me to solve all the problems I had. He pointed me to the bugs in my code. Yes! There was no problem with Boost.Python, but with my code. I attach to this mail few files: * smart_ptr.h - contains definition of my smart pointer * classes.h - contains definition of base( abstract ) and derived classes * bindings.cpp - contains code that exposes base and derived classes to Python * test.py - test suite, defines 2 classes, one that derived from base and another one that derives from "derived" class * sconstruct - scons build script I hope you( Boost.Python user) will be able to learn from the code and also to use it to test your own smart pointer. You can find the source code in Py++ repository: http://svn.sourceforge.net/viewvc/pygccxml/pyplusplus_dev/unittests/smart_p… Comments, suggestions, new tests or just thoughts are welcome. -- Roman Yakovenko C++ Python language binding http://www.language-binding.net/

1 0

Smart Pointers
by Lakin Wecker Sept. 20, 2006

Sept. 20, 2006

David Abrahams & boost.python/py++ contributors, I just wanted to thank you for the time and effort you put forth helping Roman with the custom smart pointers. It has really helped the python-ogre project. For those of us involved, it has saved us quite a bit of time. I also just wanted to say kudos to everyone involved with Boost.python and with py++. I'm very happy with the ease that it allows wrapping C++ code, and it keeps getting better everyday. Cheers! Lakin

1 0

custom smart pointers
by Roman Yakovenko Sept. 20, 2006

Sept. 20, 2006

Hi. I am working on some project that defines smart pointer and uses it allover the place. I have a problem for some reason I am not able to call functions that takes the smart pointer instantiation as argument. The error I get is ArgumentError: ArgumentError: Python argument types in custom_smart_ptr_classes.ref_get_value(add_x_t) did not match C++ signature: ref_get_value(my_smart_ptr_t<controllers::controller_i> {lvalue} a) Code high level: template<T> class my_smart_ptr_t{...} struct controller_i{ virtual int get_value() const = 0; }; struct add_x_t : controller_i{ int m_value; virtual int get_value( ){ return m_value + get_add_value(); } virtual int get_add_value(){ return 0; } }; int get_value( my_smar_ptr_t<controller_i> c ){ return c->get_value(); } In order to export this code, I wrote next code: namespace boost{ namespace python{ controller_i* get_pointer( my_smart_ptr_t< controller_i > const& p ) { return p.get(); } template <> struct pointee< my_smart_ptr_t< controller_i > > { typedef controller_i type; }; //same code for add_x_t class ... }} Because both classes contains virtual function I create class-wrapper for every class. Classes and function registration is pretty simple. After this I add next registration code: register_ptr_to_python< my_smart_ptr_t< controller_i > >(); register_ptr_to_python< my_smart_ptr_t< add_x_t > >(); implicitly_convertible< my_smart_ptr_t< add_x_t >, my_smart_ptr_t< controller_i > >(); Python code: import my_module add_x = my_module.add_x_t() my_module.get_value( add_x ) # raises the error I described earlier. I think, I exposed my classes "by the book". May be I am missing something? Thanks in advance. -- Roman Yakovenko C++ Python language binding http://www.language-binding.net/

3 15

Boost.Python and threads
by thor.arne.johansen＠ibas.no Sept. 19, 2006

Sept. 19, 2006

What is the status for threads and Boost.Python. I wish to have multiple (Python) threads. Each thread owns an instance of a wrapped C++ object doing extensive disk IO. Will Boost.Python release the GIL when executing the C++ code? If I make a C++ wrapper which releases the GIL before executing the code, will that be safe? (Can I be sure the internals of Boost.Python does not call into Python Interpreter, without holding the GIL?) Note I'm extending, and not embedding, so I will only have one Python interpreter with multiple threads, each running C++ code wrapped with Boost.Python. -- Thor Arne Johansen R&D Director Ibas AS

2 1

Exception 'No to_python (by-value) converter found - 'garbage collection'
by Eduardo Picado Sept. 19, 2006

Sept. 19, 2006

Hi, Im writing the following kind of python code using a C++ wrapped library: system = TASRSystem.getASRSystem() system.setParameter(TASRSystem.Option.asrServersListenPort, 32000) try: system.initialize(1, 3000) except Exception, e: sys.stderr.write("[E] : " + e.__class__.__name__ + str(e) ) print "ok" Running it I get: ok Exception exceptions.TypeError: 'No to_python (by-value) converter found for C++ type: class TASRSystem' in 'garbage collection' ignored Fatal Python error: unexpected exception during garbage collection The declaration of TASRSystem is like: BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS(TASRSystem_initialize_overloads_0_2, initialize, 0, 2) class_< TASRSystem, boost::noncopyable >("TASRSystem", no_init) .def("getASRSystem", &TASRSystem::getASRSystem, return_value_policy<reference_existing_object>()) .def("initialize", &TASRSystem::initialize, TASRSystem_initialize_overloads_0_2()) .staticmethod("getASRSystem") - TASRSystem.getASRSystem() is an static method, that creates and returns an instance of an ASRSystem. Seems ok. I've tried to use with_custodian_and_ward_postcall <1, 0>, to ensure that the lifetime of the created instance is as long as that of the Python object, but it didn't worked - gives "argument index out of range". - setParameter() seems good, as I can check with a .getParameter() call. - system.initialize() It doesn't return anything, I think it just tries to find/connect to remote ressources based on the parameters. It seems to work, the "ok" is printed but after it gives the "No to_python (by-value) converter found"... associated with the Python garbage collection... Some ideas? Thanks very much, Eduardo -- Eduardo PICADO MSN eduardo.picado(a)gmail.com portable +33 6 32 47 10 10 (NEW) fix +33 1 45 03 52 54 |̲̅<̲̅Θ̲̅>̲̅|

2 4

Assert fail during deletion for an obj containing external object ptr
by Gangadhar NPK Sept. 19, 2006

Sept. 19, 2006

All, I am a beginner for Boost.Python, so please be kind on the responses. I am trying to create python wrappers for an existing library. I can't change the source of the library ( I don't even have the src to change). Also, since most of the classes are via factories, I can't try to derive from the base classes, so am providing the wrappers by containership. The python wrappers will contain the pointers to the 3rd party libs. When I do this, I am able to get the interface to the third party library's base interface, but during destruction (during the interpretor exit), I get this assert failure: --------------------------- Microsoft Visual C++ Debug Library --------------------------- Debug Assertion Failed! Program: c:\Python24\python.exe File: dbgdel.cpp Line: 51 Expression: _BLOCK_TYPE_IS_VALID(pHead->nBlockUse) For information on how your program can cause an assertion failure, see the Visual C++ documentation on asserts. (Press Retry to debug the application) --------------------------- Abort Retry Ignore --------------------------- I understand that there might be a double delete or invalid deletion that is happening, but I can't point exactly where. The code that generates is attached as a CPP file. This must be something basic that I am missing, but I can't figure out from the boost.python documentation or the wiki. Can someone please point at what I am doing wrong? Thank You Gangadhar P.S:Also, ignore the fact that the 3rd party lib's static function is not exported in the Wrapper class as a member function. That is something that will be changed later.

3 7

implicitly_convertible or order of init's problem
by Roman Yakovenko Sept. 19, 2006

Sept. 19, 2006

Hi. I found some very very strange behaviour. Please, consider next code: struct vector{ vector(){} vector( double ){} vector( const vector& ){} }; struct float_vector{ float_vector(){} float_vector( const float_vector& ){} float_vector( const vector& ){} float_vector( float ){} }; namespace bp = boost::python; BOOST_PYTHON_MODULE( ic_ext ){ bp::class_< float_vector >( "float_vector" ) .def( bp::init< >() ) .def( bp::init< vector const & >() ) .def( bp::init< float >() ) .def( bp::init< const float_vector& >() ); bp::implicitly_convertible< vector const &, float_vector >(); bp::implicitly_convertible< float, float_vector >(); bp::class_< vector >( "vector" ) .def( bp::init< >() ) .def( bp::init< double >() ); bp::implicitly_convertible< double, vector >(); } import ic_ext ic_ext.float_vector( 5.0 ) The last line fails with the error: TypeError: No registered converter was able to extract a C++ reference to type vector from this Python object of type float Question: Why Boost.Python does not use float_vector( float ) constructor? If I reorder the init methods "ic_ext.float_vector( 5.0 )" expression pass: bp::class_< float_vector >( "float_vector" ) .def( bp::init< >() ) .def( bp::init< vector const & >() ) .def( bp::init< const float_vector& >() ) .def( bp::init< float >() ); If I comment out "bp::implicitly_convertible< vector const &, float_vector >();" than "ic_ext.float_vector( 5.0 )" expression pass. It could be nice to understand why this happens. I attached small example that reproduce the problem.( I hope this time I did it right :-) ) Thanks. -- Roman Yakovenko C++ Python language binding http://www.language-binding.net/

2 2

how to pass list to my c++ function which need an argument of array
by Qinfeng(Javen) Shi Sept. 19, 2006

Sept. 19, 2006

Dear all, I wrapped a c++ function into python by pyste. It is like: func1(double[] a); main() { double a[]={1,2,3}; func1(a); } I can run main in python. But I what directly call func1 in python as well. Python doesn't have a type called double[], but list(or numarray) has the similar property to double[]. So if I can pass list to func1, it will be fine. Does anybody know how can I pass list to func1 in python? Any suggestion is appreciated. Javen. Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

2 1

Returning a Python object to C++ (is get_owner the right way?)
by Jack Lloyd Sept. 19, 2006

Sept. 19, 2006

I have a scenario where I'm wrapping a virtual base class that offers cloning (here, sending a clone message to an object means "return a newly constructed object of whatever type you are"), and want to inherit from that type in Python. Simply implementing that function in Python (eg: class MySubclass(mylib.MyBaseClass): def clone(self): return MySubclass() led to "returning a dangling reference" errors, so I'm using the technique Dave suggested in http://mail.python.org/pipermail/c++-sig/2004-December/008379.html (basically, use get_owner, get ahold of the __class__ attribute, and call the constructor, then keep a reference to the copy-constructed Python object around so it is not destroyed until the lifetime of the C++ object is over, all held inside another class that simply forwards the appropriate messages to the Python object). This has been working fine for me, however since get_owner is still undocumented (AFAICT; couldn't find it anywhere in the 1.33.1 docs), and generally the technique led to some rather ugly code, I wanted to check and see if any new technique/method for dealing with this situation had developed since that discussion? I hope I've managed to explain what I'm doing clearly; if not, I'd be happy to post a minimal example of my method. Thanks, Jack

2 2

Problems Compiling Boost.Python w/ Indexing Suite v2
by Lakin Wecker Sept. 19, 2006

Sept. 19, 2006

Hi all, I've been having problems getting boost.python from CVS with the v2 indexing suite. I used the instructions from here: http://www.boost.org/more/getting_started.html#CVS To get boost.python from cvs. Then I used the instructions from here: http://mail.python.org/pipermail/c++-sig/2003-October/005802.html to switch to v2 of the indexing suite. Finally, I used http://www.boost.org/libs/python/doc/building.html as instructions to build boost.python. Specifically, I executed "bjam -sTOOLS=gcc test" from the libs/python/test directory. The following is the log of the output from the above build command: http://lakin.weckers.net/v2_log.txt I'm running Ubuntu 6.06, which has g++ (GCC) 4.0.3 (Ubuntu 4.0.3-1ubuntu5) Thanks in advance, Lakin

3 2

What if a Python override returns several values?
by Matthias Baas Sept. 19, 2006

Sept. 19, 2006

Hi, how can I obtain the result tuple from calling a Python override that returns several values? The following example is from the tutorial and works fine as long as the Python implementation of f returns an integer. struct BaseWrap : Base, wrapper<Base> { int f() { return this->get_override("f")(); } }; But if I have another virtual member function "void g(int& a, int &b)" which returns two integers, then I'd like the Python version to take no arguments and return a tuple (a,b). But how does the wrapper look like in this case? I would have expected something like: void g(int& a, int& b) { boost::python::tuple tup = this->get_override("g")(); a = ...extract an int from tup[0]... b = ...extract an int from tup[1]... } But when I run this code I get the following error at runtime: TypeError: No registered converter was able to extract a C++ reference to type boost::python::tuple from this Python object of type tuple So how is this dealt with using the "wrapper" class? (This is Boost.Python v1.33.1 on Linux) - Matthias -

3 11

Fwd: Problems Compiling Boost.Python w/ Indexing Suite v2
by Lakin Wecker Sept. 19, 2006

Sept. 19, 2006

Posting this back to the c++-sig mailing list, where it belongs. :) Lakin ---------- Forwarded message ---------- From: Lakin Wecker <lakin.wecker(a)gmail.com> Date: Sep 18, 2006 10:30 AM Subject: Re: Problems Compiling Boost.Python w/ Indexing Suite v2 To: python-ogre-developers(a)googlegroups.com I assume you meant boost/python/operators.hpp ... as that is the only file where I could find that line. I commented it out, did a bjam clean, and then rebuilt using the same command. Still no luck, but different problems this time: http://lakin.weckers.net/v2_log.2.txt At this point, I haven't had time to look into the problem, so I hope it isn't something simple. Lakin On 9/17/06, Roman Yakovenko <roman.yakovenko(a)gmail.com> wrote: > > On 9/17/06, Lakin Wecker <lakin.wecker(a)gmail.com> wrote: > > > > Hi all, > > > > I've been having problems getting boost.python from CVS with the v2 > > indexing suite. I used the instructions from here: > > http://www.boost.org/more/getting_started.html#CVS To get boost.python > > from cvs. Then I used the instructions from here: > > http://mail.python.org/pipermail/c++-sig/2003-October/005802.html to > > switch to v2 of the indexing suite. Finally, I used > > http://www.boost.org/libs/python/doc/building.html as instructions to > > build boost.python. Specifically, I executed "bjam -sTOOLS=gcc test" > > from the libs/python/test directory. > > > > The following is the log of the output from the above build command: > > http://lakin.weckers.net/v2_log.txt > > > > I'm running Ubuntu 6.06, which has g++ (GCC) 4.0.3 (Ubuntu 4.0.3-1ubuntu5) > > > > It has nothing to do with indexing suite v2. Just comment > BOOST_PYTHON_UNARY_OPERATOR(repr, lexical_cast<std::string>, repr) > line in operators.hpp in you will be okey. I found this yesterday too. > > > Thanks in advance, > > > > Lakin > > > -- > Roman Yakovenko > C++ Python language binding > http://www.language-binding.net/ > > --~--~---------~--~----~------------~-------~--~----~ > You received this message because you are subscribed to the Google Groups "Python Ogre Developers" group. > To post to this group, send email to python-ogre-developers(a)googlegroups.com > To unsubscribe from this group, send email to python-ogre-developers-unsubscribe(a)googlegroups.com > For more options, visit this group at http://groups.google.com/group/python-ogre-developers > -~----------~----~----~----~------~----~------~--~--- > >

1 0

range and iterator
by CT Sept. 18, 2006

Sept. 18, 2006

Hi , Does someone can give me some help on how to expose correctly an array of pointers using boost.python ? here is a small example of my code i have a class class A { public: A(std::string &oName); ~A(){} const std::string &GetName() const; } class B { std::vector<*A> *poNames public: std::vector<*A> GetNames() { return *poNames } } assume that there is two instances of A in the container of the object B now for the boost::python code. I would like to be able to expose class a and class b and the vector of A* typedef std::vector<A*> array_proj; array_proj::iterator begin_array_proj(array_proj &x) {return x.begin();} array_proj::iterator end_array_proj(array_proj &x) {return x.end();} BOOST_PYTHON_MODULE(pysession) { class_<A>("object a",init<std::string &) .def("GetName",&A::GetName,return_value_policy <copy_const_reference>() ) ; class_<B>("object B") .def("GetNames", &B::GetNames, return_value_policy<manage_new_object>() ) ; class_<std::vector<*A> >("array of pointer to objects A") .def("__iter__", range( begin_array_proj, end_array_proj) ) ; } Now in python for example I call #create a new object B #and iterate over all of the object A #assume that there are two object of type A in the vector oNewSession = pysession.B() oVector = oNewSession.GetNames() for oA in oVector.__iter__(): print (oA.GetName()) i have this error when i start the python script for oA in oVector.__iter__(): TypeError: No to_python (by-value) converter found for C++ type: class ::A * Does someone can give me some help on how to expose correctly an array of pointers using boost.python ?

1 0

how to use PyRun_SimpleFile in a multithread C++ program?
by jack833 Sept. 18, 2006

Sept. 18, 2006

Hi,Every one, i need help about PyRun_simplefile function,the program i'm writing is a C++ program which will start serveral threads .each thread will use PyRun_simplefile function to do something,but python script file is the same. now the program is offten crash down,but there is no problem while only thread which use PyRun_simplefile,but that is not what i wanted. is there anyone who can tell me how to resove it. -- View this message in context: http://www.nabble.com/how-to-use-PyRun_SimpleFile-in-a--multithread-C%2B%2B… Sent from the Python - c++-sig mailing list archive at Nabble.com.

2 1

how to pass function pointer to a constructor of a template class when wrap it by pyste
by Qinfeng(Javen) Shi Sept. 14, 2006

Sept. 14, 2006

Dear all, I have a class (in hello.cc file) whose constructor assgins a value to a global function pointer. It is compiled by gcc and it works well. But it always reports an error when I used bjam to build the hello.cpp file (which is produced by pyste). If I remove the template, it works. But I do need the template. Does anybody know how to wrap it correctly? Any suggestion is appreciated. The hello.cc, hello.pyste, Jamfile are as following: /********************hello.cc*************/ #include <stdio.h> float add(float a, float b) { return a+b; } float minus(float a, float b) { return a-b; } float (*op)(float a, float b) = add; template<class P> class cover_tree { P value; public: cover_tree(float (* metric)(float a, float b)) { op= metric; } cover_tree() { } }; int main() { float a,b,c; a = 1; b = 2; cover_tree<float> ct2(minus); c = op(a,b); printf("%f\n",c); } /************************hello.pyste*********/ cover_tree = Template('cover_tree','hello.cc') cover = cover_tree("float") Function("main", "hello.cc") /************************Jamfile***********/ # This is the top of our own project tree project-root ; # Include definitions needed for Python modules import python ; extension hello # Declare a Python extension called hello : hello.cpp # requirements and dependencies for Boost.Python extensions <template>@boost/libs/python/build/extension ; -- Javen Research School of Information Sciences and Engineering Australian National University Locked Bag 8001 Canberra ACT 2601

1 0