GetIndicesOfSlice(sly,self->getNumberOfTuples(),&strt,&stp,&step,"DataArray::getNumberOfItemGivenBESRelative (wrap) : the input slice is invalid !");
return DataArray::GetNumberOfItemGivenBESRelative(strt,stp,step,"");
}
+
+ PyObject *__getstate__() const throw(INTERP_KERNEL::Exception)
+ {
+ PyObject *ret(PyTuple_New(2));
+ std::string a0(self->getName());
+ const std::vector<std::string> &a1(self->getInfoOnComponents());
+ PyTuple_SetItem(ret,0,PyString_FromString(a0.c_str()));
+ //
+ int sz(a1.size());
+ PyObject *ret1(PyList_New(sz));
+ for(int i=0;i<sz;i++)
+ PyList_SetItem(ret1,i,PyString_FromString(a1[i].c_str()));
+ PyTuple_SetItem(ret,1,ret1);
+ //
+ return ret;
+ }
+
+ void __setstate__(PyObject *inp) throw(INTERP_KERNEL::Exception)
+ {
+ static const char MSG[]="DataArrayDouble.__setstate__ : expected input is a tuple of size 2 with string as 1st arg and list of string as 2nd arg !";
+ if(!PyTuple_Check(inp))
+ throw INTERP_KERNEL::Exception("DataArrayDouble.__setstate__ : invalid input ! Invalid overwrite of __getstate__ ?");
+ int sz(PyTuple_Size(inp));
+ if(sz!=2)
+ throw INTERP_KERNEL::Exception("DataArrayDouble.__setstate__ : invalid tuple in input ! Should be of size 2 ! Invalid overwrite of __getstate__ ?");
+ PyObject *a0(PyTuple_GetItem(inp,0));
+ if(!PyString_Check(a0))
+ throw INTERP_KERNEL::Exception(MSG);
+ PyObject *a1(PyTuple_GetItem(inp,1));
+ std::vector<std::string> a1cpp;
+ if(!fillStringVector(a1,a1cpp))
+ throw INTERP_KERNEL::Exception(MSG);
+ self->setName(PyString_AsString(a0));
+ self->setInfoOnComponents(a1cpp);
+ }
}
};
PyTuple_SetItem(ret,1,SWIG_NewPointerObj(SWIG_as_voidptr(ret1),SWIGTYPE_p_ParaMEDMEM__DataArrayInt, SWIG_POINTER_OWN | 0 ));
return ret;
}
+
+ // serialization
+ static PyObject *___new___(PyObject *cls, PyObject *args) throw(INTERP_KERNEL::Exception)
+ {
+ static const char MSG[]="DataArrayDouble.__new__ : the args in input is expected to be a tuple !";
+ if(!PyTuple_Check(args))
+ throw INTERP_KERNEL::Exception(MSG);
+ PyObject *builtinsd(PyEval_GetBuiltins());//borrowed
+ PyObject *obj(PyDict_GetItemString(builtinsd,"object"));//borrowed
+ PyObject *selfMeth(PyObject_GetAttrString(obj,"__new__"));
+ //
+ PyObject *tmp0(PyTuple_New(1));
+ PyTuple_SetItem(tmp0,0,cls); Py_XINCREF(cls);
+ PyObject *instance(PyObject_CallObject(selfMeth,tmp0));
+ Py_DECREF(tmp0);
+ Py_DECREF(selfMeth);
+ PyObject *initMeth(PyObject_GetAttrString(instance,"__init__"));
+ int sz(PyTuple_Size(args));
+
+ if(PyTuple_Size(args)==2 && PyDict_Check(PyTuple_GetItem(args,1)) && PyDict_Size(PyTuple_GetItem(args,1))==1 )
+ {// NOT general case. only true if in unpickeling context ! call __init__. Because for all other cases, __init__ is called right after __new__ !
+ PyObject *zeNumpyRepr(0);
+ PyObject *tmp1(PyInt_FromLong(0));
+ zeNumpyRepr=PyDict_GetItem(PyTuple_GetItem(args,1),tmp1);//borrowed
+ Py_DECREF(tmp1);
+ PyObject *tmp3(PyTuple_New(1));
+ PyTuple_SetItem(tmp3,0,zeNumpyRepr); Py_XINCREF(zeNumpyRepr);
+ PyObject *tmp2(PyObject_CallObject(initMeth,tmp3));
+ Py_XDECREF(tmp2);
+ Py_DECREF(tmp3);
+ }
+ Py_DECREF(initMeth);
+ return instance;
+ }
+
+ PyObject *__getnewargs__() throw(INTERP_KERNEL::Exception)
+ {
+#ifdef WITH_NUMPY
+ if(!self->isAllocated())
+ throw INTERP_KERNEL::Exception("PyWrap of DataArrayDouble.__getnewargs__ : self is not allocated !");
+ PyObject *ret(PyTuple_New(1));
+ PyObject *ret0(PyDict_New());
+ PyObject *numpyArryObj(ParaMEDMEM_DataArrayDouble_toNumPyArray(self));
+ {// create a dict to discriminite in __new__ if __init__ should be called. Not beautiful but not idea ...
+ PyObject *tmp1(PyInt_FromLong(0));
+ PyDict_SetItem(ret0,tmp1,numpyArryObj); Py_DECREF(tmp1); Py_DECREF(numpyArryObj);
+ PyTuple_SetItem(ret,0,ret0);
+ }
+ return ret;
+#else
+ throw INTERP_KERNEL::Exception("PyWrap of DataArrayDouble.__getnewargs__ : not implemented because numpy is not active in your configuration ! No serialization/unserialization available without numpy !");
+#endif
+ }
}
};
PyTuple_SetItem(pyRet,1,ret1Py);
return pyRet;
}
+
+ // serialization
+ static PyObject *___new___(PyObject *cls, PyObject *args) throw(INTERP_KERNEL::Exception)
+ {
+ static const char MSG[]="DataArrayInt.__new__ : the args in input is expected to be a tuple !";
+ if(!PyTuple_Check(args))
+ throw INTERP_KERNEL::Exception(MSG);
+ PyObject *builtinsd(PyEval_GetBuiltins());//borrowed
+ PyObject *obj(PyDict_GetItemString(builtinsd,"object"));//borrowed
+ PyObject *selfMeth(PyObject_GetAttrString(obj,"__new__"));
+ //
+ PyObject *tmp0(PyTuple_New(1));
+ PyTuple_SetItem(tmp0,0,cls); Py_XINCREF(cls);
+ PyObject *instance(PyObject_CallObject(selfMeth,tmp0));
+ Py_DECREF(tmp0);
+ Py_DECREF(selfMeth);
+ PyObject *initMeth(PyObject_GetAttrString(instance,"__init__"));
+ int sz(PyTuple_Size(args));
+
+ if(PyTuple_Size(args)==2 && PyDict_Check(PyTuple_GetItem(args,1)) && PyDict_Size(PyTuple_GetItem(args,1))==1 )
+ {// NOT general case. only true if in unpickeling context ! call __init__. Because for all other cases, __init__ is called right after __new__ !
+ PyObject *zeNumpyRepr(0);
+ PyObject *tmp1(PyInt_FromLong(0));
+ zeNumpyRepr=PyDict_GetItem(PyTuple_GetItem(args,1),tmp1);//borrowed
+ Py_DECREF(tmp1);
+ PyObject *tmp3(PyTuple_New(1));
+ PyTuple_SetItem(tmp3,0,zeNumpyRepr); Py_XINCREF(zeNumpyRepr);
+ PyObject *tmp2(PyObject_CallObject(initMeth,tmp3));
+ Py_XDECREF(tmp2);
+ Py_DECREF(tmp3);
+ }
+ Py_DECREF(initMeth);
+ return instance;
+ }
+
+ PyObject *__getnewargs__() throw(INTERP_KERNEL::Exception)
+ {
+#ifdef WITH_NUMPY
+ if(!self->isAllocated())
+ throw INTERP_KERNEL::Exception("PyWrap of DataArrayInt.__getnewargs__ : self is not allocated !");
+ PyObject *ret(PyTuple_New(1));
+ PyObject *ret0(PyDict_New());
+ PyObject *numpyArryObj(ParaMEDMEM_DataArrayInt_toNumPyArray(self));
+ {// create a dict to discriminite in __new__ if __init__ should be called. Not beautiful but not idea ...
+ PyObject *tmp1(PyInt_FromLong(0));
+ PyDict_SetItem(ret0,tmp1,numpyArryObj); Py_DECREF(tmp1); Py_DECREF(numpyArryObj);
+ PyTuple_SetItem(ret,0,ret0);
+ }
+ return ret;
+#else
+ throw INTERP_KERNEL::Exception("PyWrap of DataArrayInt.__getnewargs__ : not implemented because numpy is not active in your configuration ! No serialization/unserialization available without numpy !");
+#endif
+ }
}
};
