diff options
Diffstat (limited to 'numpy/doc/CAPI.txt')
| -rw-r--r-- | numpy/doc/CAPI.txt | 24 |
1 files changed, 12 insertions, 12 deletions
diff --git a/numpy/doc/CAPI.txt b/numpy/doc/CAPI.txt index 9286ee050..bfe062c66 100644 --- a/numpy/doc/CAPI.txt +++ b/numpy/doc/CAPI.txt @@ -26,7 +26,7 @@ of the API) that will need to be changed: * If you passed ``array->dimensions`` and ``array->strides`` around to functions, you will need to fix some code. These are now - ``intp*`` pointers. On 32-bit systems there won't be a problem. + ``npy_intp*`` pointers. On 32-bit systems there won't be a problem. However, on 64-bit systems, you will need to make changes to avoid errors and segfaults. @@ -54,8 +54,8 @@ This is a very flexible function. :: PyObject * PyArray_NewFromDescr(PyTypeObject *subtype, PyArray_Descr *descr, - int nd, intp *dims, - intp *strides, char *data, + int nd, npy_intp *dims, + npy_intp *strides, char *data, int flags, PyObject *obj); ``subtype`` : ``PyTypeObject *`` @@ -76,11 +76,11 @@ This is a very flexible function. ``nd`` : ``int`` The number of dimensions (<``MAX_DIMS``) -``*dims`` : ``intp *`` +``*dims`` : ``npy_intp *`` A pointer to the size in each dimension. Information will be copied from here. -``*strides`` : ``intp *`` +``*strides`` : ``npy_intp *`` The strides this array should have. For new arrays created by this routine, this should be ``NULL``. If you pass in memory for this array to use, then you can pass in the strides information as well @@ -123,7 +123,7 @@ Note: The returned array object will be unitialized unless the type is ``PyArray_OBJECT`` in which case the memory will be set to ``NULL``. ``PyArray_SimpleNew(nd, dims, typenum)`` is a drop-in replacement for -``PyArray_FromDims`` (except it takes ``intp*`` dims instead of ``int*`` dims +``PyArray_FromDims`` (except it takes ``npy_intp*`` dims instead of ``int*`` dims which matters on 64-bit systems) and it does not initialize the memory to zero. @@ -135,7 +135,7 @@ are loose wrappers around this function. These functions still take ``int *`` arguments. This should be fine on 32-bit systems, but on 64-bit systems you may run into trouble if you frequently passed ``PyArray_FromDims`` the dimensions member of the old ``PyArrayObject`` structure -because ``sizeof(intp) != sizeof(int)``. +because ``sizeof(npy_intp) != sizeof(int)``. Getting an arrayobject from an arbitrary Python object @@ -292,11 +292,11 @@ array, or part of some larger record array. But, they may have other uses... Some useful combinations of these flags: -- ``BEHAVED_FLAGS = ALIGNED | WRITEABLE`` -- ``CARRAY_FLAGS = DEFAULT_FLAGS = CONTIGUOUS | BEHAVED_FLAGS`` -- ``CARRAY_FLAGS_RO = CONTIGUOUS | ALIGNED`` -- ``FARRAY_FLAGS = FORTRAN | BEHAVED_FLAGS`` -- ``FARRAY_FLAGS_RO = FORTRAN | ALIGNED`` +- ``BEHAVED = ALIGNED | WRITEABLE`` +- ``CARRAY = DEFAULT = CONTIGUOUS | BEHAVED`` +- ``CARRAY_RO = CONTIGUOUS | ALIGNED`` +- ``FARRAY = FORTRAN | BEHAVED`` +- ``FARRAY_RO = FORTRAN | ALIGNED`` The macro ``PyArray_CHECKFLAGS(obj, flags)`` can test any combination of flags. There are several default combinations defined as macros already |