feat: modify VectorInit to handle numpy multi dimensional arrays #28
Conversation
|
Two points: first, don't change comment indentation (I know the style is a bit unusual, but the reason for it is that it makes super-easy to read either the code or the comments, instead of having both in the same flow) and don't remove white lines if it changes the style. Second, as discussed, the NumPy folks designed the "new-style" buffer protocol: https://docs.python.org/3/c-api/buffer.html . There are examples of its use in As for the use of creating the |
Code doesn't execute due to `PyObject_GetBuffer` non-matching function call, even though it is correct way as per the numpy buffer protocol docs
Code executes but Pycode return an error to handle TemplateProxy
Logic to implement the initialisation still remains
It results in TypeError: a bytes-like object is required, not 'cppyy.TemplateProxy'
Khushiyant
changed the title
Khushiyant
force-pushed
the
numpy-dim
branch
2 times, most recently
from
e249453 to
2d5f981
Compare
| @@ -52,6 +52,24 @@ bool HasAttrDirect(PyObject* pyclass, PyObject* pyname, bool mustBeCPyCppyy = fa | |||
| return false; | |||
| } | |||
|
|
|||
| template <typename T> | |||
| T Get_IndexValue(Py_buffer *view, int i) | |||
There was a problem hiding this comment.
| T Get_IndexValue(Py_buffer *view, int i) | |
| bool Get_IndexValue(Py_buffer *view, T* value) |
| for (Py_ssize_t i = 0; i < fillsz; i++) | ||
| { | ||
| int val = Get_IndexValue<int>(view, i); | ||
| PyObject *item = PyLong_FromLong(val); |
There was a problem hiding this comment.
Mapping of view->format to typename (e.g. 'i' -> 'int'; see low-level view type_traits).
TConverter* conv = CreateConverter(type_returned);
PyObject* item = conv->FromMemory(view->buf + i*view->itemsize);
if (conv->HasState()) delete conv;
|
Implemented this PR logic in cppyy (frontend) - cppyy#255 instead |
No description provided.