Brett Calcott

@wjakob I've added an `nb::writeable` flag in tensor creation so far. Let me know if this is on track.

I think I have the same problem. I'm converting code from pybind11, and I pass in a numpy array which is constructed from a buffer: ```python buf = fstream.read(RAW_CHUNK_SIZE) arr...

I spent a bit of time futzing around trying to debug this. If I change this line: https://github.com/wjakob/nanobind/blob/8555cf8d22336fc0008c4a21fe708723ad0a4721/src/tensor.cpp#L152 to this: ```c++ if (PyObject_GetBuffer(o, view.get(), PyBUF_RECORDS_RO)) { ``` Then my code...

It appears the readonly problem is not just with nanobind. ```python import numpy as np w = np.array([[2.0,2.0]]) w.setflags(write=False) w.__dlpack__() ``` gives ``` TypeError: NumPy currently only supports dlpack for...

13.4 has just been released and the ability to control the hardware key mapping has been kept -- no more workarounds are needed. Everything seems to be working beautifully. Thanks...

Bear with me, I'm new to spark. All I know (and I show above) is that it works perfectly with 1.1.9, and precisely the same thing breaks with 1.1.10. What...

I'm starting to add some custom spans to our FastAPI app, and my immediate thought was: "where are the decorators for this?" What is the status of the PR #1089...

There is now a ```readonly``` flag that allows read-only numpy arrays to be passed through. It forces the translation through the buffer protocol, and has a static-assert guard so that...

Sorry, I don't understand the table -- it looks misleading. Currently nanobind throws an exception if you send it a RO buffer from python. So neither RO -> RO or...

Hi @wjakob. Thanks for the comments. I'll have a look at the DLPack changes, to try and see how they would fit.