Hi,
I’d like to achieve the following:
Upon attempting to open a file with an extension/format not supported by Jupyter, I’d like to open a custom notebook instead. For example, I’d like to be able to open a .npz (numpy archive) file and, instead of an error saying that the file format isn’t utf-8 encoded, create a notebook that I could just run to visualize the content of the .npz file.
Is it achievable?
Thanks!
Helping folks to work with files in templated notebooks is an interesting idea, but we’d want to be careful that Lab, etc. didn’t think the npz actually was a notebook… more like it would be a command (which ideally could be opened from that failure dialog), e.g.
foo.npz is not UTF-8 Encoded
[CANCEL] [Open with example notebook...]
A potential home for such an effort is jupyterlab-starters which is triggered by commands and launcher items, and creates templated files/folders. A starter is currently parametrized by (at least) the current working directory… but an alternate command would make it possible to pre-seed with a file.
In the case of an .npz, one would need, well, numpy, but other libraries might offer other interesting cases… i guess one would make packages that provided default notebooks (or really any activity), and ensure they brought the correct dependencies along for the ride.