David Li
David Li
The bug seems to be that newlines are not being escaped in the output, leading to invalid JSON.
Other things I noticed during testing: - IPython expects the JSON to be correctly indented. So use `json.dumps(..., indent=4)`. - Characters must be escaped; use `str(...).encode('string_escape')`. - On the JSON...
Oh, and all the results are empty.
Some of the output cells are being treated as headings.
Sorry, yeah, I think I never got around to it. As I remember, I just never got around to testing the `ALLOWED_HOSTS` setting, which can't really be tested without actually...
@ashutoshsaboo I believe the PR did work, so you could try merging it into your pull request. If the `ALLOWED_HOSTS` stuff is too difficult to work through, we could just...
That would be nice. I'll look into this (currently I'm working on another set of improvements for Gamma).
I think [`NumPyDocString`](https://github.com/numpy/numpydoc/blob/master/numpydoc/docscrape.py#L88) will apparently parse a docstring and expose its info [in a reasonable way](https://github.com/numpy/numpydoc/blob/master/numpydoc/tests/test_docscrape.py#L489). `docscrape.py` also doesn't seem to be dependent on anything else in `numpydoc`, which makes...
We might even be able to parse some of the examples (perhaps only the one-liners) and link them as Gamma queries as well.
I can take a look at the mobile site sometime (probably this weekend, since school has started).