Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Ask HN: Better Jupyter code completion?
4 points by leblancfg on Feb 18, 2019 | hide | past | favorite | 1 comment
I've been consistently hitting a wall with Jupyter's autocomplete and docstrings capabilities, especially when working with pandas:

* No autocomplete when kernel is busy

* No docstrings or autocomplete for dot-calls (e.g. `Series.str.contains`)

So I end up having a devdocs.io tab open for just those two situations.

Is anyone aware of efforts to lighten this situation? Say, plug-ins, or integration to other services that would overhaul the default results from Jupyter's <Tab> and <Shift-Tab>? Thanks.



Jupyter has docstrings and autocomplete for dot calls - I just tried your Series.str.contains example in JupyterLab 0.32.

And in general my experience with the documentation and autocomplete has been the exact opposite - way better than the other environments that I've used (VScode, Atom, Spyder). Also nicer than RStudio, which isn't RStudio's fault but more due to R not having dot calls and the documentation being less user-friendly than Python's.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: