Thanks -- I've pushed this to master now.
Jack
Bastien writes:
> Hi Jack,
>
> Jack Kamm writes:
>
>> For now, I think I prefer to keep ob-python leaner, so am going to hold
>> off on this.
>
> The leaner the less maintainance ahead :)
>
>> I'll wait a week or so for comments before merging thi
Hi Jack,
Jack Kamm writes:
> For now, I think I prefer to keep ob-python leaner, so am going to hold
> off on this.
The leaner the less maintainance ahead :)
> I'll wait a week or so for comments before merging this new, more
> limited patch into master.
LGTM, thanks!
--
Bastien
After letting it sit, I'm not sure that my patch above is a good idea
anymore. While it would be useful, it also adds substantial complexity
to ob-python.
For now, I think I prefer to keep ob-python leaner, so am going to hold
off on this.
An alternative approach is to have the user handle graphi
After taking another look at my patch, I realized that I was not quite
converting dictionaries to proper alists.
Attached is a tweak to do this properly. The printing of dictionaries is
not quite as pretty, in particular it's not a table anymore:
#+begin_src python
return {"a": 1, "b": 2}
#+end
Hi Kyle,
Thanks for the comments, I'm attaching an updated patch.
Kyle Meyer writes:
> ModuleNotFoundError wasn't added until Python 3.6, so I think it'd be
> better to use its parent class, ImportError.
I did not know this, thanks for the tip.
> Should handling of Series also be added?
Yes,
Jack Kamm writes:
> The attached patch adds ob-python value results handling for the
> following types of results:
>
> - Dictionaries
> - Numpy arrays
> - Pandas dataframes
> - Matplotlib figures
Thanks. Just a couple of drive-by comments...
> +try:
> +import
The attached patch adds ob-python value results handling for the
following types of results:
- Dictionaries
- Numpy arrays
- Pandas dataframes
- Matplotlib figures
This is a bigger commit than I'm used to, so I thought I better send it
out before merging, in case someone notices obvious problems
