DOC: Add example to extract docstring, and re-explain change to match. #5826
Conversation
pandas/core/strings.py
Outdated
There was a problem hiding this comment.
No need for the #. Just the explanation line, optionally ending with a :, then a blank line, and then the code example. In that case, the code example will be rendered nicely as code. Like this:
A pattern with more than one group will return a DataFrame:
>>> Series(['a1', 'b2', 'c3']).str.extract('([ab])(\d)')
Named groups will become column names in the result:
...
|
Nice examples in the docstring! I added some comments related to numpy's docstring standard. And a suggestion, while you are at it, maybe you could also add a 'See also' section to refer to |
|
looks good.... lmk when you are ready for merging... |
|
Thanks for the help, @jorisvandenbossche . I read up on the numpy documentation standard. Ready to merge. |
DOC: Add example to extract docstring, and re-explain change to match.
|
@danielballan thanks...the py2.6 failure was bogus... |
There was a problem hiding this comment.
You forgot to remove the old 'return'
|
@danielballan I added some more comments. Do you want to handle it a new PR (otherwise I can also do it)? |
|
Ah, I thought you wanted output only for the first example, for brevity. I should have asked for clarification. See #5838 . |
|
@danielballan yeah, maybe that wasn't very clear. But I think the examples are much more valuable to users if you can see in a glimpse the possible outputs. |
3 participants
@jreback, I added examples per your request in #5099. Also, I rewrote the explanation of how
str.matchis changing, which I left confusing and wordy before.