DOC: Improve replace docstring #18100
Conversation
reidy-p
force-pushed
the
replace_docstring
branch
from
2e64b2a to
e67276e
Compare
Codecov Report
@@ Coverage Diff @@
## master #18100 +/- ##
==========================================
- Coverage 91.25% 91.14% -0.11%
==========================================
Files 163 163
Lines 50120 50225 +105
==========================================
+ Hits 45737 45780 +43
- Misses 4383 4445 +62
Continue to review full report at Codecov.
|
Codecov Report
@@ Coverage Diff @@
## master #18100 +/- ##
==========================================
+ Coverage 91.62% 91.62% +<.01%
==========================================
Files 150 150
Lines 48681 48689 +8
==========================================
+ Hits 44604 44612 +8
Misses 4077 4077
Continue to review full report at Codecov.
|
There was a problem hiding this comment.
Haven't gone through this in detail yet, but I think it'd be cleaner to leave the implementation in pandas/core/generic.py, and to only overwrite the docstring in series.py and frame.py.
So in, e.g. series.py it'd look like
def replace(self, to_replace=None, value=None, inplace=False, limit=None,
regex=False, method='pad', axis=None):
"""add your docstring"""
super().replace(to_replace= to_replace, value=value...)Then you don't have all the duplicated code.
pandas/core/frame.py
Outdated
| @@ -227,6 +232,30 @@ | |||
|
|
|||
| """ | |||
|
|
|||
| def _single_replace(self, to_replace, method, inplace, limit): | |||
There was a problem hiding this comment.
Could define this in one place and then import it in frame.py and series.py.
Also a short docstring would still be nice, even though it's internal.
pandas/core/frame.py
Outdated
|
|
||
| See Also | ||
| -------- | ||
| :func:`DataFrame.fillna` : Fill NA/NaN values |
There was a problem hiding this comment.
You do't need the
:func:`
. Just DataFrame.fillna should be enough for numpydoc.
|
@TomAugspurger thanks for the review, made some updates. I wasn't sure what to do with the docstring for |
reidy-p
force-pushed
the
replace_docstring
branch
2 times, most recently
from
9254f5c to
2b4cefc
Compare
|
can you rebase |
reidy-p
force-pushed
the
replace_docstring
branch
from
2b4cefc to
8ab24cd
Compare
pandas/core/series.py
Outdated
| @@ -2658,6 +2661,193 @@ def fillna(self, value=None, method=None, axis=None, inplace=False, | |||
| limit=limit, downcast=downcast, | |||
| **kwargs) | |||
|
|
|||
| def replace(self, to_replace=None, value=None, inplace=False, limit=None, | |||
There was a problem hiding this comment.
can we share most / all of this via _shared_docs? otherwise end up with lots of duplication here
reidy-p
force-pushed
the
replace_docstring
branch
from
e05597f to
420db01
Compare
|
I removed the duplication by using |
|
@jorisvandenbossche if you'd have a look. I only glanced briefly :> |
pandas/core/generic.py
Outdated
| **must** be the same length. | ||
| - Second, if ``regex=True`` then all of the strings in **both** | ||
| lists will be interpreted as regexs otherwise they will match | ||
| directly. This doesn't matter much for `value` since there | ||
| directly. This doesn't matter much for ``value`` since there | ||
| are only a few possible substitution regexes you can use. | ||
| - str and regex rules apply as above. |
There was a problem hiding this comment.
I suppose this is "str, regex, and numeric rules apply as above." now?
reidy-p
force-pushed
the
replace_docstring
branch
from
0772078 to
008588c
Compare
|
thanks! |
xref #17673
closes #13852
git diff upstream/master -u -- "*.py" | flake8 --diffI have tried to separate the docstrings for series and dataframe as suggested in the original issue but can try to combine them if that works better.
As noted in the original issue,
replacecan do a lot things and I've tried to cover most of them with the examples but would welcome other suggestions.