DOC: fix Dataframe.to_records examples #22419
Conversation
There was a problem hiding this comment.
Thanks for the fixes. If you want, you can also replace in this PR, boolean by bool in the parameters types, remove the y in Returns and just leave the type, and add a description of what is returned in the next line.
This way the docstring will follow all the standards.
Thanks!
pandas/core/frame.py
Outdated
|
|
||
| >>> df.to_records(convert_datetime64=False) | ||
| rec.array([('2018-01-01T09:00:00.000000000', 1, 0.5 ), | ||
| ('2018-01-01T09:01:00.000000000', 2, 0.75)], | ||
| dtype=[('index', '<M8[ns]'), ('A', '<i8'), ('B', '<f8')]) | ||
|
|
There was a problem hiding this comment.
As convert_datetime64 is now deprecated, I'd get rid of the examples for it. I feel like they encourage writing code that will be broken soon.
Also, do you mind removing the blank lines before and after the closing """ of the docstring?
Codecov Report
@@ Coverage Diff @@
## master #22419 +/- ##
=======================================
Coverage 92.05% 92.05%
=======================================
Files 169 169
Lines 50733 50733
=======================================
Hits 46702 46702
Misses 4031 4031
Continue to review full report at Codecov.
|
|
Changes made, as requested. I also added an example of what happens when the index of the dataframe is given a label. As expected, the attribute of the recarray is named using the label of the index, but I thought it was ambiguous from the docs what might happen. |
pandas/core/frame.py
Outdated
| is set to 'index'. If the index has a label then this is used as the | ||
| field name: | ||
|
|
||
| >>> df.index.rename("I", inplace=True) |
There was a problem hiding this comment.
don't use inplace in examples, pls just reassign the index
… DataFrame.to_records
* Fixed incorrect case in DataFrame * Changed 'attributes' to 'fields' when describing recarray features * Improved description of how the index is set in the recarray
| @@ -21,7 +21,7 @@ if [ "$DOCTEST" ]; then | |||
|
|
|||
| # DataFrame / Series docstrings | |||
| pytest --doctest-modules -v pandas/core/frame.py \ | |||
| -k"-assign -axes -combine -isin -itertuples -join -nlargest -nsmallest -nunique -pivot_table -quantile -query -reindex -reindex_axis -replace -round -set_index -stack -to_dict -to_records -to_stata -transform" | |||
| -k"-assign -axes -combine -isin -itertuples -join -nlargest -nsmallest -nunique -pivot_table -quantile -query -reindex -reindex_axis -replace -round -set_index -stack -to_dict -to_stata -transform" | |||
There was a problem hiding this comment.
I was asked to make this change by @TomAugspurger in issue #18160. I believe the point is to check that the examples are correct.
There was a problem hiding this comment.
This is a whitelist of doctests that are allowed to fail till we fix them.
There was a problem hiding this comment.
Yep, going to open an issue about this
|
thanks @H0R5E @pandas-dev/pandas-core merge on green. |
|
Happy to help. Keep up the great work! |
jorisvandenbossche
changed the title
|
Thanks for the contribution @H0R5E |
git diff upstream/master -u -- "*.py" | flake8 --diffFixed the examples of Dataframe.to_records related to timestamp conversion following the changes in default behaviour made in pull request #18902. As requested in issue #18160.