DOC: Add more documentation showcasing CalendarDay #22633
Conversation
Codecov Report
@@ Coverage Diff @@
## master #22633 +/- ##
=======================================
Coverage 92.16% 92.16%
=======================================
Files 169 169
Lines 50708 50708
=======================================
Hits 46734 46734
Misses 3974 3974
Continue to review full report at Codecov.
|
| idx = pd.date_range("2016-10-30", freq='H', periods=4*24, tz='Europe/Helsinki') | ||
| s = pd.Series(range(len(idx)), index=idx) | ||
| s.resample('D').count() | ||
| s.resample('CD').count() |
There was a problem hiding this comment.
Is this supposed to work on master? After a clean build locally I get a ValueError: Invalid frequency: CD.
Don't think doctest covers these files; if it does then my mistake but figured it was worth asking
There was a problem hiding this comment.
This currently works on master:
In [3]: pd.__version__
Out[3]: '0.24.0.dev0+563.g0976e1261'
In [4]: idx = pd.date_range("2016-10-30", freq='H', periods=4*24, tz='Europe/
...: Helsinki')
...: s = pd.Series(range(len(idx)), index=idx)
...: s.resample('D').count()
...: s.resample('CD').count()
...:
...:
Out[4]:
2016-10-30 00:00:00+03:00 25
2016-10-31 00:00:00+02:00 24
2016-11-01 00:00:00+02:00 24
2016-11-02 00:00:00+02:00 23
Freq: CD, dtype: int64
I am not sure if the doctest uses master or the the latest tagged branch.
| @@ -317,14 +317,34 @@ and respect calendar day arithmetic while :class:`Day` and frequency alias ``'D' | |||
| will now respect absolute time (:issue:`22274`, :issue:`20596`, :issue:`16980`, :issue:`8774`) | |||
| See the :ref:`documentation here <timeseries.dayvscalendarday>` for more information. | |||
|
|
|||
| Addition with :class:`CalendarDay` across a daylight savings time transition: | |||
| The difference between :class:`Day` vs :class:`CalendarDay` is most apparent | |||
| with timezone-aware datetime data with a daylight savings time transition: | |||
There was a problem hiding this comment.
You say "most apparent with timezone-aware data". But is there a case where this is actually relevant for tz naive data?
There was a problem hiding this comment.
There is no difference between Day and CalendarDay for tz naive data. I've parameterized some tz naive data tests for 'D' and 'CD' to ensure this behavior.
|
Hello @mroeschke! Thanks for updating the PR.
|
|
Closing as we're not moving forward with "CD" |
xref #22288 (comment)
Highlighting usages of
CalendarDaywithdate_rangeandresamplespecifically.