BUG: Add limit_area to EA ffill/bfill #56616
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rhshadrach
added
Bug
Missing-data
phofl
reviewed
Dec 28, 2023
doc/source/whatsnew/v2.2.0.rst
Outdated
| @@ -313,7 +313,7 @@ Other enhancements | |||
| - :meth:`DataFrame.apply` now allows the usage of numba (via ``engine="numba"``) to JIT compile the passed function, allowing for potential speedups (:issue:`54666`) | |||
| - :meth:`ExtensionArray._explode` interface method added to allow extension type implementations of the ``explode`` method (:issue:`54833`) | |||
| - :meth:`ExtensionArray.duplicated` added to allow extension type implementations of the ``duplicated`` method (:issue:`55255`) | |||
| - :meth:`Series.ffill`, :meth:`Series.bfill`, :meth:`DataFrame.ffill`, and :meth:`DataFrame.bfill` have gained the argument ``limit_area`` (:issue:`56492`) | |||
| - :meth:`Series.ffill`, :meth:`Series.bfill`, :meth:`DataFrame.ffill`, and :meth:`DataFrame.bfill` have gained the argument ``limit_area``; 3rd part :class:`.ExtensionArray` authors need to add this argument to the method ``_pad_or_backfill`` (:issue:`56492`) | |||
There was a problem hiding this comment.
I would put this into the ExtensionArray section as well
phofl
reviewed
Dec 28, 2023
| @@ -1012,6 +1018,12 @@ def _pad_or_backfill( | |||
| DeprecationWarning, | |||
| stacklevel=find_stack_level(), | |||
| ) | |||
| if limit_area is not None: | |||
| raise NotImplementedError( | |||
There was a problem hiding this comment.
We only get here if 3rd party authors didn't implement this themselves, correct?
There was a problem hiding this comment.
Right - this is only hit when EA author overrides .fillna but not ._pad_or_backfill. Currently pandas .ffill will call the EA's .fillna in such a case, which can only be correctly done when limit_area is None.
|
Friendly ping @phofl @jbrockmendel |
phofl
reviewed
Jan 3, 2024
| @@ -205,7 +210,21 @@ def _pad_or_backfill( | |||
| if copy: | |||
| npvalues = npvalues.copy() | |||
| new_mask = new_mask.copy() | |||
| elif limit_area is not None: | |||
| mask = mask.copy() | |||
There was a problem hiding this comment.
Non-blocking, but why do we need this copy here?
phofl
approved these changes
Jan 3, 2024
|
thx @rhshadrach |
meeseeksmachine
pushed a commit
to meeseeksmachine/pandas
that referenced
this pull request
Jan 3, 2024
|
Opened #56729 to track the deprecation |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
limit_areapassed. #41813 (Replace xxxx with the GitHub issue number)doc/source/whatsnew/vX.X.X.rstfile if fixing a bug or adding a new feature.Follow up to #56531.
This is slightly complicated by the
EA.fillnadeprecation.EA.fillnais implemented, we raise wheneverlimit_areais not NoneEA._pad_or_backfillis implemented and does not have alimit_areaarg, we raise whenlimit_areais not NoneMy plan is to introduce a deprecation in 3.1 whenever
_pad_or_backfillgets called (regardless of whetherlimit_areaisNone) informing EA authors they need to add this argument. Assuming this is a good way to do, I'll open a tracking issue for this.For dtypes where the corresponding
maskis not used after filling (e.g.NDArrayBackedExtensionArray), this takes a shortcut by modifying mask prior to filling. Compared to implementinglimit_areaafter filling, this saves 1 or 2 copies (depending on the EA) and extra computation, at the cost of makingmaskno longer necessarily represent NA values.