ENH: Add Styler.pipe() method (#23229) #23384

nmusolino · 2018-10-27T20:52:18Z

Added Styler.pipe() method. This allows users to easily apply and compose functions that operate on Styler objects, just like the DataFrame.pipe() method does for dataframes.

closes Styler class should have a pipe() method, akin to DataFrame.pipe() #23229
tests added / passed
passes git diff upstream/master -u -- "*.py" | flake8 --diff
whatsnew entry

pep8speaks · 2018-10-27T20:52:22Z

Hello @nmusolino! Thanks for updating the PR.

There are no PEP8 issues in the file pandas/io/formats/style.py !
There are no PEP8 issues in the file pandas/tests/io/formats/test_style.py !

Comment last updated on November 18, 2018 at 02:04 Hours UTC

codecov · 2018-10-27T21:34:07Z

Codecov Report

Merging #23384 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #23384      +/-   ##
==========================================
+ Coverage   92.31%   92.31%   +<.01%     
==========================================
  Files         161      161              
  Lines       51471    51473       +2     
==========================================
+ Hits        47515    47517       +2     
  Misses       3956     3956

Flag	Coverage Δ
#multiple	`90.7% <100%> (ø)`	⬆️
#single	`42.43% <50%> (ø)`	⬆️

Impacted Files	Coverage Δ
pandas/io/formats/style.py	`96.71% <100%> (+0.01%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 0e7cf48...726d01d. Read the comment docs.

gfyoung · 2018-10-27T21:56:53Z

doc/source/whatsnew/v0.24.0.txt

@@ -214,6 +214,7 @@ Other Enhancements
 - Compatibility with Matplotlib 3.0 (:issue:`22790`).
 - Added :meth:`Interval.overlaps`, :meth:`IntervalArray.overlaps`, and :meth:`IntervalIndex.overlaps` for determining overlaps between interval-like objects (:issue:`21998`)
 - :meth:`Timestamp.tz_localize`, :meth:`DatetimeIndex.tz_localize`, and :meth:`Series.tz_localize` have gained the ``nonexistent`` argument for alternative handling of nonexistent times. See :ref:`timeseries.timezone_nonexsistent` (:issue:`8917`)
+- :meth:`Styler.pipe` method added, to simplify application of user-defined functions that operate on stylers.


Reference the issue number.

Also, you should create a mini-section explaining the enhancement in more detail so that end users can understand the benefit.

Okay, I can do that.

gfyoung · 2018-10-27T21:58:31Z

pandas/io/formats/style.py

@@ -1222,6 +1222,35 @@ class MyStyler(cls):

        return MyStyler

+    def pipe(self, func, *args, **kwargs):
+        """
+        Apply func(self, *args, **kwargs)


This should be a little more descriptive. Apply func to what? Also, for what purpose?

(yes, as reviewers, we can see what the rationale was from your PR, so just put that rationale into the docstring)

Okay, I'll change the doctoring, and try to provide some additional context/background.

gfyoung · 2018-10-27T22:00:16Z

pandas/tests/io/formats/test_style.py

+        def g(**kwargs):
+            assert 'styler' in kwargs
+            return kwargs['styler'].data
+        assert self.df.style.pipe((g, 'styler')) is self.df


Can you comment here explaining why you're using is instead of tm.assert_frame_equal ?

Also, add a comment explaining why you're creating this g function in the first place (perhaps a more descriptive function name is needed).

Yes, I can rewrite this a bit to simplify it. Essentially, I wanted to test that when pipe() is called with a (callable, string) tuple, it works as advertised.

pandas/io/formats/style.py

nmusolino

Thank you, @gfyoung, for the prompt review and feedback. All of your comments make sense, and I'll try to make the suggested changes.

nmusolino · 2018-10-27T23:19:00Z

doc/source/whatsnew/v0.24.0.txt

@@ -214,6 +214,7 @@ Other Enhancements
 - Compatibility with Matplotlib 3.0 (:issue:`22790`).
 - Added :meth:`Interval.overlaps`, :meth:`IntervalArray.overlaps`, and :meth:`IntervalIndex.overlaps` for determining overlaps between interval-like objects (:issue:`21998`)
 - :meth:`Timestamp.tz_localize`, :meth:`DatetimeIndex.tz_localize`, and :meth:`Series.tz_localize` have gained the ``nonexistent`` argument for alternative handling of nonexistent times. See :ref:`timeseries.timezone_nonexsistent` (:issue:`8917`)
+- :meth:`Styler.pipe` method added, to simplify application of user-defined functions that operate on stylers.


Okay, I can do that.

nmusolino · 2018-10-27T23:22:20Z

pandas/tests/io/formats/test_style.py

+        def g(**kwargs):
+            assert 'styler' in kwargs
+            return kwargs['styler'].data
+        assert self.df.style.pipe((g, 'styler')) is self.df


Yes, I can rewrite this a bit to simplify it. Essentially, I wanted to test that when pipe() is called with a (callable, string) tuple, it works as advertised.

nmusolino · 2018-10-27T23:24:14Z

pandas/io/formats/style.py

@@ -1222,6 +1222,35 @@ class MyStyler(cls):

        return MyStyler

+    def pipe(self, func, *args, **kwargs):
+        """
+        Apply func(self, *args, **kwargs)


Okay, I'll change the doctoring, and try to provide some additional context/background.

nmusolino · 2018-10-28T02:04:49Z

I have not pushed any changes yet, but this what the pipe() docs look like after local changes:

nmusolino · 2018-10-28T02:47:02Z

I'll summarize the changes made since the original push:

Simplified the second part of the unit test
Added "Notes" and "Examples" section to the docstring. A rendering of the HTML documentation can be found in an earlier comment.
Added a "version added" directive to the docstring (caught that one myself).
Moved the "whatsnew" entry from a one-liner to a short paragraph/example.

I wasn't able to test how the whatsnew will be rendered.

At this point, the main questions are:

Do the notes/examples for the new method look reasonable? I'm open to any suggestions there.
Should I leave this PR as five distinct commits, or manually rebase it into one commit? I'm reasonably familiar with different approaches, but don't know what the common practice is for pandas.

pandas/io/formats/style.py

doc/source/whatsnew/v0.24.0.txt

jreback · 2018-10-28T03:21:06Z

doc/source/whatsnew/v0.24.0.txt

@@ -180,6 +180,26 @@ array, but rather an ``ExtensionArray``:
 This is the same behavior as ``Series.values`` for categorical data. See
 :ref:`whatsnew_0240.api_breaking.interval_values` for more.

+New ``Styler.pipe()`` method
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+A new method :meth:`pandas.formats.style.Styler.pipe()` was added (:issue:`#23229`).


Styler as gained a pipe method

doc/source/whatsnew/v0.24.0.txt

datapythonista

Added some comments about the docstring.

datapythonista · 2018-10-28T03:27:02Z

pandas/io/formats/style.py

@@ -1222,6 +1222,76 @@ class MyStyler(cls):

        return MyStyler

+    def pipe(self, func, *args, **kwargs):
+        """
+        Apply func(self, \*args, \*\*kwargs), and return the result.


Can you add double backticks around func(...)

datapythonista · 2018-10-28T03:32:29Z

pandas/io/formats/style.py

+            ``callable`` that expects the Styler.
+        args : iterable, optional
+            positional arguments passed into ``func``.
+        kwargs : mapping, optional


Add the * before args and kwargs. I'd prefer dict intead of mapping. Parameter descriptions should start with a capital letter.

If you run ./scripts/validate_docstrings.py pandas.Styler.pipe you should get a report with most of the docstring issues.

Done. As mentioned above, now validating.

datapythonista · 2018-10-28T03:34:19Z

pandas/io/formats/style.py

+
+        Returns
+        -------
+        object : the value returned by ``func``.


Is this not always returning a pandas.Styler object?

The description should go in the next line (indented), and without the colon. And it should start by a capital letter.

The return type is determined by func, the user-provided function.

datapythonista · 2018-10-28T03:35:28Z

pandas/io/formats/style.py

+        --------
+        Styler.apply
+        Styler.applymap
+        pandas.DataFrame.pipe


pandas. prefix not needed. Please add a description on why those are relevant in the context of Styler.pipe.

pandas/io/formats/style.py

datapythonista · 2018-10-28T03:40:23Z

pandas/io/formats/style.py

+        ...     ''' Highlight the indicated element, and its containing row. '''
+        ...     cell, row = pd.IndexSlice[row_label, column], pd.IndexSlice[row_label, :]
+        ...     return (styler.set_properties(cell, **{'background-color': '#ffffcc'})
+        ...                   .set_properties(row, **{'background-color': '#ffccc0'}))


It'd be great if you can find a simpler example to illustrate .pipe(). Something like adding quotes around the values, or something as simple as that, so the function does not distract on what .pipe() does.

That is a good suggestion. I chose this to illustrate functionality that seems to belong in a function, and to use a function that could sensibly be applied twice.

BUT that's not the clearest example for the docs. I'll use a much simpler function as you suggest.

nmusolino

Thanks for the feedback. I've made changes as requested. Here is an image of the updated documentation:

The pandas.DataFrame links are not resolving, but I think that's because I am only building part of the docs, using python make.py --single api.

pandas/io/formats/style.py

nmusolino · 2018-10-30T02:03:39Z

pandas/io/formats/style.py

@@ -1222,6 +1222,76 @@ class MyStyler(cls):

        return MyStyler

+    def pipe(self, func, *args, **kwargs):
+        """
+        Apply func(self, \*args, \*\*kwargs), and return the result.


nmusolino · 2018-10-30T02:11:16Z

pandas/io/formats/style.py

+            ``callable`` that expects the Styler.
+        args : iterable, optional
+            positional arguments passed into ``func``.
+        kwargs : mapping, optional


Done. As mentioned above, now validating.

nmusolino · 2018-10-30T02:12:09Z

pandas/io/formats/style.py

+
+        Returns
+        -------
+        object : the value returned by ``func``.


The return type is determined by func, the user-provided function.

nmusolino · 2018-10-30T02:12:16Z

pandas/io/formats/style.py

+        --------
+        Styler.apply
+        Styler.applymap
+        pandas.DataFrame.pipe


pandas/io/formats/style.py

nmusolino · 2018-10-30T02:17:50Z

pandas/io/formats/style.py

+        ...     ''' Highlight the indicated element, and its containing row. '''
+        ...     cell, row = pd.IndexSlice[row_label, column], pd.IndexSlice[row_label, :]
+        ...     return (styler.set_properties(cell, **{'background-color': '#ffffcc'})
+        ...                   .set_properties(row, **{'background-color': '#ffccc0'}))


That is a good suggestion. I chose this to illustrate functionality that seems to belong in a function, and to use a function that could sensibly be applied twice.

BUT that's not the clearest example for the docs. I'll use a much simpler function as you suggest.

doc/source/whatsnew/v0.24.0.txt

nmusolino · 2018-10-30T02:19:19Z

doc/source/whatsnew/v0.24.0.txt

@@ -180,6 +180,26 @@ array, but rather an ``ExtensionArray``:
 This is the same behavior as ``Series.values`` for categorical data. See
 :ref:`whatsnew_0240.api_breaking.interval_values` for more.

+New ``Styler.pipe()`` method
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+A new method :meth:`pandas.formats.style.Styler.pipe()` was added (:issue:`#23229`).


TomAugspurger · 2018-10-31T19:16:17Z

doc/source/whatsnew/v0.24.0.txt

@@ -180,6 +180,27 @@ array, but rather an ``ExtensionArray``:
 This is the same behavior as ``Series.values`` for categorical data. See
 :ref:`whatsnew_0240.api_breaking.interval_values` for more.

+New ``Styler.pipe()`` method
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+The dataframe Styler class has gained a :meth:`~pandas.io.formats.style.Styler.pipe` method


Newline before this line.

You should be able to link to Styler class here with :class:`pandas.io.formats.style.Styler`

I can make those changes. I’ll also (a) add an anchor, like .. _whatsnew_0240.enhancements.styler_pipe: and (b) change the code block style to match everything else in the file.

pandas/io/formats/style.py

TomAugspurger · 2018-11-01T15:19:35Z

I'm fine with skipping the example with f and g. That could also be made into a ..code-block:: python directive.

…

On Thu, Nov 1, 2018 at 8:03 AM Nicholas Musolino ***@***.***> wrote: ***@***.**** commented on this pull request. ------------------------------ In pandas/io/formats/style.py <#23384 (comment)>: > @@ -1222,6 +1222,76 @@ class MyStyler(cls): return MyStyler + def pipe(self, func, *args, **kwargs): I am traveling at the moment and am not able to copy-paste the output at this moment. But my recollection is: 1. There were no docstring style issues identified by the validate_docstrings.py script, in my most recent version. 2. The initial doctests on line 1260 (namely >>> f(g(df.style.set_precision(3), arg1=a), arg2=b, arg3=c)) and the subsequent rewritten version fail because the functions f and g are not defined. I suppose the same thing happens in the analogous DataFrame.pipe doctest <https://github.com/pandas-dev/pandas/blob/0409521665bd436a10aea7e06336066bf07ff057/pandas/core/generic.py#L4230> . 3. The later doctest on line 1287 executes as intended, but fails when the actual output (a string like <pandas.io.formats.style.Styler at 0xaabcd538364>) does not match the listed output in the doctest (no output included). I could change this to make it formally pass, but it would require a doctest: +ELLIPSIS directive (to accommodate the memory address in the string), and it would be inconsistent with other doctests in this module (which don’t show the styler’s string representation as output). — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub <#23384 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABQHImhjtI1ukK57kFlbGC6WyVYQWrNxks5uqw0lgaJpZM4X9o-F> .

TomAugspurger · 2018-11-02T17:06:40Z

Merge conflict in 0.24.0, if you could update.

nmusolino · 2018-11-03T01:34:58Z

I can resolve the merge conflict and turn the first example into a code-block instead of a doctest. I’ll also clean up the whatsnew entry as discussed earlier. I will push some changes Sunday night or Monday night.

nmusolino · 2018-11-06T02:44:07Z

Here is the updated method documentation:

[Large, outdated image removed for brevity]

Here is the output from the validate_docstring.py script:

$ python ./scripts/validate_docstrings.py pandas.io.formats.style.Styler.pipe

################################################################################
############### Docstring (pandas.io.formats.style.Styler.pipe)  ###############
################################################################################

Apply ``func(self, *args, **kwargs)``, and return the result.

.. versionadded:: 0.24.0

Parameters
----------
func : function
    Function to apply to the Styler.
    ``*args``, and ``**kwargs`` are passed into ``func``.
    Alternatively a ``(callable, data_keyword)`` tuple where
    ``data_keyword`` is a string indicating the keyword of
    ``callable`` that expects the Styler.
*args : iterable, optional
    Positional arguments passed into ``func``.
**kwargs : dict, optional
    Dictionary of keyword arguments passed into ``func``.

Returns
-------
result : object
    The value returned by ``func``.

See Also
--------
DataFrame.pipe : Analogous method for DataFrame.
Styler.apply : Apply a function row-wise, column-wise, or table-wise to
    modify the dataframe's styling.

Notes
-----
Like :meth:`DataFrame.pipe`, this method can simplify the
application of several user-defined functions to a styler.  Instead
of writing:

.. code-block:: python

    f(g(df.style.set_precision(3), arg1=a), arg2=b, arg3=c)

users can write:

.. code-block:: python

    (df.style.set_precision(3)
       .pipe(g, arg1=a)
       .pipe(f, arg2=b, arg3=c))

In particular, this allows users to define functions that take a
styler object, along with other parameters, and return the styler after
making styling changes (such as calling :meth:`Styler.apply` or
:meth:`Styler.set_properties`).  Using ``.pipe``, these user-defined
style "transformations" can be interleaved with calls to the built-in
Styler interface.

Examples
--------
>>> def set_standard_formatting(styler):
...     return (styler.set_properties(**{'text-align': 'right'})
...                   .format({'X': '{:.1%}'}))

The user-defined highlight function above can be called within a
sequence of other style modifications:

>>> df = pd.DataFrame({'A': list(range(-1, 4)), 'X': np.arange(0.2, 1.2, 0.2)})
>>> (df.style
...    .set_properties(subset=['X'], **{'background-color': 'yellow'})
...    .pipe(set_standard_formatting)
...    .set_caption("Results with column 'X' highlighted."))

################################################################################
################################## Validation ##################################
################################################################################

Errors found:
	Examples do not pass tests

################################################################################
################################### Doctests ###################################
################################################################################

**********************************************************************
Line 65, in pandas.io.formats.style.Styler.pipe
Failed example:
    (df.style
       .set_properties(subset=['X'], **{'background-color': 'yellow'})
       .pipe(set_standard_formatting)
       .set_caption("Results with column 'X' highlighted."))
Expected nothing
Got:
    <pandas.io.formats.style.Styler object at 0x11c982b00>

datapythonista

Added some comments, still needs some work.

Make sure that the docstring validation does not have any error (and that the CI is green).

Thanks for working on this!

datapythonista · 2018-11-06T06:52:24Z

pandas/io/formats/style.py

+        *args : iterable, optional
+            Positional arguments passed into ``func``.
+        **kwargs : dict, optional
+            Dictionary of keyword arguments passed into ``func``.


You can merge args and kwargs in a single line, and you don't need types or optional, as those are always the caseÑ:

*args, **kwargs Arguments passed to `func`.

Changed as suggested. As a result of these changes, the validate_docstring.py script reports an error, but I agree this is better.

pandas/io/formats/style.py

datapythonista · 2018-11-06T07:09:57Z

pandas/io/formats/style.py

+        The user-defined highlight function above can be called within a
+        sequence of other style modifications:
+
+        >>> df = pd.DataFrame({'A': list(range(-1, 4)), 'X': np.arange(0.2, 1.2, 0.2)})


Use something that looks real, no range fuctions, and no a, x, foo columns. We start to have many examples with animal names (e.g. cat, penguin, falcon...). I'd use that for consistency (hopefully at some point we have all the examples with similar data). Just have couple of columns that are percentages, and 2 or 3 rows.

Then rename set_standard_formatting to percentage_format (if I understand correctly, that's what it is, right?

I made some changes here, and fixed the reference in the text, which was incorrect. I changed the function name as you suggested, while trying to make it clear that the function was user-defined and application-specific.

I'll be honest, using animal names in examples seems a little silly to me, but that's up to the pandas developers.

In this case, using numeric data rather than strings made more sense to me, since numeric data presents styling/formatting choices, and special values (min/max/null) can be highlighted.

…_pipe

nmusolino

I think we're reaching the point of diminishing marginal returns to the refinements in this PR, and it's about ready to move across the finish line.

Here's the latest docstring as rendered:

pandas/io/formats/style.py

nmusolino · 2018-11-10T16:20:15Z

pandas/io/formats/style.py

+        *args : iterable, optional
+            Positional arguments passed into ``func``.
+        **kwargs : dict, optional
+            Dictionary of keyword arguments passed into ``func``.


Changed as suggested. As a result of these changes, the validate_docstring.py script reports an error, but I agree this is better.

pandas/io/formats/style.py

nmusolino · 2018-11-10T16:44:29Z

pandas/io/formats/style.py

+        The user-defined highlight function above can be called within a
+        sequence of other style modifications:
+
+        >>> df = pd.DataFrame({'A': list(range(-1, 4)), 'X': np.arange(0.2, 1.2, 0.2)})


I made some changes here, and fixed the reference in the text, which was incorrect. I changed the function name as you suggested, while trying to make it clear that the function was user-defined and application-specific.

I'll be honest, using animal names in examples seems a little silly to me, but that's up to the pandas developers.

In this case, using numeric data rather than strings made more sense to me, since numeric data presents styling/formatting choices, and special values (min/max/null) can be highlighted.

TomAugspurger · 2018-11-10T20:18:24Z

@nmusolino a few lines are too long:
#23384 (comment)

TomAugspurger · 2018-11-10T20:21:37Z

doc/source/whatsnew/v0.24.0.txt

+        return (styler.format({'N': '{:,}', 'X': '{:.1%}'})
+                      .set_properties(**{'text-align': 'right'}))
+
+    (df.style.set_precision(1)


This code is executed. I think the exception at https://travis-ci.org/pandas-dev/pandas/jobs/453316565#L2108 is from this. I'd recommend defining df in this code block.

nmusolino · 2018-11-18T02:09:37Z

@TomAugspurger, I've fixed the line length issues, and made sure the "whatsnew" documentation builds correctly, and looks reasonable.

Any other changes you would suggest? And would you like me to squash, rebase, and force-push to clean up the commit history?

An updated docstring rendering can be found below.

nmusolino · 2018-11-27T14:26:00Z

@jreback @TomAugspurger , could you please take another look at this? I have rebased twice in order to keep up with changes in the whatsnew file, and I'd like to get this PR ready for completion before rebasing again.

TomAugspurger

LGTM, if you could fix the conflict and ping when the CI passes I'll merge.

nmusolino · 2018-11-28T14:33:38Z

@TomAugspurger , I resolved the conflict, and CI is passing. Would you be able to merge?

TomAugspurger · 2018-11-28T14:48:33Z

Looks good, thanks @nmusolino!

* Add Styler.pipe() method, akin to DataFrame.pipe()

nmusolino added 2 commits October 27, 2018 17:33

Add Styler.pipe() method, akin to DataFrame.pipe()

00701c7

Add whatsnew entry for Styler.pipe()

c6a92cd

nmusolino force-pushed the styler_pipe branch from 2047052 to c6a92cd Compare October 27, 2018 21:33

gfyoung added Enhancement Code Style Code style, linting, code_checks labels Oct 27, 2018

gfyoung reviewed Oct 27, 2018

View reviewed changes

gfyoung requested review from jreback and TomAugspurger October 27, 2018 22:02

gfyoung reviewed Oct 27, 2018

View reviewed changes

pandas/io/formats/style.py Show resolved Hide resolved

nmusolino commented Oct 27, 2018

View reviewed changes

nmusolino mentioned this pull request Oct 28, 2018

Styler class should have a pipe() method, akin to DataFrame.pipe() #23229

Closed

Simplify Styler.pipe() unit tests

36606ee

nmusolino added 2 commits October 27, 2018 22:35

Styler.pipe() docstring improvements

73821ba

Expand whatsnew entry per review feedback

df6fb42

jreback reviewed Oct 28, 2018

View reviewed changes

pandas/io/formats/style.py Show resolved Hide resolved

jreback requested changes Oct 28, 2018

View reviewed changes

datapythonista reviewed Oct 28, 2018

View reviewed changes

nmusolino added 3 commits October 29, 2018 22:21

Revise whatsnew message

e5eacbe

Improve Styler.pipe() docstring

51b21ac

Final docstring tweak

a0e758e

nmusolino force-pushed the styler_pipe branch from 1f99d7b to a0e758e Compare October 30, 2018 02:31

nmusolino commented Oct 30, 2018

View reviewed changes

TomAugspurger reviewed Oct 31, 2018

View reviewed changes

nmusolino added 3 commits November 5, 2018 21:31

More whatsnew revisions

21403f7

Change non-runnable doctests to code blocks

0363ea8

Merge branch 'master' into styler_pipe

d1cf095

datapythonista requested changes Nov 6, 2018

View reviewed changes

nmusolino added 4 commits November 10, 2018 11:23

Simplify docstring parameters/return sections

0798c73

Modify doctest example

7cbb6b0

Merge branch 'styler_pipe' of github.com:nmusolino/pandas into styler…

6c79509

…_pipe

Use natural subset for doctest example

ae83236

nmusolino commented Nov 10, 2018

View reviewed changes

TomAugspurger reviewed Nov 10, 2018

View reviewed changes

nmusolino added 2 commits November 17, 2018 20:41

Ensure whatsnew entry runs cleanly

706f3a6

Fix docstring line length, format issues

28d7098

TomAugspurger approved these changes Nov 27, 2018

View reviewed changes

Merge remote-tracking branch 'origin/master' into styler_pipe

726d01d

TomAugspurger merged commit 66abbc3 into pandas-dev:master Nov 28, 2018

Pingviinituutti pushed a commit to Pingviinituutti/pandas that referenced this pull request Feb 28, 2019

ENH: Add Styler.pipe() method (pandas-dev#23229) (pandas-dev#23384)

634470e

* Add Styler.pipe() method, akin to DataFrame.pipe()

Pingviinituutti pushed a commit to Pingviinituutti/pandas that referenced this pull request Feb 28, 2019

ENH: Add Styler.pipe() method (pandas-dev#23229) (pandas-dev#23384)

a0532d3

* Add Styler.pipe() method, akin to DataFrame.pipe()

ENH: Add Styler.pipe() method (#23229) #23384

ENH: Add Styler.pipe() method (#23229) #23384

Conversation

nmusolino commented Oct 27, 2018 • edited Loading

pep8speaks commented Oct 27, 2018 • edited Loading

Comment last updated on November 18, 2018 at 02:04 Hours UTC

codecov bot commented Oct 27, 2018 • edited Loading

Codecov Report

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

gfyoung Oct 27, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

gfyoung Oct 27, 2018 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nmusolino left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nmusolino commented Oct 28, 2018 • edited Loading

nmusolino commented Oct 28, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

datapythonista left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nmusolino left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

TomAugspurger commented Nov 1, 2018 via email

TomAugspurger commented Nov 2, 2018

nmusolino commented Nov 3, 2018

nmusolino commented Nov 6, 2018 • edited Loading

datapythonista left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nmusolino left a comment • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

TomAugspurger commented Nov 10, 2018

Choose a reason for hiding this comment

Choose a reason for hiding this comment

nmusolino commented Nov 18, 2018

nmusolino commented Nov 27, 2018

TomAugspurger left a comment

Choose a reason for hiding this comment

nmusolino commented Nov 28, 2018

TomAugspurger commented Nov 28, 2018

nmusolino commented Oct 27, 2018 •

edited

Loading

pep8speaks commented Oct 27, 2018 •

edited

Loading

codecov bot commented Oct 27, 2018 •

edited

Loading

gfyoung Oct 27, 2018 •

edited

Loading

gfyoung Oct 27, 2018 •

edited

Loading

nmusolino commented Oct 28, 2018 •

edited

Loading

nmusolino commented Nov 6, 2018 •

edited

Loading

nmusolino left a comment •

edited

Loading