ENH 14194: add style option for hiding index and columns #16141

nchmura4 · 2017-04-26T04:56:04Z

[ x] closes ENH: DataFrame.style option for hiding the index #14194
[ x] tests added / passed
[ x] passes git diff upstream/master --name-only -- '*.py' | flake8 --diff
whatsnew entry

codecov · 2017-04-26T05:29:53Z

Codecov Report

Merging #16141 into master will decrease coverage by 0.03%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master   #16141      +/-   ##
==========================================
- Coverage   91.39%   91.36%   -0.04%     
==========================================
  Files         164      164              
  Lines       49854    49790      -64     
==========================================
- Hits        45566    45492      -74     
- Misses       4288     4298      +10

Flag	Coverage Δ
#multiple	`89.17% <ø> (-0.02%)`	⬇️
#single	`39.49% <ø> (-0.01%)`	⬇️

Impacted Files	Coverage Δ
pandas/io/formats/style.py	`100% <ø> (ø)`	⬆️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/tseries/offsets.py	`96.91% <0%> (-0.16%)`	⬇️
pandas/core/panel.py	`97.14% <0%> (-0.15%)`	⬇️
pandas/core/frame.py	`97.8% <0%> (-0.1%)`	⬇️
pandas/core/indexes/datetimes.py	`95.48% <0%> (ø)`	⬆️
pandas/core/indexes/range.py	`95.66% <0%> (ø)`	⬆️
pandas/core/reshape/reshape.py	`100% <0%> (ø)`	⬆️
pandas/core/indexes/timedeltas.py	`91.14% <0%> (ø)`	⬆️
pandas/core/indexes/multi.py	`96.38% <0%> (ø)`	⬆️
... and 10 more

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 9c799e2...cb779ec. Read the comment docs.

jreback · 2017-04-26T13:23:48Z

pandas/io/formats/style.py

@@ -69,6 +69,9 @@ class Styler(object):
        a unique identifier to avoid CSS collisons; generated automatically
    caption: str, default None
        caption to attach to the table
+    index_hidden: bool, default False
+        hides the index in the html output


index=True is our standard way of doing this

add a blank line before version added

And add an entry for hidden_columns

jreback · 2017-04-26T13:24:44Z

pandas/tests/io/formats/test_style.py

+        ctx = self.df.style._translate()
+        self.assertTrue(ctx['body'][0][0]['is_visible'])
+        self.assertTrue(ctx['head'][0][0]['is_visible'])
+        ctx2 = self.df.style.hide_index()._translate()


use bare asset here

Thanks. Do you mean use ctx intead of ctx#?

no I mean

assert cts['body'][0][0]['is_visible']

got it, thanks!

TomAugspurger · 2017-04-26T18:06:41Z

pandas/io/formats/style.py

@@ -223,7 +234,7 @@ def format_attr(pair):
            row_es = [{"type": "th",
                       "value": BLANK_VALUE,
                       "display_value": BLANK_VALUE,
-                       "is_visible": True,
+                       "is_visible": (not index_hidden),


You shouldn't need the parentheses here.

TomAugspurger · 2017-04-26T18:19:15Z

pandas/io/formats/style.py

@@ -69,6 +69,9 @@ class Styler(object):
        a unique identifier to avoid CSS collisons; generated automatically
    caption: str, default None
        caption to attach to the table
+    index_hidden: bool, default False
+        hides the index in the html output


And add an entry for hidden_columns

gfyoung · 2017-04-29T20:46:59Z

pandas/tests/io/formats/test_style.py

+        assert(ctx['body'][0][0]['is_visible'])
+        assert(ctx['head'][0][0]['is_visible'])
+        ctx2 = self.df.style.hide_index()._translate()
+        self.assertFalse(ctx2['body'][0][0]['is_visible'])


Don't use self.assertFalse. Use assert not ... instead.

gfyoung · 2017-04-29T20:47:38Z

pandas/tests/io/formats/test_style.py

+        # single named index
+        ctx3 = self.df.set_index('A').style._translate()
+        assert(ctx3['body'][0][0]['is_visible'])
+        self.assertEqual(len(ctx3['head']), 2)  # 2 header levels


Don't use self.assertEqual. Use assert ... == ... instead.

gfyoung · 2017-04-29T20:50:01Z

@nchmura4 : My two comments regarding self.assertEqual and self.assertFalse should be applied to ALL places where you wrote that. We're making a massive effort to remove function calls like this, so before this is merged, the diff should be checked to make sure that those (and any others) are not used.

nchmura4 · 2017-05-01T03:25:30Z

@gfyoung got it, thanks for explaining

TomAugspurger

Could you clear all the cell output in the notebook? We commit just the source to the repository, then nbsphinx executes it while building the docs.

TomAugspurger · 2017-05-12T11:46:42Z

pandas/io/formats/style.py

@@ -257,8 +274,8 @@ def format_attr(pair):
                row_es.append(es)
            head.append(row_es)

-        if self.data.index.names and not all(x is None
-                                             for x in self.data.index.names):
+        if self.data.index.names and index_visible\


Could you write this is

if (self.data.index.names and index_visible and not all(x is None for x in self.data.index.names)):

we try to avoid continuation backslashes

jreback · 2017-05-12T11:49:33Z

doc/source/whatsnew/v0.20.2.txt

@@ -49,9 +49,12 @@ Indexing
 I/O
 ^^^

+- :class:`pandas.io.formats.style.Styler` now has ``index`` parameter and corresponding method ``hide_index()`` to determine whether the index will be rendered in ouptut (:issue:`14194`)


this would be in Other Enhancements

jreback · 2017-05-12T11:50:10Z

pandas/io/formats/style.py

+
+        .. versionadded:: 0.20.2
+
+    hidden_cols: IndexSlice, default None


why is this not called columns? (or do we have that are already)

i thought hidden_cols was more intuitive of a name, since these columns won't show up. but i'm fine with changing it if it breaks the standard

jreback · 2017-05-12T11:52:22Z

pandas/io/formats/style.py

+    def hide_columns(self, subset):
+        """
+        Hide columns from rendering.
+


how does subset make sense here? does it make sense to hide some columns?

The idea is that you may have columns that are needed for formatting -- determining the color of a row, for example -- but you may not want to display them. For example, lets say you had a dataframe with columns City and AvgTemp. Avg Temp was used by Styler to color each row. But in the final rendering, you only want to display city (colored appropriately). AFAIK there is no way to do that right now, but with this change you could do df.style.hide_columns(['AvgTemp']). You only want to hide AvgTemp.

this is a bit unusual for pandas, but I think it's fine here. It's kind of like drop, except it's hiding instead of dropping.

maybe we should support df.style[list_of_selected_columns] for showing some columns. (separate discussion)

gfyoung · 2017-05-15T20:04:21Z

pandas/tests/io/formats/test_style.py

+    def test_hide_single_index(self):
+        # single unnamed index
+        ctx = self.df.style._translate()
+        assert(ctx['body'][0][0]['is_visible'])


We generally use the raw assert with no parentheses. This should be applied to ALL places where you added an assertion statement.

TomAugspurger · 2017-05-15T20:07:20Z

pandas/io/formats/style.py

@@ -118,7 +127,8 @@ class Styler(object):
    template = env.get_template("html.tpl")

    def __init__(self, data, precision=None, table_styles=None, uuid=None,
-                 caption=None, table_attributes=None):
+                 caption=None, table_attributes=None, index=True,
+                 hidden_cols=None):


Can you rename hidden_cols to hidden_columns

TomAugspurger · 2017-05-15T20:07:33Z

pandas/io/formats/style.py

+        self.index_visible = index
+        if hidden_cols is None:
+            hidden_cols = []
+        self.hidden_cols = hidden_cols


Here as well.

jreback · 2017-08-01T22:58:06Z

can you rebase

nchmura4 · 2017-08-06T04:47:21Z

thanks, done.

jreback · 2017-09-23T17:02:51Z

can you rebase / update

nchmura4 · 2017-09-25T00:13:23Z

@jreback - done. i updated all the version references, as well. let me know if you need me to do anything else.

gfyoung · 2017-09-25T00:36:10Z

https://circleci.com/gh/pandas-dev/pandas/4945#tests/containers/3

@jreback : Is this symptomatic of another numpy compatibility issue?

gfyoung · 2017-09-25T00:36:36Z

pandas/io/formats/style.py

+
+        Parameters
+        ----------
+


Remove this line.

gfyoung · 2017-09-25T00:37:17Z

pandas/tests/io/formats/test_style.py

@@ -891,6 +891,116 @@ def test_mi_sparse_column_names(self):
        ]
        assert head == expected

+    def test_hide_single_index(self):


For this test all of other tests that you've added, reference your PR (or issue number) under the function definition for that test.

jreback · 2017-09-23T17:02:17Z

doc/source/whatsnew/v0.20.2.txt

@@ -85,5 +85,12 @@ Numeric



+Other Enhancements
+^^^^^^^^^^^^^^^^^^
+


move to 0.21.0

jreback · 2017-09-23T17:02:26Z

pandas/io/formats/style.py

@@ -70,6 +70,15 @@ class Styler(object):
        a unique identifier to avoid CSS collisons; generated automatically
    caption: str, default None
        caption to attach to the table
+    index: bool, default True
+        determines if the index is rendered in the html output
+


change to 0.21.0

jreback · 2017-09-25T10:27:46Z

pandas/io/formats/style.py

@@ -70,6 +70,15 @@ class Styler(object):
        a unique identifier to avoid CSS collisons; generated automatically
    caption: str, default None
        caption to attach to the table
+    index: bool, default True
+        determines if the index is rendered in the html output


I realize these new parametrize should directly reflect the methods, so use hide_index, hide_columns

jreback · 2017-09-25T10:28:47Z

pandas/io/formats/style.py

@@ -140,6 +150,11 @@ def __init__(self, data, precision=None, table_styles=None, uuid=None,
            precision = get_option('display.precision')
        self.precision = precision
        self.table_attributes = table_attributes
+        self.index_visible = index


just call this .hidden_index/.hidden_columns to avoid the name clash

jreback · 2017-09-25T10:31:42Z

doc/source/whatsnew/v0.21.0.txt

@@ -163,6 +163,9 @@ Other Enhancements
 - :func:`Categorical.rename_categories` now accepts a dict-like argument as `new_categories` and only updates the categories found in that dict. (:issue:`17336`)
 - :func:`read_excel` raises ``ImportError`` with a better message if ``xlrd`` is not installed. (:issue:`17613`)
 - :meth:`DataFrame.assign` will preserve the original order of ``**kwargs`` for Python 3.6+ users instead of sorting the column names
+- :class:`pandas.io.formats.style.Styler` now has ``index`` parameter and corresponding method ``hide_index()`` to determine whether the index will be rendered in ouptut (:issue:`14194`)


I would just highlite the methods, which are more idiomatic.

jreback · 2017-09-25T10:32:45Z

pandas/io/formats/style.py

+        self.index_visible = index
+        if hidden_columns is None:
+            hidden_columns = []
+        self.hidden_columns = hidden_columns


I actually wouldn't add these to the Styler.__init__ prefer these as just methods. (like almost everything else). having more than 1 way to do things is not great.

TomAugspurger · 2017-09-29T22:02:42Z

Merged in master to fix the Circle-CI build. Hopefully it passes this time.

jreback

I think this only needs a bit of updating

jreback · 2017-11-12T18:57:32Z

doc/source/whatsnew/v0.21.0.txt

@@ -164,6 +164,9 @@ Other Enhancements
 - :func:`read_excel` raises ``ImportError`` with a better message if ``xlrd`` is not installed. (:issue:`17613`)
 - :func:`read_json` now accepts a ``chunksize`` parameter that can be used when ``lines=True``. If ``chunksize`` is passed, read_json now returns an iterator which reads in ``chunksize`` lines with each iteration. (:issue:`17048`)
 - :meth:`DataFrame.assign` will preserve the original order of ``**kwargs`` for Python 3.6+ users instead of sorting the column names
+- :class:`pandas.io.formats.style.Styler` now has method ``hide_index()`` to determine whether the index will be rendered in ouptut (:issue:`14194`)


move to 0.22

jreback · 2017-11-12T18:57:52Z

pandas/io/formats/style.py

+        """
+        Hide any indices from rendering.
+
+        .. versionadded:: 0.21.0


jreback · 2017-11-12T18:57:55Z

pandas/io/formats/style.py

+        """
+        Hide columns from rendering.
+
+        .. versionadded:: 0.21.0


TomAugspurger · 2017-11-15T14:12:28Z

Let us know when all the tests pass.

jreback · 2017-11-17T01:14:25Z

doc/source/whatsnew/v0.22.0.txt

@@ -24,7 +24,8 @@ Other Enhancements

 - Better support for :func:`Dataframe.style.to_excel` output with the ``xlsxwriter`` engine. (:issue:`16149`)
 - :func:`pandas.tseries.frequencies.to_offset` now accepts leading '+' signs e.g. '+1h'. (:issue:`18171`)
-
+- :class:`pandas.io.formats.style.Styler` now has ``index`` parameter and corresponding method ``hide_index()`` to determine whether the index will be rendered in ouptut (:issue:`14194`)
+- :class:`pandas.io.formats.style.Styler` now has ``hidden_cols`` parameter and corresponding method ``hide_columns()`` to determine whether columns will be hidden in output (:issue:`14194`)


hide_cols -> hide_columns

didn't we eliminate the paramaters as user facing and only use the methods? (which i think is better anyhow)

@jreback yes, thanks, i should have caught that. fixed!

jreback · 2017-11-19T16:15:38Z

thanks @nchmura4

jreback reviewed Apr 26, 2017

View reviewed changes

TomAugspurger reviewed Apr 26, 2017

View reviewed changes

jreback added the Output-Formatting label Apr 27, 2017

gfyoung reviewed Apr 29, 2017

View reviewed changes

nchmura4 force-pushed the nchmura-df-style-hide branch from 4da33af to 41dd4be Compare May 8, 2017 05:12

TomAugspurger requested changes May 12, 2017

View reviewed changes

jreback requested changes May 12, 2017

View reviewed changes

nchmura4 force-pushed the nchmura-df-style-hide branch from f06cfc3 to 806ef8c Compare May 15, 2017 11:16

gfyoung reviewed May 15, 2017

View reviewed changes

TomAugspurger reviewed May 15, 2017

View reviewed changes

nchmura4 force-pushed the nchmura-df-style-hide branch from b906173 to f5c16ae Compare August 6, 2017 04:44

nchmura4 force-pushed the nchmura-df-style-hide branch from f5c16ae to 9853bfd Compare September 24, 2017 21:48

gfyoung reviewed Sep 25, 2017

View reviewed changes

pandas/io/formats/style.py Outdated

Parameters

----------

Copy link

Member

gfyoung Sep 25, 2017

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Remove this line.

gfyoung reviewed Sep 25, 2017

View reviewed changes

nchmura4 force-pushed the nchmura-df-style-hide branch from 9853bfd to 08cf8db Compare September 25, 2017 01:32

jreback requested changes Sep 25, 2017

View reviewed changes

jreback requested changes Nov 12, 2017

View reviewed changes

nchmura4 added 3 commits November 13, 2017 23:59

ENH 14194: add style option for hiding index and columns

39dfbd5

removed aruments from init

33d5e5d

changed version to v0.22.0

72f0bc0

nchmura4 added 2 commits November 14, 2017 00:52

fix merge

650f3f7

fix typo in whatsnew

4556057

nchmura4 force-pushed the nchmura-df-style-hide branch from 8942f76 to 4556057 Compare November 15, 2017 13:24

fix merge

a5c4e9e

TomAugspurger approved these changes Nov 15, 2017

View reviewed changes

jreback requested changes Nov 17, 2017

View reviewed changes

removed reference to Styler parameters which were removed

cb779ec

jreback added this to the 0.22.0 milestone Nov 19, 2017

jreback approved these changes Nov 19, 2017

View reviewed changes

jreback merged commit b00e62c into pandas-dev:master Nov 19, 2017


		.. versionadded:: 0.20.2

		hidden_cols: IndexSlice, default None

		@@ -85,5 +85,12 @@ Numeric



		Other Enhancements
		^^^^^^^^^^^^^^^^^^

ENH 14194: add style option for hiding index and columns #16141

ENH 14194: add style option for hiding index and columns #16141

Conversation

nchmura4 commented Apr 26, 2017 • edited Loading

codecov bot commented Apr 26, 2017 • 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

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

gfyoung commented Apr 29, 2017

nchmura4 commented May 1, 2017

TomAugspurger 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

nchmura4 May 12, 2017 • edited Loading

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

jreback commented Aug 1, 2017

nchmura4 commented Aug 6, 2017

jreback commented Sep 23, 2017

nchmura4 commented Sep 25, 2017

gfyoung commented Sep 25, 2017

Choose a reason for hiding this comment

gfyoung Sep 25, 2017 • edited Loading

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 Sep 29, 2017

jreback 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

TomAugspurger commented Nov 15, 2017

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jreback commented Nov 19, 2017

nchmura4 commented Apr 26, 2017 •

edited

Loading

codecov bot commented Apr 26, 2017 •

edited

Loading

nchmura4 May 12, 2017 •

edited

Loading

gfyoung Sep 25, 2017 •

edited

Loading