ERR: raise if values passed to Categorical is a DataFrame #18207

jbrockmendel · 2017-11-10T00:42:44Z

2nd of 3 to address #17112. This raises instead of making the following mistake:

>>> df = pd.DataFrame(np.random.randn(3,2), columns=['A', 'B'])
>>> pd.Categorical(df)
[A, B]
Categories (2, object): [A, B]

jreback · 2017-11-10T01:05:31Z

pandas/core/categorical.py

@@ -308,6 +308,9 @@ def __init__(self, values, categories=None, ordered=None, dtype=None,
        elif isinstance(values, (ABCIndexClass, ABCSeries)):
            # we'll do inference later
            pass
+        elif isinstance(values, ABCDataFrame):


instead of this specific tests, rather check the ndim of the passed values, it must be 1 (or None)

e.g.
getattr(values, 'ndim', None) in [1, None]

jschendel · 2017-11-10T05:50:35Z

pandas/core/categorical.py

@@ -308,6 +308,9 @@ def __init__(self, values, categories=None, ordered=None, dtype=None,
        elif isinstance(values, (ABCIndexClass, ABCSeries)):
            # we'll do inference later
            pass
+        elif getattr(values, 'ndim', 0) > 1:
+            raise TypeError("> 1 ndim Categorical are not "


Should this be a NotImplementedError instead of a TypeError? Similar to how the existing error with this message on line 350 is a NotImplementedError?

codecov · 2017-11-10T07:47:22Z

Codecov Report

Merging #18207 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #18207      +/-   ##
==========================================
+ Coverage    91.4%    91.4%   +<.01%     
==========================================
  Files         163      163              
  Lines       50064    50066       +2     
==========================================
+ Hits        45759    45763       +4     
+ Misses       4305     4303       -2

Flag	Coverage Δ
#multiple	`89.21% <100%> (+0.02%)`	⬆️
#single	`40.36% <100%> (-0.06%)`	⬇️

Impacted Files	Coverage Δ
pandas/core/categorical.py	`95.67% <100%> (-0.13%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/frame.py	`97.8% <0%> (-0.1%)`	⬇️
pandas/plotting/_converter.py	`65.2% <0%> (+1.81%)`	⬆️

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 ca737ac...b0da084. Read the comment docs.

codecov · 2017-11-10T07:47:27Z

Codecov Report

Merging #18207 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #18207      +/-   ##
==========================================
+ Coverage    91.4%    91.4%   +<.01%     
==========================================
  Files         163      163              
  Lines       50064    50066       +2     
==========================================
+ Hits        45759    45763       +4     
+ Misses       4305     4303       -2

Flag	Coverage Δ
#multiple	`89.21% <100%> (+0.02%)`	⬆️
#single	`40.36% <100%> (-0.06%)`	⬇️

Impacted Files	Coverage Δ
pandas/core/categorical.py	`95.67% <100%> (-0.13%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/frame.py	`97.8% <0%> (-0.1%)`	⬇️
pandas/plotting/_converter.py	`65.2% <0%> (+1.81%)`	⬆️

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 ca737ac...b0da084. Read the comment docs.

codecov · 2017-11-10T07:47:34Z

Codecov Report

Merging #18207 into master will increase coverage by <.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #18207      +/-   ##
==========================================
+ Coverage    91.4%    91.4%   +<.01%     
==========================================
  Files         163      163              
  Lines       50064    50066       +2     
==========================================
+ Hits        45759    45763       +4     
+ Misses       4305     4303       -2

Flag	Coverage Δ
#multiple	`89.21% <100%> (+0.02%)`	⬆️
#single	`40.36% <100%> (-0.06%)`	⬇️

Impacted Files	Coverage Δ
pandas/core/categorical.py	`95.67% <100%> (-0.13%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/frame.py	`97.8% <0%> (-0.1%)`	⬇️
pandas/plotting/_converter.py	`65.2% <0%> (+1.81%)`	⬆️

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 ca737ac...b0da084. Read the comment docs.

gfyoung · 2017-11-10T09:01:21Z

pandas/tests/test_categorical.py

+    def test_constructor_dataframe_error(self):
+        # GH#17112
+        df = pd.DataFrame(np.random.randn(3, 2), columns=['A', 'B'])
+        with pytest.raises(NotImplementedError):


This is where checking the error message is helpful. AFAICT, your tests don't cover both places where a NotImplementedError can be raised.

Good idea, will do.

jreback · 2017-11-10T14:01:23Z

pandas/core/categorical.py

@@ -2331,6 +2334,8 @@ def _factorize_from_iterable(values):
                                      categories=values.categories,
                                      ordered=values.ordered)
        codes = values.codes
+    elif getattr(values, 'ndim', 0) > 1:


this doesn't get hit in the tests

jreback · 2017-11-10T14:01:31Z

pandas/tests/test_categorical.py

+    def test_constructor_dataframe_error(self):
+        # GH#17112
+        df = pd.DataFrame(np.random.randn(3, 2), columns=['A', 'B'])
+        with pytest.raises(NotImplementedError):


jreback · 2017-11-10T14:02:42Z

pandas/tests/test_categorical.py

@@ -173,6 +173,12 @@ def test_constructor_unsortable(self):
        pytest.raises(
            TypeError, lambda: Categorical(arr, ordered=True))

+    def test_constructor_dataframe_error(self):


paramatrize with a ndarray > 1 dim

jbrockmendel · 2017-11-12T18:56:24Z

Don't want to deal with rebase right now. Will return to this bug later. Closing.

raise if values passed to Categorical is a DataFrame

157720c

jreback reviewed Nov 10, 2017

View reviewed changes

change condition check per reviewer suggestion

5f4ef4b

jschendel reviewed Nov 10, 2017

View reviewed changes

use NotImplementedError per reviewer suggestion

b0da084

gfyoung added Error Reporting Incorrect or improved errors from pandas Categorical Categorical Data Type labels Nov 10, 2017

gfyoung reviewed Nov 10, 2017

View reviewed changes

jreback requested changes Nov 10, 2017

View reviewed changes

jreback changed the title ~~raise if values passed to Categorical is a DataFrame~~ ERR: raise if values passed to Categorical is a DataFrame Nov 10, 2017

jreback reviewed Nov 10, 2017

View reviewed changes

jbrockmendel closed this Nov 12, 2017

jbrockmendel deleted the period_bugs2 branch April 5, 2020 17:40

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

ERR: raise if values passed to Categorical is a DataFrame #18207

ERR: raise if values passed to Categorical is a DataFrame #18207

jbrockmendel commented Nov 10, 2017 •

edited by jorisvandenbossche

Loading

jreback Nov 10, 2017

jschendel Nov 10, 2017

jbrockmendel Nov 10, 2017

codecov bot commented Nov 10, 2017

codecov bot commented Nov 10, 2017

codecov bot commented Nov 10, 2017 •

edited

Loading

gfyoung Nov 10, 2017

jreback Nov 10, 2017

jbrockmendel Nov 10, 2017

jreback Nov 10, 2017

jreback Nov 10, 2017

jreback Nov 10, 2017

jbrockmendel commented Nov 12, 2017

ERR: raise if values passed to Categorical is a DataFrame #18207

ERR: raise if values passed to Categorical is a DataFrame #18207

Conversation

jbrockmendel commented Nov 10, 2017 • edited by jorisvandenbossche Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

codecov bot commented Nov 10, 2017

Codecov Report

codecov bot commented Nov 10, 2017

Codecov Report

codecov bot commented Nov 10, 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

jbrockmendel commented Nov 12, 2017

jbrockmendel commented Nov 10, 2017 •

edited by jorisvandenbossche

Loading

codecov bot commented Nov 10, 2017 •

edited

Loading