BUG: support for "level=" when reset_index() is called with a flat Index #16266

toobaz · 2017-05-06T10:03:58Z

closes DataFrame.reset_index deletes index, does not all for ints as level arg #16263
tests added / passed
passes git diff upstream/master --name-only -- '*.py' | flake8 --diff
whatsnew entry

jorisvandenbossche

Looks good

jorisvandenbossche · 2017-05-06T10:20:45Z

doc/source/whatsnew/v0.21.0.txt

@@ -77,6 +77,7 @@ Conversion
 Indexing
 ^^^^^^^^

+- Bug in ``DataFrame.reset_index(level=)`` with flat index (:issue:`16263`)


Can you move this to the 0.20.2 file?

jorisvandenbossche · 2017-05-06T10:24:35Z

pandas/tests/frame/test_alter_axes.py

+                          columns=['A', 'B', 'C', 'D'])
+
+        # With MultiIndex
+        for levels in ['A', 'B', 'E'], [0, 1, 2]:


Can you remove the 'E' and 2 from the lists? I founded it a bit confusing as you don't use it

yep, they were refactoring leftovers

codecov · 2017-05-06T10:37:36Z

Codecov Report

Merging #16266 into master will decrease coverage by 0.03%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #16266      +/-   ##
==========================================
- Coverage   90.33%    90.3%   -0.04%     
==========================================
  Files         167      167              
  Lines       50907    50908       +1     
==========================================
- Hits        45989    45974      -15     
- Misses       4918     4934      +16

Flag	Coverage Δ
#multiple	`88.09% <100%> (-0.02%)`	⬇️
#single	`40.29% <88.88%> (-0.1%)`	⬇️

Impacted Files	Coverage Δ
pandas/core/frame.py	`97.59% <100%> (-0.1%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/plotting/_converter.py	`63.23% <0%> (-1.82%)`	⬇️
pandas/_version.py	`44.65% <0%> (+1.9%)`	⬆️

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 e513430...d27ae91. Read the comment docs.

codecov · 2017-05-06T10:38:19Z

Codecov Report

Merging #16266 into master will decrease coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master   #16266      +/-   ##
==========================================
- Coverage   90.32%    90.3%   -0.02%     
==========================================
  Files         167      167              
  Lines       50907    50908       +1     
==========================================
- Hits        45982    45973       -9     
- Misses       4925     4935      +10

Flag	Coverage Δ
#multiple	`88.09% <100%> (-0.01%)`	⬇️
#single	`40.29% <88.88%> (-0.11%)`	⬇️

Impacted Files	Coverage Δ
pandas/core/frame.py	`97.59% <100%> (-0.1%)`	⬇️
pandas/io/gbq.py	`25% <0%> (-58.34%)`	⬇️
pandas/core/common.py	`90.68% <0%> (-0.35%)`	⬇️

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 ba60321...c301473. Read the comment docs.

jreback · 2017-05-06T11:24:45Z

doc/source/whatsnew/v0.20.2.txt

@@ -42,7 +42,7 @@ Conversion
 Indexing
 ^^^^^^^^

-
+- Bug in ``DataFrame.reset_index(level=)`` with flat index (:issue:`16263`)


single level index

jreback · 2017-05-06T11:25:33Z

pandas/tests/frame/test_alter_axes.py

@@ -641,6 +641,35 @@ def test_reset_index(self):
        xp = xp.set_index(['B'], append=True)
        assert_frame_equal(rs, xp, check_names=False)

+    def test_reset_index_level(self):
+        df = pd.DataFrame([[1, 2, 3, 4], [5, 6, 7, 8]],
+                          columns=['A', 'B', 'C', 'D'])


issue number

the number is few lines after, it only applies to the flat index

jreback · 2017-05-06T11:27:39Z

pandas/tests/frame/test_alter_axes.py

+            result = df.set_index(['A', 'B']).reset_index(level=levels[0])
+            tm.assert_frame_equal(result, df.set_index('B'))
+
+            result = df.set_index(['A', 'B']).reset_index(level=levels[:1])


can u add similar tests for Series (in series test hierarchy)

jreback · 2017-05-06T11:28:18Z

pandas/tests/frame/test_alter_axes.py

+
+            result = df.set_index('A').reset_index(level=levels[:1])
+            tm.assert_frame_equal(result, df)
+


add with drop as well

jreback · 2017-05-06T11:29:08Z

pandas/tests/frame/test_alter_axes.py

+            with tm.assert_raises_regex(KeyError, 'Level E '):
+                df.set_index(idx_lev).reset_index(level=['A', 'E'])
+            with tm.assert_raises_regex(IndexError, 'Too many levels'):
+                df.set_index(idx_lev).reset_index(level=[0, 1, 2])


it might be easier to parameteize these tests (snd make the errors a separate test)

can you clarify?

I guess this is fine.

closes pandas-dev#16263

jreback · 2017-05-06T12:18:41Z

pandas/tests/series/test_alter_axes.py

+            with tm.assert_raises_regex(KeyError, 'Level E '):
+                s.reset_index(level=['A', 'E'])
+
+            # With flat Index


don't use word flat, rather single level

jreback · 2017-05-06T12:19:26Z

minor comment, otherwise lgtm.

jorisvandenbossche · 2017-05-06T21:50:24Z

Thanks!

…level Index (pandas-dev#16266)

…level Index (pandas-dev#16266) (cherry picked from commit 8809b04)

…level Index (#16266) (cherry picked from commit 8809b04)

…level Index (pandas-dev#16266)

jorisvandenbossche added this to the 0.20.2 milestone May 6, 2017

jorisvandenbossche reviewed May 6, 2017

View reviewed changes

toobaz force-pushed the fix_flat_reset_level branch 2 times, most recently from fd83502 to f4c3121 Compare May 6, 2017 11:16

jreback requested changes May 6, 2017

View reviewed changes

BUG: support for "level=" when reset_index() is called with a flat Index

7cc21c2

closes pandas-dev#16263

toobaz force-pushed the fix_flat_reset_level branch from f4c3121 to 7cc21c2 Compare May 6, 2017 11:34

TST: test drop= and Series

b7be15a

jreback added Bug Indexing Related to indexing on series/frames, not to indexes themselves labels May 6, 2017

jreback reviewed May 6, 2017

View reviewed changes

jreback approved these changes May 6, 2017

View reviewed changes

TST: -flat +single-level

c301473

jorisvandenbossche merged commit 8809b04 into pandas-dev:master May 6, 2017

toobaz deleted the fix_flat_reset_level branch May 6, 2017 22:50

pawroman pushed a commit to pawroman/pandas that referenced this pull request May 9, 2017

BUG: support for "level=" when reset_index() is called with a single …

881b29c

…level Index (pandas-dev#16266)

pcluo pushed a commit to pcluo/pandas that referenced this pull request May 22, 2017

BUG: support for "level=" when reset_index() is called with a single …

ace75cc

…level Index (pandas-dev#16266)

TomAugspurger pushed a commit to TomAugspurger/pandas that referenced this pull request May 29, 2017

BUG: support for "level=" when reset_index() is called with a single …

1e146e4

…level Index (pandas-dev#16266) (cherry picked from commit 8809b04)

TomAugspurger added Backported and removed Needs Backport labels May 30, 2017

TomAugspurger pushed a commit that referenced this pull request May 30, 2017

BUG: support for "level=" when reset_index() is called with a single …

0248460

…level Index (#16266) (cherry picked from commit 8809b04)

stangirala pushed a commit to stangirala/pandas that referenced this pull request Jun 11, 2017

BUG: support for "level=" when reset_index() is called with a single …

ab6fd4b

…level Index (pandas-dev#16266)

jreback mentioned this pull request Jul 15, 2017

Reindexing raises TypeError on Py2.7 #16921

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

BUG: support for "level=" when reset_index() is called with a flat Index #16266

BUG: support for "level=" when reset_index() is called with a flat Index #16266

toobaz commented May 6, 2017

jorisvandenbossche left a comment

jorisvandenbossche May 6, 2017

jorisvandenbossche May 6, 2017

toobaz May 6, 2017

codecov bot commented May 6, 2017 •

edited

Loading

codecov bot commented May 6, 2017 •

edited

Loading

jreback May 6, 2017

jreback May 6, 2017

toobaz May 6, 2017

jreback May 6, 2017

toobaz May 6, 2017

jreback May 6, 2017

toobaz May 6, 2017

jreback May 6, 2017

toobaz May 6, 2017

jreback May 6, 2017

jreback May 6, 2017

jreback commented May 6, 2017

jorisvandenbossche commented May 6, 2017


		result = df.set_index('A').reset_index(level=levels[:1])
		tm.assert_frame_equal(result, df)

BUG: support for "level=" when reset_index() is called with a flat Index #16266

BUG: support for "level=" when reset_index() is called with a flat Index #16266

Conversation

toobaz commented May 6, 2017

jorisvandenbossche 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

codecov bot commented May 6, 2017 • edited Loading

Codecov Report

codecov bot commented May 6, 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

jreback commented May 6, 2017

jorisvandenbossche commented May 6, 2017

codecov bot commented May 6, 2017 •

edited

Loading

codecov bot commented May 6, 2017 •

edited

Loading