BUG: SparseDataFrame construction with lists not coercing to dtype (GH 15682) #15834
Conversation
pandas/sparse/frame.py
Outdated
| @@ -159,7 +159,7 @@ def _init_dict(self, data, index, columns, dtype=None): | |||
| v = [v.get(i, nan) for i in index] | |||
|
|
|||
| v = sp_maker(v) | |||
| sdict[k] = v | |||
| sdict[k] = v.astype(np.dtype(dtype)) if dtype is not None else v | |||
There was a problem hiding this comment.
you would do this in sp_maker, also don't use if/else on a single line (hard to read)
pandas/tests/sparse/test_frame.py
Outdated
| @@ -758,7 +757,8 @@ def test_sparse_frame_fillna_limit(self): | |||
| def test_rename(self): | |||
| # just check this works | |||
| renamed = self.frame.rename(index=str) # noqa | |||
| renamed = self.frame.rename(columns=lambda x: '%s%d' % (x, len(x))) # noqa | |||
There was a problem hiding this comment.
can you compare these results (as no tests before).
There was a problem hiding this comment.
ok, just figured it out, added the right test on my last commit! :)
pandas/tests/sparse/test_frame.py
Outdated
| @@ -1226,7 +1225,6 @@ def test_from_to_scipy_object(spmatrix, fill_value): | |||
|
|
|||
|
|
|||
| class TestSparseDataFrameArithmetic(tm.TestCase): | |||
|
|
|||
There was a problem hiding this comment.
try to turn off this removing blank lines like this
pandas/tests/sparse/test_frame.py
Outdated
| df = pd.SparseDataFrame( | ||
| {'a': [1, 0, 0], 'b': [0, 1, 0], 'c': [0, 0, 1]}, dtype='uint8', | ||
| default_fill_value=0) | ||
| result = df.dtypes[0] |
There was a problem hiding this comment.
create the result and use assert_sp_frame_equal
There was a problem hiding this comment.
you can use the comparison from the issue (IOW construct both ways and compare)
There was a problem hiding this comment.
move this test where other test_constructor tests are
pandas/tests/sparse/test_frame.py
Outdated
| df = pd.SparseDataFrame( | ||
| {'a': [1, 0, 0], 'b': [0, 1, 0], 'c': [0, 0, 1]}, dtype='uint8', | ||
| default_fill_value=0) | ||
| result = df.dtypes[0] |
There was a problem hiding this comment.
you can use the comparison from the issue (IOW construct both ways and compare)
pandas/tests/sparse/test_frame.py
Outdated
| df = pd.SparseDataFrame( | ||
| {'a': [1, 0, 0], 'b': [0, 1, 0], 'c': [0, 0, 1]}, dtype='uint8', | ||
| default_fill_value=0) | ||
| result = df.dtypes[0] |
There was a problem hiding this comment.
move this test where other test_constructor tests are
Codecov Report
@@ Coverage Diff @@
## master #15834 +/- ##
==========================================
- Coverage 90.97% 90.96% -0.02%
==========================================
Files 143 143
Lines 49429 49442 +13
==========================================
+ Hits 44969 44973 +4
- Misses 4460 4469 +9
Continue to review full report at Codecov.
|
…H 15682) closes pandas-dev#15682 Author: Carlos Souza <[email protected]> Closes pandas-dev#15834 from ucals/bug-fix-15682 and squashes the following commits: 04fba8d [Carlos Souza] Adding test_rename test cases (were missing) 483bb2c [Carlos Souza] Doing adjustments as per @jreback requests cc4c15b [Carlos Souza] Fixing coersion bug at SparseDataFrame construction faa5c5c [Carlos Souza] Merge remote-tracking branch 'upstream/master' 43456a5 [Carlos Souza] Merge remote-tracking branch 'upstream/master' 8b463cb [Carlos Souza] Merge remote-tracking branch 'upstream/master' 9fc617b [Carlos Souza] Merge remote-tracking branch 'upstream/master' e12bca7 [Carlos Souza] Sync fork 676a4e5 [Carlos Souza] Test
git diff upstream/master --name-only -- '*.py' | flake8 --diff