Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 6cca195

Browse files
TST: Assist Replace ensure_clean utility function with the temp_file pytest fixture #62435 (#62461)
1 parent bd88fa0 commit 6cca195

File tree

3 files changed

+540
-536
lines changed

3 files changed

+540
-536
lines changed

‎pandas/tests/frame/methods/test_to_csv.py‎

Lines changed: 138 additions & 139 deletions
Original file line numberDiff line numberDiff line change
@@ -248,6 +248,7 @@ def _return_result_expected(
248248
self,
249249
df,
250250
chunksize,
251+
temp_file,
251252
r_dtype=None,
252253
c_dtype=None,
253254
rnlvl=None,
@@ -260,15 +261,13 @@ def _return_result_expected(
260261
kwargs["index_col"] = list(range(rnlvl))
261262
kwargs["header"] = list(range(cnlvl))
262263

263-
with tm.ensure_clean("__tmp_to_csv_moar__") as path:
264-
df.to_csv(path, encoding="utf8", chunksize=chunksize)
265-
recons = self.read_csv(path, **kwargs)
264+
df.to_csv(temp_file, encoding="utf8", chunksize=chunksize)
265+
recons = self.read_csv(temp_file, **kwargs)
266266
else:
267267
kwargs["header"] = 0
268268

269-
with tm.ensure_clean("__tmp_to_csv_moar__") as path:
270-
df.to_csv(path, encoding="utf8", chunksize=chunksize)
271-
recons = self.read_csv(path, **kwargs)
269+
df.to_csv(temp_file, encoding="utf8", chunksize=chunksize)
270+
recons = self.read_csv(temp_file, **kwargs)
272271

273272
def _to_uni(x):
274273
if not isinstance(x, str):
@@ -353,13 +352,13 @@ def _to_uni(x):
353352
@pytest.mark.parametrize(
354353
"nrows", [2, 10, 99, 100, 101, 102, 198, 199, 200, 201, 202, 249, 250, 251]
355354
)
356-
def test_to_csv_nrows(self, nrows):
355+
def test_to_csv_nrows(self, nrows, temp_file):
357356
df = DataFrame(
358357
np.ones((nrows, 4)),
359358
index=date_range("2020年01月01日", periods=nrows),
360359
columns=Index(list("abcd"), dtype=object),
361360
)
362-
result, expected = self._return_result_expected(df, 1000, "dt", "s")
361+
result, expected = self._return_result_expected(df, 1000, temp_file, "dt", "s")
363362
expected.index = expected.index.astype("M8[ns]")
364363
tm.assert_frame_equal(result, expected, check_names=False)
365364

@@ -372,7 +371,7 @@ def test_to_csv_nrows(self, nrows):
372371
)
373372
@pytest.mark.parametrize("ncols", [1, 2, 3, 4])
374373
@pytest.mark.filterwarnings(r"ignore:PeriodDtype\[B\] is deprecated:FutureWarning")
375-
def test_to_csv_idx_types(self, nrows, r_idx_type, c_idx_type, ncols):
374+
def test_to_csv_idx_types(self, nrows, r_idx_type, c_idx_type, ncols, temp_file):
376375
axes = {
377376
"i": lambda n: Index(np.arange(n), dtype=np.int64),
378377
"s": lambda n: Index([f"{i}_{chr(i)}" for i in range(97, 97 + n)]),
@@ -387,6 +386,7 @@ def test_to_csv_idx_types(self, nrows, r_idx_type, c_idx_type, ncols):
387386
result, expected = self._return_result_expected(
388387
df,
389388
1000,
389+
temp_file,
390390
r_idx_type,
391391
c_idx_type,
392392
)
@@ -401,18 +401,18 @@ def test_to_csv_idx_types(self, nrows, r_idx_type, c_idx_type, ncols):
401401
"nrows", [10, 98, 99, 100, 101, 102, 198, 199, 200, 201, 202, 249, 250, 251]
402402
)
403403
@pytest.mark.parametrize("ncols", [1, 2, 3, 4])
404-
def test_to_csv_idx_ncols(self, nrows, ncols):
404+
def test_to_csv_idx_ncols(self, nrows, ncols, temp_file):
405405
df = DataFrame(
406406
np.ones((nrows, ncols)),
407407
index=Index([f"i-{i}" for i in range(nrows)], name="a"),
408408
columns=Index([f"i-{i}" for i in range(ncols)], name="a"),
409409
)
410-
result, expected = self._return_result_expected(df, 1000)
410+
result, expected = self._return_result_expected(df, 1000, temp_file)
411411
tm.assert_frame_equal(result, expected, check_names=False)
412412

413413
@pytest.mark.slow
414414
@pytest.mark.parametrize("nrows", [10, 98, 99, 100, 101, 102])
415-
def test_to_csv_dup_cols(self, nrows):
415+
def test_to_csv_dup_cols(self, nrows, temp_file):
416416
df = DataFrame(
417417
np.ones((nrows, 3)),
418418
index=Index([f"i-{i}" for i in range(nrows)], name="a"),
@@ -427,25 +427,29 @@ def test_to_csv_dup_cols(self, nrows):
427427
ix[-2:] = ["rdupe", "rdupe"]
428428
df.index = ix
429429
df.columns = cols
430-
result, expected = self._return_result_expected(df, 1000, dupe_col=True)
430+
result, expected = self._return_result_expected(
431+
df, 1000, temp_file, dupe_col=True
432+
)
431433
tm.assert_frame_equal(result, expected, check_names=False)
432434

433435
@pytest.mark.slow
434-
def test_to_csv_empty(self):
436+
def test_to_csv_empty(self, temp_file):
435437
df = DataFrame(index=np.arange(10, dtype=np.int64))
436-
result, expected = self._return_result_expected(df, 1000)
438+
result, expected = self._return_result_expected(df, 1000, temp_file)
437439
tm.assert_frame_equal(result, expected, check_column_type=False)
438440

439441
@pytest.mark.slow
440-
def test_to_csv_chunksize(self):
442+
def test_to_csv_chunksize(self, temp_file):
441443
chunksize = 1000
442444
rows = chunksize // 2 + 1
443445
df = DataFrame(
444446
np.ones((rows, 2)),
445447
columns=Index(list("ab")),
446448
index=MultiIndex.from_arrays([range(rows) for _ in range(2)]),
447449
)
448-
result, expected = self._return_result_expected(df, chunksize, rnlvl=2)
450+
result, expected = self._return_result_expected(
451+
df, chunksize, temp_file, rnlvl=2
452+
)
449453
tm.assert_frame_equal(result, expected, check_names=False)
450454

451455
@pytest.mark.slow
@@ -461,7 +465,7 @@ def test_to_csv_chunksize(self):
461465
[{"r_idx_nlevels": 2, "c_idx_nlevels": 2}, {"rnlvl": 2, "cnlvl": 2}],
462466
],
463467
)
464-
def test_to_csv_params(self, nrows, df_params, func_params, ncols):
468+
def test_to_csv_params(self, nrows, df_params, func_params, ncols, temp_file):
465469
if df_params.get("r_idx_nlevels"):
466470
index = MultiIndex.from_arrays(
467471
[f"i-{i}" for i in range(nrows)]
@@ -478,7 +482,9 @@ def test_to_csv_params(self, nrows, df_params, func_params, ncols):
478482
else:
479483
columns = Index([f"i-{i}" for i in range(ncols)])
480484
df = DataFrame(np.ones((nrows, ncols)), index=index, columns=columns)
481-
result, expected = self._return_result_expected(df, 1000, **func_params)
485+
result, expected = self._return_result_expected(
486+
df, 1000, temp_file, **func_params
487+
)
482488
tm.assert_frame_equal(result, expected, check_names=False)
483489

484490
def test_to_csv_from_csv_w_some_infs(self, temp_file, float_frame):
@@ -595,108 +601,104 @@ def test_to_csv_multiindex(self, temp_file, float_frame, datetime_frame):
595601
# needed if setUp becomes class method
596602
datetime_frame.index = old_index
597603

598-
with tm.ensure_clean("__tmp_to_csv_multiindex__") as path:
599-
# GH3571, GH1651, GH3141
600-
601-
def _make_frame(names=None):
602-
if names is True:
603-
names = ["first", "second"]
604-
return DataFrame(
605-
np.random.default_rng(2).integers(0, 10, size=(3, 3)),
606-
columns=MultiIndex.from_tuples(
607-
[("bah", "foo"), ("bah", "bar"), ("ban", "baz")], names=names
608-
),
609-
dtype="int64",
610-
)
611-
612-
# column & index are multi-index
613-
df = DataFrame(
614-
np.ones((5, 3)),
615-
columns=MultiIndex.from_arrays(
616-
[[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
617-
),
618-
index=MultiIndex.from_arrays(
619-
[[f"i-{i}" for i in range(5)] for _ in range(2)], names=list("ab")
620-
),
621-
)
622-
df.to_csv(path)
623-
result = read_csv(path, header=[0, 1, 2, 3], index_col=[0, 1])
624-
tm.assert_frame_equal(df, result)
625-
626-
# column is mi
627-
df = DataFrame(
628-
np.ones((5, 3)),
629-
columns=MultiIndex.from_arrays(
630-
[[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
604+
def _make_frame(names=None):
605+
if names is True:
606+
names = ["first", "second"]
607+
return DataFrame(
608+
np.random.default_rng(2).integers(0, 10, size=(3, 3)),
609+
columns=MultiIndex.from_tuples(
610+
[("bah", "foo"), ("bah", "bar"), ("ban", "baz")], names=names
631611
),
612+
dtype="int64",
632613
)
633-
df.to_csv(path)
634-
result = read_csv(path, header=[0, 1, 2, 3], index_col=0)
635-
tm.assert_frame_equal(df, result)
636-
637-
# dup column names?
638-
df = DataFrame(
639-
np.ones((5, 3)),
640-
columns=MultiIndex.from_arrays(
641-
[[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
642-
),
643-
index=MultiIndex.from_arrays(
644-
[[f"i-{i}" for i in range(5)] for _ in range(3)], names=list("abc")
645-
),
646-
)
647-
df.to_csv(path)
648-
result = read_csv(path, header=[0, 1, 2, 3], index_col=[0, 1, 2])
649-
tm.assert_frame_equal(df, result)
650-
651-
# writing with no index
652-
df = _make_frame()
653-
df.to_csv(path, index=False)
654-
result = read_csv(path, header=[0, 1])
655-
tm.assert_frame_equal(df, result)
656-
657-
# we lose the names here
658-
df = _make_frame(True)
659-
df.to_csv(path, index=False)
660-
result = read_csv(path, header=[0, 1])
661-
assert com.all_none(*result.columns.names)
662-
result.columns.names = df.columns.names
663-
tm.assert_frame_equal(df, result)
664-
665-
# whatsnew example
666-
df = _make_frame()
667-
df.to_csv(path)
668-
result = read_csv(path, header=[0, 1], index_col=[0])
669-
tm.assert_frame_equal(df, result)
670-
671-
df = _make_frame(True)
672-
df.to_csv(path)
673-
result = read_csv(path, header=[0, 1], index_col=[0])
674-
tm.assert_frame_equal(df, result)
675-
676-
# invalid options
677-
df = _make_frame(True)
678-
df.to_csv(path)
679-
680-
for i in [6, 7]:
681-
msg = f"len of {i}, but only 5 lines in file"
682-
with pytest.raises(ParserError, match=msg):
683-
read_csv(path, header=list(range(i)), index_col=0)
684-
685-
# write with cols
686-
msg = "cannot specify cols with a MultiIndex"
687-
with pytest.raises(TypeError, match=msg):
688-
df.to_csv(path, columns=["foo", "bar"])
689-
690-
with tm.ensure_clean("__tmp_to_csv_multiindex__") as path:
691-
# empty
692-
tsframe[:0].to_csv(path)
693-
recons = self.read_csv(path)
694-
695-
exp = tsframe[:0]
696-
exp.index = []
697-
698-
tm.assert_index_equal(recons.columns, exp.columns)
699-
assert len(recons) == 0
614+
615+
# column & index are multi-index
616+
df = DataFrame(
617+
np.ones((5, 3)),
618+
columns=MultiIndex.from_arrays(
619+
[[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
620+
),
621+
index=MultiIndex.from_arrays(
622+
[[f"i-{i}" for i in range(5)] for _ in range(2)], names=list("ab")
623+
),
624+
)
625+
df.to_csv(temp_file)
626+
result = read_csv(temp_file, header=[0, 1, 2, 3], index_col=[0, 1])
627+
tm.assert_frame_equal(df, result)
628+
629+
# column is mi
630+
df = DataFrame(
631+
np.ones((5, 3)),
632+
columns=MultiIndex.from_arrays(
633+
[[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
634+
),
635+
)
636+
df.to_csv(temp_file)
637+
result = read_csv(temp_file, header=[0, 1, 2, 3], index_col=0)
638+
tm.assert_frame_equal(df, result)
639+
640+
# dup column names?
641+
df = DataFrame(
642+
np.ones((5, 3)),
643+
columns=MultiIndex.from_arrays(
644+
[[f"i-{i}" for i in range(3)] for _ in range(4)], names=list("abcd")
645+
),
646+
index=MultiIndex.from_arrays(
647+
[[f"i-{i}" for i in range(5)] for _ in range(3)], names=list("abc")
648+
),
649+
)
650+
df.to_csv(temp_file)
651+
result = read_csv(temp_file, header=[0, 1, 2, 3], index_col=[0, 1, 2])
652+
tm.assert_frame_equal(df, result)
653+
654+
# writing with no index
655+
df = _make_frame()
656+
df.to_csv(temp_file, index=False)
657+
result = read_csv(temp_file, header=[0, 1])
658+
tm.assert_frame_equal(df, result)
659+
660+
# we lose the names here
661+
df = _make_frame(True)
662+
df.to_csv(temp_file, index=False)
663+
result = read_csv(temp_file, header=[0, 1])
664+
assert com.all_none(*result.columns.names)
665+
result.columns.names = df.columns.names
666+
tm.assert_frame_equal(df, result)
667+
668+
# whatsnew example
669+
df = _make_frame()
670+
df.to_csv(temp_file)
671+
result = read_csv(temp_file, header=[0, 1], index_col=[0])
672+
tm.assert_frame_equal(df, result)
673+
674+
df = _make_frame(True)
675+
df.to_csv(temp_file)
676+
result = read_csv(temp_file, header=[0, 1], index_col=[0])
677+
tm.assert_frame_equal(df, result)
678+
679+
# invalid options
680+
df = _make_frame(True)
681+
df.to_csv(temp_file)
682+
683+
for i in [6, 7]:
684+
msg = f"len of {i}, but only 5 lines in file"
685+
with pytest.raises(ParserError, match=msg):
686+
read_csv(temp_file, header=list(range(i)), index_col=0)
687+
688+
# write with cols
689+
msg = "cannot specify cols with a MultiIndex"
690+
with pytest.raises(TypeError, match=msg):
691+
df.to_csv(temp_file, columns=["foo", "bar"])
692+
693+
# empty
694+
tsframe[:0].to_csv(temp_file)
695+
recons = self.read_csv(temp_file)
696+
697+
exp = tsframe[:0]
698+
exp.index = []
699+
700+
tm.assert_index_equal(recons.columns, exp.columns)
701+
assert len(recons) == 0
700702

701703
def test_to_csv_interval_index(self, temp_file, using_infer_string):
702704
# GH 28210
@@ -808,16 +810,15 @@ def test_to_csv_dups_cols(self, temp_file):
808810

809811
df.columns = [0, 1, 2] * 5
810812

811-
with tm.ensure_clean() as filename:
812-
df.to_csv(filename)
813-
result = read_csv(filename, index_col=0)
813+
df.to_csv(temp_file)
814+
result = read_csv(temp_file, index_col=0)
814815

815-
# date cols
816-
for i in ["0.4", "1.4", "2.4"]:
817-
result[i] = to_datetime(result[i])
816+
# date cols
817+
for i in ["0.4", "1.4", "2.4"]:
818+
result[i] = to_datetime(result[i])
818819

819-
result.columns = df.columns
820-
tm.assert_frame_equal(result, df)
820+
result.columns = df.columns
821+
tm.assert_frame_equal(result, df)
821822

822823
def test_to_csv_dups_cols2(self, temp_file):
823824
# GH3457
@@ -1197,18 +1198,16 @@ def test_to_csv_with_dst_transitions_with_pickle(self, start, end, temp_file):
11971198
idx = idx._with_freq(None) # freq does not round-trip
11981199
idx._data._freq = None # otherwise there is trouble on unpickle
11991200
df = DataFrame({"values": 1, "idx": idx}, index=idx)
1200-
with tm.ensure_clean("csv_date_format_with_dst") as path:
1201-
df.to_csv(path, index=True)
1202-
result = read_csv(path, index_col=0)
1203-
result.index = (
1204-
to_datetime(result.index, utc=True)
1205-
.tz_convert("Europe/Paris")
1206-
.as_unit("ns")
1207-
)
1208-
result["idx"] = to_datetime(result["idx"], utc=True).astype(
1209-
"datetime64[ns, Europe/Paris]"
1210-
)
1211-
tm.assert_frame_equal(result, df)
1201+
1202+
df.to_csv(temp_file, index=True)
1203+
result = read_csv(temp_file, index_col=0)
1204+
result.index = (
1205+
to_datetime(result.index, utc=True).tz_convert("Europe/Paris").as_unit("ns")
1206+
)
1207+
result["idx"] = to_datetime(result["idx"], utc=True).astype(
1208+
"datetime64[ns, Europe/Paris]"
1209+
)
1210+
tm.assert_frame_equal(result, df)
12121211

12131212
# assert working
12141213
df.astype(str)

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /