BUG: astype(str) on datetimelike index #10442 #11148
Conversation
| import pandas as pd | ||
| pd.date_range('2012-01-01', periods=4).astype(str) | ||
| result = ['2012-01-01', '2012-01-02', '2012-01-03', '2012-01-04'] | ||
| assert(type(result[0]) == np.str) |
There was a problem hiding this comment.
The test should compare the result and expected as below. I understand the result will be np.object dtype, not numpy.str.
result = pd.date_range('2012-01-01', periods=4).astype(str)
expected = pd.Index(['2012-01-01', '2012-01-02', '2012-01-03', '2012-01-04'], dtype=object)
tm.assert_index_equal(result, expected)
Also, can you add other test cases, such as different freq like H, and with timezone.
|
I added all the tests..please let me know if I need to add anything else. |
jreback
added
Datetime
|
|
||
|
|
||
| def test_datetimelike_tostring_astype_freqH(): | ||
| import pandas as pd |
There was a problem hiding this comment.
you don't need any of the imports, though you may need _skip_if_no_pytz. below
|
sounds fine now, let me know if I need to add anythingelse |
| @@ -504,6 +504,29 @@ def test_infer_freq(self): | |||
| self.assertEqual(result.freq, freq) | |||
|
|
|||
|
|
|||
| def test_datetimelike_tostring_astype(): | |||
There was a problem hiding this comment.
these should be under the TestDatetimeIndexOps class
|
pls add a note in whatsnew for 0.17.1 (enhancements). squash as well. |
|
going to need tests for |
|
@jreback |
|
rebase on master |
| tm.assert_index_equal(result, expected) | ||
|
|
||
| # test astype string with freqH and timezone | ||
| import pytz |
|
@hamedhsn can you update |
|
@jreback sorry, what do I need to update? |
|
comments above, needs a rebase as well |
|
@jreback did that. |
| @@ -17,6 +17,7 @@ Highlights include: | |||
|
|
|||
| Enhancements | |||
| ~~~~~~~~~~~~ | |||
| - adding conversion of ``DatetimeIndex`` values to strings using ``astype(str)`` (:issue:`10442`) | |||
There was a problem hiding this comment.
say DatetimeIndex now supports conversion to strings with astype(str)
|
couple of comments. pls squash as well |
|
@jreback sorry that I am asking..how can I squash the last few commits to one. I tried a few different approaches without a luck. |
|
@jreback squashed all the commits into one. |
|
you need to make sure you force push too. squash (put a s by each commit) |
|
@jreback did that. thanks |
| result = ['2012-01-01', '2012-01-02', '2012-01-03', '2012-01-04'] | ||
| assert(type(result[0]) == np.str) | ||
|
|
||
|
|
There was a problem hiding this comment.
the above test doesn't do anything. pls make it more like @sinhrks suggested
|
@jreback should be fine now. |
|
ok @hamedhsn ping on green. |
|
@jreback strange, not sure why test is failed. both sides are object type but it raises this error.. I tried the test with both side same nothing to do with astype but still get the same error |
| # BUG#10442 : testing astype(str) is correct for Series/DatetimeIndex | ||
| result = pd.Series(pd.date_range('2012-01-01', periods=3)).astype(str) | ||
| expected = pd.Series(['2012-01-01', '2012-01-02', '2012-01-03'], dtype=object) | ||
| tm.assert_index_equal(result, expected) |
There was a problem hiding this comment.
you need to use tm.assert_series_equal here (and below)
BUG pandas-dev#10442(test) : Convert datetimelike index to strings with astype(str) BUG#10422: note added bug#10442 : tests added bug#10442 : note udated BUG pandas-dev#10442(test) : Convert datetimelike index to strings with astype(str) bug#10442: fix, adding note and test bug#10442: fix, adding note and test
|
@jreback it is fine now. |
BUG: astype(str) on datetimelike index #10442
|
thanks! |
closes #10442
fix for the bug: Convert datetimelike index to strings with astype(str)
going to send pull request for test