Skip to content

BUG: Fix index order for Index.intersection() #15583

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 23 commits into from
Closed

BUG: Fix index order for Index.intersection() #15583

Changes from 1 commit
Commits
Show all changes
23 commits
Select commit Hold shift + click to select a range
c12bb3f
BUG: Fix index order for Index.intersection()
Mar 5, 2017
c2a8dc3
Implement tests
Mar 6, 2017
a4ead99
Fix typo
Mar 6, 2017
e7bcd28
Fix typo in documentation
Mar 8, 2017
d9e29f8
Create new DatetimeIndex from the Index.intersection result
Mar 8, 2017
1197b99
Fix DataFrame column order when read from HDF file
Mar 12, 2017
784fe75
Fix error: line too long
Mar 12, 2017
b977278
Address requested changes
Mar 14, 2017
ec836bd
Fix Travis errors
Mar 15, 2017
047b513
Address requested changes
Mar 15, 2017
c96306d
Add sort argument to Index.join
Mar 15, 2017
f0d9d03
Add whatsnew
Mar 15, 2017
ef2581e
Fix Travis error
Mar 15, 2017
3c200fe
Add new tests
Mar 18, 2017
33eb740
Address requested changes
Mar 18, 2017
654288b
Fix Travis errors
Mar 18, 2017
5bf1508
Address requested changes
Mar 20, 2017
9e39794
Address requested changes
Mar 20, 2017
968c7f1
DOC/TST: change to use parameterization
jreback Mar 25, 2017
8d2e9cc
Address requested changes
Mar 28, 2017
73df69e
Address requested changes
Mar 28, 2017
64e86a4
Fix test on right join
Mar 28, 2017
2d4e143
Fix pytest fixture name collision
Mar 28, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Fix error: line too long
  • Loading branch information
@jreback
Albert Villanova del Moral authored and jreback committed Mar 25, 2017
commit 784fe75997c88a2298f2cea634e6f22a9fcc1af6
3 changes: 2 additions & 1 deletion pandas/tseries/index.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1203,7 +1203,8 @@ def intersection(self, other):
not other.offset.isAnchored() or
(not self.is_monotonic or not other.is_monotonic)):
result = Index.intersection(self, other)
result = self._simple_new(result._values, name=result.name, tz=result.tz)
result = self._simple_new(result._values, name=result.name,
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

use _shallow_copy rather than _simple_new

Copy link
Contributor Author

@albertvillanova albertvillanova Mar 13, 2017 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If I use _shallow_copy, the result will inherit all self attributes (e.g., freq) and this might not be the expected result. For example, this test would fail:

base = date_range('6/1/2000', '6/30/2000', freq='D', name='idx')

rng4 = date_range('7/1/2000', '7/31/2000', freq='D', name='idx') 
expected4 = DatetimeIndex([], name='idx')

because result would inherit self freq = 'D'.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

maybe you misunderstand, simply pass the new freq

result = self._shallow_copy(...., freq=....)

We don't use _simple_new at all (outside of a few specific places), this is not one of them. This is all documented next to these functions.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Might I simply use the constructor DatetimeIndex()? I ask this because I think the result freq is not known in advance.

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

no, use _shallow_copy with freq=None, which will not infer it, and then you can reset the freq after

tz=result.tz)
if result.freq is None:
result.offset = to_offset(result.inferred_freq)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this change related to the PR?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, as the simple result of Index.intersection inherits all attributes from other, and this is not the expected result. Therefore, it is necessary to create a new DatetimeIndex using only three of the Index.intersection attributes: _values, name and tz.

return result
Expand Down