Skip to content

Fix _assert_shared #2136

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

Merged
marqh merged 2 commits into from
Sep 14, 2016
Merged

Fix _assert_shared #2136

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6253fff
Fix _assert_shared
djkirkham Sep 8, 2016
aaa8c04
Modify fix and add dev comment
djkirkham Sep 14, 2016
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
11 changes: 10 additions & 1 deletion lib/iris/pandas.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
# (C) British Crown Copyright 2013 - 2015, Met Office
# (C) British Crown Copyright 2013 - 2016, Met Office
#
# This file is part of Iris.
#
Expand Down Expand Up @@ -129,6 +129,15 @@ def _assert_shared(np_obj, pandas_obj):
base = pandas_obj.base
else:
base = pandas_obj[0].base

# Prior to Pandas 0.17, when pandas_obj is a Series, pandas_obj.values
# returns a view of the underlying array, and pandas_obj.base, which calls
# pandas_obj.values.base, returns the underlying array. In 0.17 and 0.18
# pandas_obj.values returns the underlying array, so base may be None even
# if the array is shared.
if base is None:
base = pandas_obj.values

# Chase the stack of NumPy `base` references back to see if any of
# them are our original array.
while base is not None:
Expand Down