Skip to content

Conversation

SylvainCorlay
Copy link
Contributor

This solves #8737 in the case where there is at least one boolean indexer. The patch of #8739 does not allow the following:

import pandas as pd
import numpy as np
multi_index = pd.MultiIndex.from_product((['foo', 'bar', 'baz'], ['alpha', 'beta']))
df = pd.DataFrame(np.random.randn(5, 6), index=range(5), columns=multi_index)
df = df.sortlevel(0, axis=1)
df.loc[:, (['foo'], [])]

@SylvainCorlay
Copy link
Contributor Author

The Travis failure seems unrelated to this change.

@jreback
Copy link
Contributor

jreback commented Nov 10, 2014

pls add your example as a test (and put it after the 8737 test)

@SylvainCorlay
Copy link
Contributor Author

Done.

@jreback
Copy link
Contributor

jreback commented Nov 10, 2014

don't worry about that failing test - sep issue

also pls add a release note in v0.15.2 (u can ref this issue)

@SylvainCorlay
Copy link
Contributor Author

here it is

@jreback jreback added Bug Indexing Related to indexing on series/frames, not to indexes themselves MultiIndex labels Nov 11, 2014
@jreback jreback added this to the 0.15.2 milestone Nov 11, 2014
@jreback
Copy link
Contributor

jreback commented Nov 11, 2014

can you sqush the commits? thanks.

@SylvainCorlay
Copy link
Contributor Author

Rebased.

jreback added a commit that referenced this pull request Nov 11, 2014
Allowing empty slice in multi-indices
@jreback jreback merged commit 81c8a5f into pandas-dev:master Nov 11, 2014
@jreback
Copy link
Contributor

jreback commented Nov 11, 2014

thank you for the fix!

@SylvainCorlay SylvainCorlay deleted the empty_slice branch November 11, 2014 00:12
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Bug Indexing Related to indexing on series/frames, not to indexes themselves MultiIndex
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants