BUG: DataFrame.__setitem__ raising ValueError with string indexer and empty df and df to set #38931
Conversation
… empty df and df to set
phofl
added
the
Indexing
| # GH#38831 | ||
| df = DataFrame(columns=["A", "B"]) | ||
| other = DataFrame({"B": [1, 2]}) | ||
| df[indexer] = other |
There was a problem hiding this comment.
do we need/want to check any of the other indexers?
There was a problem hiding this comment.
Don't know if this would be helpful, the string-DataFrame case is a special case while indexer in [list, Index, Series, array] is all the same (which was working previously)
There was a problem hiding this comment.
hmm, can you check df.loc[:, indexer] ? should be the same result i think.
There was a problem hiding this comment.
No unfortunately not, in case of indexer=["B"] we align lhs and rhs, which leads to an empty result. This is losely related to our dicussion in #30439 @jbrockmendel
When indexer="B" this works as the case above
| # GH#38831 | ||
| df = DataFrame(columns=["A", "B"]) | ||
| other = DataFrame({"B": [1, 2]}) | ||
| df[indexer] = other |
There was a problem hiding this comment.
hmm, can you check df.loc[:, indexer] ? should be the same result i think.
� Conflicts: � doc/source/whatsnew/v1.3.0.rst
…d empty df and df to set (pandas-dev#38931)
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diffDont have to convert df to series, df has an index.
cc @jbrockmendel