This is an automated email from the ASF dual-hosted git repository.
wesm pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/arrow.git
The following commit(s) were added to refs/heads/master by this push:
new d64e144 ARROW-2170: [Python] construct_metadata fails on reading
files where no index was preserved
d64e144 is described below
commit d64e144cd4de03ad517f0d38ea72c2f308a8ea35
Author: Uwe L. Korn <uw...@xhochy.com>
AuthorDate: Mon Feb 19 11:14:15 2018 -0500
ARROW-2170: [Python] construct_metadata fails on reading files where no
index was preserved
cc @cpcloud
The result was that we only persisted empty Pandas metadata for these files.
Author: Uwe L. Korn <uw...@xhochy.com>
Closes #1623 from xhochy/ARROW-2170 and squashes the following commits:
ba1ce3f3 [Uwe L. Korn] ARROW-2170: [Python] construct_metadata fails on
reading files where no index was preserved
---
python/pyarrow/pandas_compat.py | 7 +++++--
python/pyarrow/tests/test_parquet.py | 3 +++
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/python/pyarrow/pandas_compat.py b/python/pyarrow/pandas_compat.py
index f5e56a9..e8fa83f 100644
--- a/python/pyarrow/pandas_compat.py
+++ b/python/pyarrow/pandas_compat.py
@@ -197,8 +197,11 @@ def construct_metadata(df, column_names, index_levels,
index_column_names,
-------
dict
"""
- df_types = types[:-len(index_levels)]
- index_types = types[-len(index_levels):]
+ # Use ntypes instead of Python shorthand notation [:-len(x)] as [:-0]
+ # behaves differently to what we want.
+ ntypes = len(types)
+ df_types = types[:ntypes - len(index_levels)]
+ index_types = types[ntypes - len(index_levels):]
column_metadata = [
get_column_metadata(
diff --git a/python/pyarrow/tests/test_parquet.py
b/python/pyarrow/tests/test_parquet.py
index c49f3d3..bd76feb 100644
--- a/python/pyarrow/tests/test_parquet.py
+++ b/python/pyarrow/tests/test_parquet.py
@@ -215,6 +215,9 @@ def
test_pandas_parquet_2_0_rountrip_read_pandas_no_index_written(tmpdir):
arrow_table = pa.Table.from_pandas(df, preserve_index=False)
js = json.loads(arrow_table.schema.metadata[b'pandas'].decode('utf8'))
assert not js['index_columns']
+ # ARROW-2170
+ # While index_columns should be empty, columns needs to be filled still.
+ assert js['columns']
_write_table(arrow_table, filename.strpath, version="2.0",
coerce_timestamps='ms')
--
To stop receiving notification emails like this one, please contact
w...@apache.org.
