Repository: arrow Updated Branches: refs/heads/master 81be9c667 -> 03dce9dca
ARROW-900: [Python] Fix UnboundLocalError in ParquetDatasetPiece.read Author: Wes McKinney <wes.mckin...@twosigma.com> Closes #607 from wesm/ARROW-900 and squashes the following commits: 81f8394 [Wes McKinney] Fix UnboundLocalError in ParquetDatasetPiece.read Project: http://git-wip-us.apache.org/repos/asf/arrow/repo Commit: http://git-wip-us.apache.org/repos/asf/arrow/commit/03dce9dc Tree: http://git-wip-us.apache.org/repos/asf/arrow/tree/03dce9dc Diff: http://git-wip-us.apache.org/repos/asf/arrow/diff/03dce9dc Branch: refs/heads/master Commit: 03dce9dcab1df587f2293decf49708f872aaad3d Parents: 81be9c6 Author: Wes McKinney <wes.mckin...@twosigma.com> Authored: Thu Apr 27 18:11:44 2017 +0200 Committer: Uwe L. Korn <uw...@xhochy.com> Committed: Thu Apr 27 18:11:44 2017 +0200 ---------------------------------------------------------------------- python/pyarrow/parquet.py | 3 +++ python/pyarrow/tests/test_parquet.py | 14 ++++++++++++++ 2 files changed, 17 insertions(+) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/arrow/blob/03dce9dc/python/pyarrow/parquet.py ---------------------------------------------------------------------- diff --git a/python/pyarrow/parquet.py b/python/pyarrow/parquet.py index 94ad227..21359f1 100644 --- a/python/pyarrow/parquet.py +++ b/python/pyarrow/parquet.py @@ -208,6 +208,9 @@ class ParquetDatasetPiece(object): reader = self._open(open_file_func) elif file is not None: reader = ParquetFile(file) + else: + # try to read the local path + reader = ParquetFile(self.path) if self.row_group is not None: table = reader.read_row_group(self.row_group, columns=columns, http://git-wip-us.apache.org/repos/asf/arrow/blob/03dce9dc/python/pyarrow/tests/test_parquet.py ---------------------------------------------------------------------- diff --git a/python/pyarrow/tests/test_parquet.py b/python/pyarrow/tests/test_parquet.py index 8c446af..bb3a9ed 100644 --- a/python/pyarrow/tests/test_parquet.py +++ b/python/pyarrow/tests/test_parquet.py @@ -493,6 +493,20 @@ def test_read_single_row_group(): @parquet +def test_parquet_piece_read(tmpdir): + df = _test_dataframe(1000) + table = pa.Table.from_pandas(df) + + path = tmpdir.join('parquet_piece_read.parquet').strpath + pq.write_table(table, path, version='2.0') + + piece1 = pq.ParquetDatasetPiece(path) + + result = piece1.read() + assert result.equals(table) + + +@parquet def test_parquet_piece_basics(): path = '/baz.parq'