[ARROW-10264] [C++][Python] Parquet test failing with HadoopFileSystem URI

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 3.0.0
• Component/s: Python
• Labels: filesystem hdfs pull-request-available
• External issue URL: https://github.com/apache/arrow/issues/26260
• Assignee: Antoine Pitrou
• Reporter: Joris Van den Bossche
• Created: 10/Oct/20 08:16
• Updated: 11/Jan/23 08:12
• Resolved: 14/Jan/21 17:02

Description

Follow-up on ARROW-10175. In the HDFS integration tests, there is a test using a URI failing if we use the new filesystem / dataset implementation:

FAILED opt/conda/envs/arrow/lib/python3.7/site-packages/pyarrow/tests/test_hdfs.py::TestLibHdfs::test_read_multiple_parquet_files_with_uri

fails with

pyarrow.lib.ArrowInvalid: Path '/tmp/pyarrow-test-838/multi-parquet-uri-48569714efc74397816722c9c6723191/0.parquet' is not relative to '/user/root'

while it is passing a URI (and not a filesystem object) to parquet.read_table, and the new filesystems/dataset implementation should be able to handle URIs.

cc apitrou

Issue Links

• links to: GitHub Pull Request #9192