maxburke edited a comment on pull request #9331:
URL: https://github.com/apache/arrow/pull/9331#issuecomment-770552255
@jorgecarleitao Here's the schema:
```
>>> t = pq.ParquetFile('problem.parquet')
>>> t
<pyarrow.parquet.ParquetFile object at 0x10f345f40>
>>> t.schema
<pyarrow._parquet.ParquetSchema object at 0x10f350a40>
required group field_id=0 schema {
optional int96 field_id=1 ul_observation_date;
optional binary field_id=2 document;
optional binary field_id=3 report_id (String);
optional binary field_id=4 type (String);
optional binary field_id=5 name (String);
optional binary field_id=6 year (String);
optional binary field_id=7 vendor (String);
optional binary field_id=8 direction (String);
optional binary field_id=9 geometry (String);
optional binary field_id=10 ul_node_id (String);
optional binary field_id=11 classification (String);
optional double field_id=12 posted_speed;
optional int64 field_id=13 __index_level_0__;
}
```
The query:
`SELECT "ul_node_id", "ul_observation_date", "vendor" FROM parquet_table
WHERE (("vendor" = 'foo') AND "ul_observation_date" >=
to_timestamp('2020-09-01T14:30:00.000Z') AND "ul_observation_date" <=
to_timestamp('2020-09-04T00:00:00.000Z')) ORDER BY ul_node_id ASC NULLS LAST;`
(in this file there are no rows where vendor = 'foo')
The error message received is:
```
thread 'tokio-runtime-worker' panicked at 'compute_op failed to downcast
array', datafusion/src/physical_plan/expressions.rs:1522:31
```
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
