Fokko commented on code in PR #7163:
URL: https://github.com/apache/iceberg/pull/7163#discussion_r1145289020
##########
python/pyiceberg/io/pyarrow.py:
##########
@@ -517,15 +536,22 @@ def _file_to_table(
if file_schema is None:
raise ValueError(f"Missing Iceberg schema in Metadata for file:
{path}")
- arrow_table = pq.read_table(
- source=fout,
- schema=parquet_schema,
- pre_buffer=True,
- buffer_size=8 * ONE_MEGABYTE,
- filters=pyarrow_filter,
+ fragment_scanner = ds.Scanner.from_fragment(
+ fragment=fragment,
+ schema=physical_schema,
+ filter=pyarrow_filter,
columns=[col.name for col in file_project_schema.columns],
)
+ if limit:
+ arrow_table = fragment_scanner.head(limit)
+ with rows_counter.get_lock():
Review Comment:
Ah, thanks for the explanation. Currently, we don't do multi-processing, but
multi-threading. I did some extensive testing and noticed that multi-processing
wasn't substantially faster than multithreading. Probably because most time is
spent in fetching the files, and reading the data, which all happens in Arrow,
which bypasses the GIL.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
