geoffreyclaude commented on code in PR #15560:
URL: https://github.com/apache/datafusion/pull/15560#discussion_r2027071698
##########
benchmarks/src/tpch/convert.rs:
##########
@@ -99,6 +110,11 @@ impl ConvertOpt {
if partitions > 1 {
csv =
csv.repartition(Partitioning::RoundRobinBatch(partitions))?
}
+ let csv = if self.sort {
+ csv.sort_by(vec![col(key_column_name)])?
Review Comment:
the `sort_by` means there is a single output partition, which translates to
a single output Parquet file. Enforcing multiple files while maintaining the
ordering information needs something a bit more complex than the simple
original `ctx.write_parquet(...)`...
##########
benchmarks/src/tpch/convert.rs:
##########
@@ -99,6 +110,11 @@ impl ConvertOpt {
if partitions > 1 {
csv =
csv.repartition(Partitioning::RoundRobinBatch(partitions))?
}
+ let csv = if self.sort {
+ csv.sort_by(vec![col(key_column_name)])?
Review Comment:
the `sort_by` means there is a single output partition, which translates to
a single output Parquet file. Enforcing multiple output files per table while
maintaining the ordering information needs something a bit more complex than
the simple original `ctx.write_parquet(...)`...
--
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]
