Github user davies commented on a diff in the pull request:
https://github.com/apache/spark/pull/13442#discussion_r65937515
--- Diff:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/FileSourceStrategy.scala
---
@@ -143,8 +143,18 @@ private[sql] object FileSourceStrategy extends
Strategy with Logging {
val defaultParallelism =
files.sparkSession.sparkContext.defaultParallelism
val totalBytes = selectedPartitions.flatMap(_.files.map(_.getLen
+ openCostInBytes)).sum
val bytesPerCore = totalBytes / defaultParallelism
- val maxSplitBytes = Math.min(defaultMaxSplitBytes,
- Math.max(openCostInBytes, bytesPerCore))
+ val maxSplitBytes = {
+ // Since `LineRecordReader` in hadoop cannot split files
compressed by some codecs
+ // (e.g., gzip), check if all the input files are splittable
here.
+ if (files.fileFormat.canSplitFiles(
--- End diff --
I think we should check the file whether it's splittable or not one by one,
do not split it if it's not.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
