Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Sanitize column names in ParquetCachedBatchSerializer before writing to Parquet [databricks] #4258

Merged
merged 3 commits into from
Dec 7, 2021
Merged

Sanitize column names in ParquetCachedBatchSerializer before writing to Parquet [databricks] #4258

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c9229bd
Sanitize col names
razajafri Dec 1, 2021
0437c8f
addressed review comments
razajafri Dec 3, 2021
8a9a789
addressed review comments
razajafri Dec 6, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 14 additions & 2 deletions ...c/main/311+-all/scala/com/nvidia/spark/rapids/shims/v2/ParquetCachedBatchSerializer.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -565,11 +565,14 @@ class ParquetCachedBatchSerializer extends GpuCachedBatchSerializer with Arm {
})
cbRdd.mapPartitions(iter => CloseableColumnBatchIterator(iter))
} else {
val origSelectedAttributesWithUnambiguousNames =
sanitizeColumnNames(newSelectedAttributes, selectedSchemaWithNames)
val broadcastedConf = SparkSession.active.sparkContext.broadcast(conf.getAllConfs)
input.mapPartitions {
cbIter => {
new CachedBatchIteratorConsumer(cbIter, cachedSchemaWithNames, selectedSchemaWithNames,
cacheAttributes, newSelectedAttributes, broadcastedConf).getColumnBatchIterator
cacheAttributes, origSelectedAttributesWithUnambiguousNames, broadcastedConf)
.getColumnBatchIterator
}
}
}
Expand All @@ -592,11 +595,12 @@ class ParquetCachedBatchSerializer extends GpuCachedBatchSerializer with Arm {
conf: SQLConf): RDD[InternalRow] = {
val (cachedSchemaWithNames, selectedSchemaWithNames) =
getSupportedSchemaFromUnsupported(cacheAttributes, selectedAttributes)
val newSelectedAttributes = sanitizeColumnNames(selectedAttributes, selectedSchemaWithNames)
val broadcastedConf = SparkSession.active.sparkContext.broadcast(conf.getAllConfs)
input.mapPartitions {
cbIter => {
new CachedBatchIteratorConsumer(cbIter, cachedSchemaWithNames, selectedSchemaWithNames,
cacheAttributes, selectedAttributes, broadcastedConf).getInternalRowIterator
cacheAttributes, newSelectedAttributes, broadcastedConf).getInternalRowIterator
}
}
}
Expand Down Expand Up @@ -1357,6 +1361,14 @@ class ParquetCachedBatchSerializer extends GpuCachedBatchSerializer with Arm {
}
}

// We want to change the original schema to have the new names as well
private def sanitizeColumnNames(originalSchema: Seq[Attribute],
schemaToCopyNamesFrom: Seq[Attribute]): Seq[Attribute] = {
originalSchema.zip(schemaToCopyNamesFrom).map {
case (origAttribute, newAttribute) => origAttribute.withName(newAttribute.name)
}
}

private def getSupportedSchemaFromUnsupported(
cachedAttributes: Seq[Attribute],
requestedAttributes: Seq[Attribute] = Seq.empty): (Seq[Attribute], Seq[Attribute]) = {
Expand Down