You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@spark.apache.org by GitBox <gi...@apache.org> on 2019/04/28 15:47:55 UTC
[GitHub] [spark] HyukjinKwon commented on a change in pull request #24387:
[SPARK-27473][SQL] Support filter push down for status fields in binary
file data source
HyukjinKwon commented on a change in pull request #24387: [SPARK-27473][SQL] Support filter push down for status fields in binary file data source
URL: https://github.com/apache/spark/pull/24387#discussion_r279200295
##########
File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/binaryfile/BinaryFileFormat.scala
##########
@@ -142,26 +150,62 @@ class BinaryFileFormat extends FileFormat with DataSourceRegister {
object BinaryFileFormat {
- private val fileStatusSchema = StructType(
- StructField("path", StringType, false) ::
- StructField("modificationTime", TimestampType, false) ::
- StructField("length", LongType, false) :: Nil)
+ private[binaryfile] val PATH = "path"
+ private[binaryfile] val MODIFICATION_TIME = "modificationTime"
+ private[binaryfile] val LENGTH = "length"
+ private[binaryfile] val CONTENT = "content"
/**
* Schema for the binary file data source.
*
* Schema:
+ * - path (StringType): The path of the file.
+ * - modificationTime (TimestampType): The modification time of the file.
+ * In some Hadoop FileSystem implementation, this might be unavailable and fallback to some
+ * default value.
+ * - length (LongType): The length of the file in bytes.
* - content (BinaryType): The content of the file.
- * - status (StructType): The status of the file.
- * - path (StringType): The path of the file.
- * - modificationTime (TimestampType): The modification time of the file.
- * In some Hadoop FileSystem implementation, this might be unavailable and fallback to some
- * default value.
- * - length (LongType): The length of the file in bytes.
*/
val schema = StructType(
- StructField("content", BinaryType, true) ::
- StructField("status", fileStatusSchema, false) :: Nil)
+ StructField(PATH, StringType, false) ::
+ StructField(MODIFICATION_TIME, TimestampType, false) ::
+ StructField(LENGTH, LongType, false) ::
+ StructField(CONTENT, BinaryType, true) :: Nil)
+
+ private[binaryfile] def createFilterFunction(filter: Filter): FileStatus => Boolean = {
+ filter match {
+ case And(left, right) =>
+ s => createFilterFunction(left)(s) && createFilterFunction(right)(s)
+ case Or(left, right) =>
+ s => createFilterFunction(left)(s) || createFilterFunction(right)(s)
+ case Not(child) =>
+ s => !createFilterFunction(child)(s)
+
+ case LessThan(LENGTH, value: Long) =>
+ _.getLen < value
+ case LessThanOrEqual(LENGTH, value: Long) =>
+ _.getLen <= value
+ case GreaterThan(LENGTH, value: Long) =>
+ _.getLen > value
+ case GreaterThanOrEqual(LENGTH, value: Long) =>
+ _.getLen >= value
+ case EqualTo(LENGTH, value: Long) =>
+ _.getLen == value
+
+ case LessThan(MODIFICATION_TIME, value: Timestamp) =>
+ _.getModificationTime < value.getTime
+ case LessThanOrEqual(MODIFICATION_TIME, value: Timestamp) =>
+ _.getModificationTime <= value.getTime
+ case GreaterThan(MODIFICATION_TIME, value: Timestamp) =>
+ _.getModificationTime > value.getTime
+ case GreaterThanOrEqual(MODIFICATION_TIME, value: Timestamp) =>
+ _.getModificationTime >= value.getTime
+ case EqualTo(MODIFICATION_TIME, value: Timestamp) =>
+ _.getModificationTime == value.getTime
Review comment:
Looks `EqualNullSafe` and `In` are missing.
----------------------------------------------------------------
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:
users@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org