Github user HyukjinKwon commented on a diff in the pull request:
https://github.com/apache/spark/pull/22611#discussion_r222167036
--- Diff:
external/avro/src/main/scala/org/apache/spark/sql/avro/AvroFileFormat.scala ---
@@ -100,6 +77,50 @@ private[avro] class AvroFileFormat extends FileFormat
}
}
+ private def inferAvroSchemaFromFiles(
+ files: Seq[FileStatus],
+ conf: Configuration,
+ ignoreExtension: Boolean): Schema = {
+ val ignoreCorruptFiles = SQLConf.get.ignoreCorruptFiles
+ // Schema evolution is not supported yet. Here we only pick first
random readable sample file to
+ // figure out the schema of the whole dataset.
+ val avroReader = files.iterator.map { f =>
+ val path = f.getPath
+ if (!ignoreExtension && !path.getName.endsWith(".avro")) {
+ None
+ } else {
+ val in = new FsInput(path, conf)
+ try {
+ Some(DataFileReader.openReader(in, new
GenericDatumReader[GenericRecord]()))
+ } catch {
+ case e: IOException =>
+ if (ignoreCorruptFiles) {
+ logWarning(s"Skipped the footer in the corrupted file:
$path", e)
+ None
+ } else {
+ throw new SparkException(s"Could not read file: $path", e)
+ }
+ } finally {
+ in.close()
+ }
+ }
+ }.collectFirst {
+ case Some(reader) => reader
+ }
+
+ avroReader match {
+ case Some(reader) =>
+ try {
--- End diff --
ditto for `tryWithResource`
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
