pengzhiwei2018 commented on a change in pull request #2651:
URL: https://github.com/apache/hudi/pull/2651#discussion_r599318844
##########
File path:
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/hudi/DefaultSource.scala
##########
@@ -79,39 +82,58 @@ class DefaultSource extends RelationProvider
val allPaths = path.map(p => Seq(p)).getOrElse(Seq()) ++ readPaths
val fs = FSUtils.getFs(allPaths.head,
sqlContext.sparkContext.hadoopConfiguration)
- val globPaths = HoodieSparkUtils.checkAndGlobPathIfNecessary(allPaths, fs)
-
- val tablePath = DataSourceUtils.getTablePath(fs, globPaths.toArray)
+ // Use the HoodieFileIndex only if the 'path' has specified with no "*"
contains.
+ // And READ_PATHS_OPT_KEY has not specified.
+ // Or else we use the original way to read hoodie table.
+ val useHoodieFileIndex = path.isDefined && !path.get.contains("*") &&
+ !parameters.contains(DataSourceReadOptions.READ_PATHS_OPT_KEY)
+ val globPaths = if (useHoodieFileIndex) {
+ None
+ } else {
+ Some(HoodieSparkUtils.checkAndGlobPathIfNecessary(allPaths, fs))
+ }
+ // Get the table base path
+ val tablePath = if (globPaths.isDefined) {
+ DataSourceUtils.getTablePath(fs, globPaths.get.toArray)
+ } else {
+ val tablePathOpt = TablePathUtils.getTablePath(fs, new Path(path.get))
+ if (tablePathOpt.isPresent) {
+ tablePathOpt.get().toString
+ } else {
+ throw new HoodieException(s"Failed to get the table path, " +
+ s"Maybe ${allPaths.head} is not a hudi table path")
+ }
Review comment:
Yes, I will refactor this code.
--
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:
us...@infra.apache.org
