cozos commented on a change in pull request #25899: [SPARK-29089][SQL] Parallelize blocking FileSystem calls in DataSource#checkAndGlobPathIfNecessary URL: https://github.com/apache/spark/pull/25899#discussion_r344914942
########## File path: sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/DataSource.scala ########## @@ -734,30 +735,52 @@ object DataSource extends Logging { * Checks and returns files in all the paths. */ private[sql] def checkAndGlobPathIfNecessary( - paths: Seq[String], + pathStrings: Seq[String], hadoopConf: Configuration, checkEmptyGlobPath: Boolean, checkFilesExist: Boolean): Seq[Path] = { - val allGlobPath = paths.flatMap { path => - val hdfsPath = new Path(path) - val fs = hdfsPath.getFileSystem(hadoopConf) - val qualified = hdfsPath.makeQualified(fs.getUri, fs.getWorkingDirectory) - val globPath = SparkHadoopUtil.get.globPathIfNecessary(fs, qualified) - - if (checkEmptyGlobPath && globPath.isEmpty) { - throw new AnalysisException(s"Path does not exist: $qualified") + val qualifiedPaths = pathStrings.map { pathString => + val path = new Path(pathString) + val fs = path.getFileSystem(hadoopConf) + path.makeQualified(fs.getUri, fs.getWorkingDirectory) + } + + // Split the paths into glob and non glob paths, because we don't need to do an existence check + // for globbed paths. + val (globPaths, nonGlobPaths) = qualifiedPaths.partition(SparkHadoopUtil.get.isGlobPath) + + val globbedPaths = + try { + ThreadUtils.parmap(globPaths, "globPath", 20) { globPath => + val fs = globPath.getFileSystem(hadoopConf) + val globResult = SparkHadoopUtil.get.globPath(fs, globPath) + + if (checkEmptyGlobPath && globResult.isEmpty) { + throw new AnalysisException(s"Path does not exist: $globPath") + } + + globResult + }.flatten + } catch { + case e: SparkException => throw e.getCause Review comment: SparkException comes from [ThreadUtils#parmap](https://github.com/apache/spark/blob/a6a27485851a160b9fc4bb86481c60f10573aad0/core/src/main/scala/org/apache/spark/util/ThreadUtils.scala#L285) And [ThreadUtils#awaitResult](https://github.com/apache/spark/blob/master/core/src/main/scala/org/apache/spark/util/ThreadUtils.scala#L227) Which always seems to wrap another exception i.e. never null ---------------------------------------------------------------- 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 With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org