Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/3670#discussion_r23972092
--- Diff: core/src/main/scala/org/apache/spark/util/Utils.scala ---
@@ -555,17 +586,42 @@ private[spark] object Utils extends Logging {
uc.setReadTimeout(timeout)
uc.connect()
val in = uc.getInputStream()
- downloadFile(url, in, tempFile, targetFile, fileOverwrite)
+ downloadFile(url, in, targetFile, fileOverwrite)
case "file" =>
// In the case of a local file, copy the local file to the target
directory.
// Note the difference between uri vs url.
val sourceFile = if (uri.isAbsolute) new File(uri) else new
File(url)
copyFile(url, sourceFile, targetFile, fileOverwrite)
case _ =>
- // Use the Hadoop filesystem library, which supports file://,
hdfs://, s3://, and others
val fs = getHadoopFileSystem(uri, hadoopConf)
- val in = fs.open(new Path(uri))
- downloadFile(url, in, tempFile, targetFile, fileOverwrite)
+ val path = new Path(uri)
+ fetchHcfsFile(path, new File(targetDir, path.getName), fs, conf,
hadoopConf, fileOverwrite)
+ }
+ }
+
+ /**
+ * Fetch a file or directory from a Hadoop-compatible filesystem.
+ *
+ * Visible for testing
+ */
+ private[spark] def fetchHcfsFile(
+ path: Path,
+ targetDir: File,
+ fs: FileSystem,
+ conf: SparkConf,
+ hadoopConf: Configuration,
+ fileOverwrite: Boolean): Unit = {
+ targetDir.mkdir()
--- End diff --
Should we check the return value here so that we get a more informative
message if an error occurs?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
