[GitHub] [incubator-pinot] kkrugler commented on a change in pull request #6506: Fix hadoop batch ingest

[GitBox <gi...@apache.org>]

kkrugler commented on a change in pull request #6506:
URL: https://github.com/apache/incubator-pinot/pull/6506#discussion_r568254617



##########
File path: pinot-plugins/pinot-batch-ingestion/pinot-batch-ingestion-hadoop/src/main/java/org/apache/pinot/plugin/ingestion/batch/hadoop/HadoopSegmentGenerationJobRunner.java
##########
@@ -208,11 +220,12 @@ public void run()
     } else {
       LOGGER.info("Creating segments with data files: {}", filteredFiles);
       for (int i = 0; i < numDataFiles; i++) {
-        String dataFilePath = filteredFiles.get(i);
-
-        File localFile = new File("tmp");
+        // Typically PinotFS implementations list files without a protocol, so we lose (for example) the
+        // hdfs:// portion of the path. Call getFileURI() to fix this up.
+        URI inputFileURI = SegmentGenerationUtils.getFileURI(filteredFiles.get(i), inputDirURI);

Review comment:
       It should be fixed in Standalone. I could file an issue for Spark, lmk.




----------------------------------------------------------------
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



---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@pinot.apache.org
For additional commands, e-mail: commits-help@pinot.apache.org