This is an automated email from the ASF dual-hosted git repository.

kabhwan pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/spark.git


The following commit(s) were added to refs/heads/master by this push:
     new 3cdbc176337 [SPARK-42668][SS][FOLLOW-UP] Do not add file scheme to the 
Windows path
3cdbc176337 is described below

commit 3cdbc176337bdb0dbe11a120b0416b4f9014d8d9
Author: Hyukjin Kwon <gurwls...@apache.org>
AuthorDate: Tue May 9 15:51:16 2023 +0900

    [SPARK-42668][SS][FOLLOW-UP] Do not add file scheme to the Windows path
    
    ### What changes were proposed in this pull request?
    
    This PR is a followup of https://github.com/apache/spark/pull/40273 that 
avoids adding `file://` into Windows path.
    
    ### Why are the changes needed?
    
    To fix the broken AppVeyor build
    
    ```
    == Failed 
======================================================================
    -- 1. Error ('test_streaming.R:55'): read.stream, write.stream, 
awaitTermination
    Error: Error in start : org.apache.hadoop.HadoopIllegalArgumentException: 
Unsupported name: has scheme but relative path-part
    ...
    ```
    
    See also 
https://ci.appveyor.com/project/ApacheSoftwareFoundation/spark/builds/46451196
    
    ### Does this PR introduce _any_ user-facing change?
    
    No to end users because the change has not been released out yet.
    
    ### How was this patch tested?
    
    AppVeyor build on windows.
    
    Closes #41098 from HyukjinKwon/appveyor-testing.
    
    Authored-by: Hyukjin Kwon <gurwls...@apache.org>
    Signed-off-by: Jungtaek Lim <kabhwan.opensou...@gmail.com>
---
 .../spark/sql/execution/streaming/ResolveWriteToStream.scala  | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git 
a/sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/ResolveWriteToStream.scala
 
b/sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/ResolveWriteToStream.scala
index 9cd1e60b564..35bb7db6a6e 100644
--- 
a/sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/ResolveWriteToStream.scala
+++ 
b/sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/ResolveWriteToStream.scala
@@ -82,8 +82,15 @@ object ResolveWriteToStream extends Rule[LogicalPlan] with 
SQLConfHelper {
           s" circumstances, please set 
${SQLConf.FORCE_DELETE_TEMP_CHECKPOINT_LOCATION.key} to" +
           s" true. Important to know deleting temp checkpoint folder is best 
effort.")
         // SPARK-42676 - Write temp checkpoints for streaming queries to local 
filesystem
-        // even if default FS is set differently
-        "file://" + tempDir
+        // even if default FS is set differently.
+        // This is a band-aid fix. Ideally we should convert `tempDir` to 
URIs, but there
+        // are many legacy behaviors related to this.
+        if (Utils.isWindows) {
+          // For Windows local path, we can't treat that as a URI with file 
scheme.
+          tempDir
+        } else {
+          "file://" + tempDir
+        }
       } else {
         throw QueryCompilationErrors.checkpointLocationNotSpecifiedError()
       }


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@spark.apache.org
For additional commands, e-mail: commits-h...@spark.apache.org

Reply via email to