[GitHub] [hadoop] mukund-thakur commented on a change in pull request #3534: HADOOP-17409. Remove s3guard from S3A module

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

mukund-thakur commented on a change in pull request #3534:
URL: https://github.com/apache/hadoop/pull/3534#discussion_r783712717



##########
File path: hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/impl/RenameOperation.java
##########
@@ -724,4 +608,22 @@ private String maybeAddTrailingSlash(String key) {
     }
   }
 
+  /**
+   * Convert a passed in exception (expected to be an IOE or AWS exception)
+   * into an IOException.
+   * @param ex exception caught
+   * @return the exception to throw in the failure handler.
+   */
+  protected IOException convertToIOException(final Exception ex) {
+    if (ex instanceof IOException) {
+      return (IOException) ex;

Review comment:
       Why casting again?

##########
File path: hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AFileSystem.java
##########
@@ -3821,23 +3481,14 @@ S3AFileStatus s3GetFileStatus(final Path path,
         // by the time of listing so that the response includes some
         // which have not.
 
-        int listSize;
-        if (tombstones == null) {
-          // no tombstones so look for a marker and at least one child.
-          listSize = 2;
-        } else {
-          // build a listing > tombstones. If the caller has many thousands
-          // of tombstones this won't work properly, which is why pruning
-          // of expired tombstones matters.
-          listSize = Math.min(2 + tombstones.size(), Math.max(2, maxKeys));
-        }
+        final int listSize = 2;
         S3ListRequest request = createListObjectsRequest(dirKey, "/",
             listSize);
         // execute the request
         S3ListResult listResult = listObjects(request,
             getDurationTrackerFactory());
 
-        if (listResult.hasPrefixesOrObjects(contextAccessors, tombstones)) {
+        if (listResult.hasPrefixesOrObjects(contextAccessors, null)) {

Review comment:
       Can we remove the tombstones field altogether? I suppose this was only for S3Guard entires.

##########
File path: hadoop-tools/hadoop-aws/src/main/java/org/apache/hadoop/fs/s3a/S3AFileSystem.java
##########
@@ -3316,31 +3111,19 @@ protected void maybeCreateFakeParentDirectory(Path path)
       }
     }
     // Here we have a directory which may or may not be empty.
-    // So we update the metastore and return.
-    return S3Guard.dirListingUnion(
-            metadataStore,
-            path,
-            statusesAssumingNonEmptyDir.getLeft(),
-            statusesAssumingNonEmptyDir.getMiddle(),
-            allowAuthoritative(path),
-            ttlTimeProvider, p ->
-                    listing.createProvidedFileStatusIterator(
-                    dirMetaToStatuses(statusesAssumingNonEmptyDir.getMiddle()),
-                    ACCEPT_ALL,
-                    Listing.ACCEPT_ALL_BUT_S3N));
+    return statusIt;
   }
 
   /**
    * Is a path to be considered as authoritative?
-   * True iff there is an authoritative metastore or if there
-   * is a non-auth store with the supplied path under
+   * is a  store with the supplied path under
    * one of the paths declared as authoritative.
    * @param path path
    * @return true if the path is auth
    */
   public boolean allowAuthoritative(final Path path) {
     return S3Guard.allowAuthoritative(path, this,
-        allowAuthoritativeMetadataStore, allowAuthoritativePaths);
+        allowAuthoritativePaths);

Review comment:
       Should this be deprecated?




-- 
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.

To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org