pvary commented on code in PR #10935:
URL: https://github.com/apache/iceberg/pull/10935#discussion_r1767544018
##########
core/src/test/java/org/apache/iceberg/TestBaseIncrementalChangelogScan.java:
##########
@@ -132,6 +131,175 @@ public void testFileDeletes() {
assertThat(t1.existingDeletes()).as("Must be no deletes").isEmpty();
}
+ @TestTemplate
+ public void testRowDeletes() {
+ assumeThat(formatVersion).isEqualTo(2);
+
+ table
+ .newFastAppend()
+ .appendFile(FILE_A)
+ .appendFile(FILE_A2)
+ .appendFile(FILE_B)
+ .appendFile(FILE_C)
+ .commit();
+ Snapshot snap1 = table.currentSnapshot();
+
+ // position delete
+ table.newRowDelta().addDeletes(FILE_B_DELETES).commit();
+ Snapshot snap2 = table.currentSnapshot();
+
+ // equality delete
+ table.newRowDelta().addDeletes(FILE_C2_DELETES).commit();
+ Snapshot snap3 = table.currentSnapshot();
+
+ // mix of position and equality deletes
+
table.newRowDelta().addDeletes(FILE_A_DELETES).addDeletes(FILE_A2_DELETES).commit();
Review Comment:
This is the question about how you see the compaction where we rewrite the
files with the expressed goal to keep the actual content unchanged.
- Is this a change where we remove and add files which are coincidentally
contain the same data?
- Is this a change where no data has changed?
I see the reasoning behind both cases, but using the first approach would
most probably make the changelog scan unfesable for cases where compaction
happens on the table.
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
