Anurag-S1ngh commented on code in PR #4407:
URL: https://github.com/apache/iceberg-python/pull/4407#discussion_r4640182625


##########
tests/expressions/test_projection.py:
##########
@@ -320,8 +347,8 @@ def test_int_truncate_projection(schema: Schema, 
truncate_int_spec: PartitionSpe
     expected = [
         NotNull("id_trunc"),
         IsNull("id_trunc"),
-        LessThanOrEqual("id_trunc", 0),
-        LessThanOrEqual("id_trunc", 10),
+        Or(LessThanOrEqual("id_trunc", 0), EqualTo("id_trunc", 
9223372036854775806)),
+        Or(LessThanOrEqual("id_trunc", 10), EqualTo("id_trunc", 
9223372036854775806)),

Review Comment:
   The wrapped value is what both writers actually produce for the type 
minimum: Java's truncate uses floorMod and wraps the same way, and our writer 
does too, so files already in the wild can carry this partition. On read, the 
current planning silently drops those rows. That is why I think the read side 
should handle it, even if the partition value itself looks odd. Java's range 
projection has the same gap, so this also goes a bit beyond what Java does 
today.



-- 
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]

Reply via email to