impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tim Armstrong (Code Review)" <>
Subject [Impala-ASF-CR] IMPALA-3931: arbitrary fixed-size uda intermediate types
Date Thu, 03 Aug 2017 17:27:45 GMT
Tim Armstrong has posted comments on this change.

Change subject: IMPALA-3931: arbitrary fixed-size uda intermediate types

Patch Set 6:

Commit Message:

> I think FIXED_UDA_INTERMEDIATE would be more precise/clear, especially sinc
It wasn't too hard with some sed -i.

PS6, Line 37: Testing:
> any corner cases to consider for analytic fns? I think it should work fine,
I couldn't think of any. I think it should work fine too. I agree we don't have many tests
where analytics spill. I added a couple more  tests in
but we could probably do with more.

At least the spilling logic in the analytic is straightforward - most of the complication
is in BufferedTupleStream which has a reasonable number of unit tests.
File be/src/runtime/

PS6, Line 172: TypeToOdbcString
> should types that aren't exposed raise a dcheck?
Probably. Changed that in this case because we never will expose it.
File fe/src/main/java/org/apache/impala/catalog/

PS6, Line 1000: TODO
> ?

To view, visit
To unsubscribe, visit

Gerrit-MessageType: comment
Gerrit-Change-Id: Ife90cf27989f98ffb5ef5c39f1e09ce92e8cb87c
Gerrit-PatchSet: 6
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Tim Armstrong <>
Gerrit-Reviewer: Matthew Jacobs <>
Gerrit-Reviewer: Michael Ho
Gerrit-Reviewer: Tim Armstrong <>
Gerrit-HasComments: Yes

View raw message