impala-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alex Behm (Code Review)" <>
Subject [Impala-ASF-CR] IMPALA-3726: Add support for Kudu-specific column options
Date Mon, 14 Nov 2016 20:07:13 GMT
Alex Behm has posted comments on this change.

Change subject: IMPALA-3726: Add support for Kudu-specific column options

Patch Set 5:

File fe/src/main/cup/sql-parser.cup:

Line 402: nonterminal String ident_or_keyword, ident_or_default;
Can you move ident_or_keyword after the terminals and add a comment that this should be used
instead of IDENT with an explanation why.

I think we should consider reverting these changes in a future compatibility breaking release
because it adds a burden to all future parser changes. Please file a JIRA for that labeled
with "incompatibility".

Line 1498: // TODO: Consider introducing a DEFAULT keyword
File fe/src/main/java/org/apache/impala/analysis/

Line 242:             "on non-nullable column: '%s'", getColName()));
> Agree. I was thinking of a scenario where Kudu is configured to make column
Let's settle on what to do here.
File fe/src/main/java/org/apache/impala/analysis/

Line 220:     // TODO: Similar checks are applied for range partition values in
Can we address this immediately after this patch goes in? The code is somewhat tricky/subtle
and we now have it in two places. I still think there should be a final cast after L252 although
it's not strictly necessary to make the currently desired features work.
File tests/query_test/

Line 59:     encodings = ["ENCODING AUTO_ENCODING", ""]
try some non-default ones?

should we really print AUTO or DEFAULT values?

To view, visit
To unsubscribe, visit

Gerrit-MessageType: comment
Gerrit-Change-Id: I727b9ae1b7b2387db752b58081398dd3f3449c02
Gerrit-PatchSet: 5
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-Owner: Dimitris Tsirogiannis <>
Gerrit-Reviewer: Alex Behm <>
Gerrit-Reviewer: Dimitris Tsirogiannis <>
Gerrit-HasComments: Yes

View raw message