[ https://issues.apache.org/jira/browse/CALCITE-2708?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16699487#comment-16699487 ]
Julian Hyde commented on CALCITE-2708: -------------------------------------- Sounds reasonable. [~dawidwys], On a separate matter, I'm working on CALCITE-1935 again, which duplicates the work you've done in Flink somewhat. (By the way, congratulations on all the good work you've been doing there.) I don't think it's possible to share code, but I do think it's possible to share test cases. Is there any chance you contribute to [match.iq|https://github.com/julianhyde/calcite/blob/1935-match-recognize/core/src/test/resources/sql/match.iq]? Then we'll end up with a set of SQL tests and results (including answers we've checked against Oracle) so we'd know that both Calcite's and Flink's implementation are compliant? > UDAGs are not verified for single pattern variable in MATCH_RECOGNIZE > --------------------------------------------------------------------- > > Key: CALCITE-2708 > URL: https://issues.apache.org/jira/browse/CALCITE-2708 > Project: Calcite > Issue Type: Bug > Reporter: Dawid Wysakowicz > Assignee: Julian Hyde > Priority: Major > Labels: match > > Query: > {code} > SELECT * > FROM Ticker > MATCH_RECOGNIZE ( > ORDER BY proctime > MEASURES > udagg(A.price + B.tax) AS taxedPrice > PATTERN (A B) > DEFINE > A AS A.symbol = 'a' > ) AS T > {code} > should fail with ValidationException, because udagg is applied to different > pattern variables. -- This message was sent by Atlassian JIRA (v7.6.3#76005)