[
https://issues.apache.org/jira/browse/PHOENIX-2351?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14983698#comment-14983698
]
Maryann Xue commented on PHOENIX-2351:
--------------------------------------
Sorry, [~aliciashu], my description wasn't accurate. I suppose your example
works fine with the current implementation, and actually the change was to make
it work for both rs.getString("t1.a") and rs.getString("a") if there's no
conflict of "a" (even there is, it should be OK too according to JDBC spec).
And I was NOT saying you should not use XML output. My suggestion was that you
find out which code in Phoenix is responsible for printing the column names and
make changes there.
> Phoenix query table alias name is not coming the output xml file
> ----------------------------------------------------------------
>
> Key: PHOENIX-2351
> URL: https://issues.apache.org/jira/browse/PHOENIX-2351
> Project: Phoenix
> Issue Type: Bug
> Reporter: Alicia Ying Shu
> Assignee: Alicia Ying Shu
> Attachments: PHOENIX-2351.patch
>
>
> Phoenix query table alias name is not coming the output xml file.
> From sqlline:
> !outputformat xmlattr
> SELECT /*+ SORTED_MERGE_JOIN */ * FROM LARGE_TABLE AS L INNER JOIN
> GIGANTIC_TABLE AS G ON L.UNSIG_ID=G.BIG_ID LIMIT 10000;
> When alias name is used in the query (in this example, table alias name is
> “L”) then the expected xml output need to have the alias name,
> Current output:
> <resultset>
> <result ID="0" UNSIG_ID="0" BIG_ID="0" UNSIG_LONG_ID="0" ...
> ......
> Expected output:
> <resultset>
> <result L.ID="0" UNSIG_ID="0" G.BIG_ID="0" UNSIG_LONG_ID="0" ...
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
