[
https://issues.apache.org/jira/browse/DRILL-1954?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14269475#comment-14269475
]
Jacques Nadeau commented on DRILL-1954:
---------------------------------------
I just ran across this recently as well. Drill's text/csv representation when
reading is a repeated varchar but output is expected to be delineated columns.
We should really have better handling so that a single repeated varchar output
is also a handled correctly in CSV.
An ugly work around for now if doing a ctas on the individual columns. For
example:
CREATE TABLE fcc_cell AS select columns[0], columns[1], columns[2], columns[3],
..., columns[n] from dfs.`/Users/Documents/fcc/fcc_lic_vw.csv` WHERE
columns[10] IN ('"Mobile/Fixed Broadband"', '"Fixed Wireless"') AND columns[17]
= '"A"';
> Update CSV Writer to treat a single RepeatedVarChar column as separate
> columns on output rather than Erroring
> -------------------------------------------------------------------------------------------------------------
>
> Key: DRILL-1954
> URL: https://issues.apache.org/jira/browse/DRILL-1954
> Project: Apache Drill
> Issue Type: Improvement
> Components: Storage - Text & CSV
> Affects Versions: 0.6.0, 0.7.0
> Reporter: Cary Greenwood
> Fix For: 1.0.0
>
>
> I am experiencing this error when attempting to create a table. When the
> same query is executed using SELECT * the results display fine.
> 0: jdbc:drill:zk=local> CREATE TABLE fcc_cell AS select * from
> dfs.`/Users/Documents/fcc/fcc_lic_vw.csv` WHERE columns[10] IN
> ('"Mobile/Fixed Broadband"', '"Fixed Wireless"') AND columns[17] = '"A"';
> Query failed: Query stopped., Repeated types are not supported. [
> cd76d8fd-b00e-47d5-962c-6566381b17fb on 192.168.1.4:31010 ]
> Error: exception while executing query: Failure while executing query.
> (state=,code=0)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
