[jira] [Commented] (HAWQ-1092) lc_collate and lc_ctype do not work after setting through hawq init

["Paul Guo (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HAWQ-1092?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15559751#comment-15559751 ] 

Paul Guo commented on HAWQ-1092:
--------------------------------

With a preliminary fix,
https://github.com/apache/incubator-hawq/pull/953

Now the results look like below:

# psql -d postgres -c "SELECT name, setting from pg_settings where name like 'lc%'"
    name     |  setting
-------------+------------
 lc_collate  | ja_JP.utf8
 lc_ctype    | ja_JP.utf8
 lc_messages | ja_JP.utf8
 lc_monetary | ja_JP.utf8
 lc_numeric  | ja_JP.utf8
 lc_time     | ja_JP.utf8
(6 rows)

# psql postgres -f sort2.sql
....... <ignored>
 id | name
----+------
  4 | DEF
  3 | abd
  5 | ghi
  1 | ウ
  2 | チ
(5 rows)

> lc_collate and lc_ctype do not work after setting through hawq init
> -------------------------------------------------------------------
>
>                 Key: HAWQ-1092
>                 URL: https://issues.apache.org/jira/browse/HAWQ-1092
>             Project: Apache HAWQ
>          Issue Type: Bug
>          Components: Catalog, Command Line Tools
>            Reporter: Paul Guo
>            Assignee: Lei Chang
>
> Users reports this issue to us, to reproduce:
> {noformat}
> # hawq init cluster -a --locale ja_JP.utf8 --lc-collate ja_JP.utf8 --lc-ctype ja_JP.utf8 --lc-messages ja_JP.utf8 --lc-monetary ja_JP.utf8 --lc-numeric ja_JP.utf8 --lc-time ja_JP.utf8
> #psql -d postgres -c "SELECT name, setting from pg_settings where name like 'lc%'"
>     name     |  setting
> -------------+------------
>  lc_collate  | C
>  lc_ctype    | C
>  lc_messages | ja_JP.utf8
>  lc_monetary | ja_JP.utf8
>  lc_numeric  | ja_JP.utf8
>  lc_time     | ja_JP.utf8
> (6 rows)
> {noformat}
> Besides, the sorting result is also not correct in some cases.
> {noformat}
> # cat sort2.sql
> DROP TABLE IF EXISTS t;
> CREATE TABLE t (id INT, name TEXT);
> INSERT INTO t VALUES (1, 'ウ');
> INSERT INTO t VALUES (2, 'チ');
> INSERT INTO t VALUES (3, 'abd');
> INSERT INTO t VALUES (4, 'DEF');
> INSERT INTO t VALUES (5, 'ghi');
> SELECT * FROM t ORDER BY name;
> #  psql postgres -f sort2.sql
> ........
>  id | name
> ----+------
>   1 | ウ
>   2 | チ
>   3 | abd
>   4 | DEF
>   5 | ghi
> (5 rows)
> {noformat}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)