[jira] [Commented] (PHOENIX-1733) sqlline.py doesn't allow connecting with a tenant-specific connection

["Abhishek Sreenivasa (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/PHOENIX-1733?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14498581#comment-14498581 ] 

Abhishek Sreenivasa commented on PHOENIX-1733:
----------------------------------------------

Wrapping the argument passed to sqlline.py with pipes.quote() as [~gabriel.reid] mentioned made it work. Will submit a patch via pull request.

> sqlline.py doesn't allow connecting with a tenant-specific connection
> ---------------------------------------------------------------------
>
>                 Key: PHOENIX-1733
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-1733
>             Project: Phoenix
>          Issue Type: Bug
>            Reporter: Gabriel Reid
>            Assignee: Abhishek Sreenivasa
>              Labels: Newbie
>
> Connecting with a client-specific connection via sqlline.py should work with a command like the following:
> {code}
> ./bin/sqlline.py 'localhost:2181;TenantId=t1'
> {code}
> However, the command that is built up within sqlline.py doesn't correctly quote the JDBC url, so the JDBC url parameter ends up as:
> {code}
> jdbc:phoenix:localhost:2181;TenantId=t1
> {code}
> without quoting. This causes the semicolon to be interpreted as a command separator by the shell, effectively ignoring it (and any other parameters).
> Proper parameter quoting (as done with the pipes module in psql.py) should be used for sqlline.py to get around this and similar issues.



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