[jira] [Updated] (PHOENIX-1523) Make it easy to provide a tab literal as separator for CSV imports

["Sergey Soldatov (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/PHOENIX-1523?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Sergey Soldatov updated PHOENIX-1523:
-------------------------------------
    Attachment: PHOENIX-1523.patch

Added support for escape sequences as the delimiter.

> Make it easy to provide a tab literal as separator for CSV imports
> ------------------------------------------------------------------
>
>                 Key: PHOENIX-1523
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-1523
>             Project: Phoenix
>          Issue Type: New Feature
>            Reporter: Gabriel Reid
>            Assignee: Sergey Soldatov
>         Attachments: PHOENIX-1523.patch
>
>
> When importing CSV data that is tab-separated, it's quite inconvenient to supply a tab literal on the command line to override the default separator (you need to press Ctrl-V, followed by Tab, which isn't well known).
> It would be better if it was possible to supply something that looks like a tab literal, i.e. '\t', or something else like '<tab>', to reduce the complexity in supplying a tab separator.
> Regardless of the approach, the way to supply a tab character should also be added to the documentation (or to the command line docs provided by the tool).
> This should be done for both the map reduce bulk loader, and psql.



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