[jira] Updated: (MAPREDUCE-941) vaidya script calls awk instead of nawk

["Chad Metcalf (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MAPREDUCE-941?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chad Metcalf updated MAPREDUCE-941:
-----------------------------------

    Attachment: MapReduce941.patch

This is perhaps a better alternative. The script current already makes an assumption that the version information is in the second half of the first line. This changes the assumption to second half of a line starting with Hadoop. So it isn't any less maintainable. It improves readability. And doesn't require a user to manipulate paths or have nawk installed. 

The patch uses the most basic (read portable) functionality from both grep and awk.

  

> vaidya script calls awk instead of nawk
> ---------------------------------------
>
>                 Key: MAPREDUCE-941
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-941
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>         Environment: Solaris
>            Reporter: Allen Wittenauer
>            Priority: Trivial
>         Attachments: MapReduce941.patch
>
>
> The vaidya script uses this construction:
> awk 'BEGIN { RS = "" ; FS = "\n" } ; { print $1 }'
> Under "old awk", this fails.  Under "new awk", this works.  Depending upon the OS, awk may point to either old or new awk.  In the Solaris case, awk defaults to oawk.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.