[
https://issues.apache.org/jira/browse/HADOOP-2865?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12588832#action_12588832
]
Konstantin Shvachko commented on HADOOP-2865:
---------------------------------------------
+0.7
I think this is good as a first step in improving the formatting of ls.
There are 3 cases when formatting brakes
# when file size is larger than 7 digits
# when user names substantially differ in length
# when group names differ
{code}
2 168 2007-10-09 20:21 -rw-rw-rw- shv supergroup
/README.txt
0 2008-04-09 18:47 drwxr-xr-x administrator
supergroup /conf
3 12345678 2008-04-09 18:46 -rw-r--r-- shv
supergroup /conf/.svn/README.txt
3 1234567 2008-04-09 18:47 -rw-r--r-- shv supergroup
/conf/.svn/dir-prop-base
3 90 2008-04-09 18:46 -rw-r--r-- shv user
/conf/.svn/dir-wcprops
3 9012345 2008-04-09 18:47 -rw-r--r-- shv supergroup
/conf/.svn/dir-props
{code}
The first one is going to be very common, since we expect large files and
design for it. And we should fix it.
The best way would be to convert bytes into MB, GB, TB, or even PB if needed,
but at least using "%-XXd" as in DU would be better.
This should also produce right aligned column.
Dynamically calculating column widths is probably the next logical improvement
step, which belongs to a different issue.
We should mark this issue as an incomatible change.
> FsShell.ls() should print file attributes first then the path name.
> -------------------------------------------------------------------
>
> Key: HADOOP-2865
> URL: https://issues.apache.org/jira/browse/HADOOP-2865
> Project: Hadoop Core
> Issue Type: Improvement
> Affects Versions: 0.15.0
> Reporter: Konstantin Shvachko
> Assignee: Edward J. Yoon
> Fix For: 0.18.0
>
> Attachments: 2865_v01.patch
>
>
> When we had a handful of attributes this looked OK. But after the permissions
> the ls output is just unreadable.
> This how the output looks now:
> {code}
> /CHANGES.txt <r 2> 174779 2008-01-29 10:00 rw-r--r-- shv
> supergroup
> /LICENSE.txt <r 2> 11358 2008-01-28 17:27 rw-r--r-- shv
> supergroup
> /NOTICE.txt <r 2> 101 2008-01-29 12:06 rw-r--r-- shv
> supergroup
> /Work <dir> 2008-01-23 17:43 rwxr-xr-x shv
> supergroup
> /Work/hadoop-data <dir> 2008-01-23 17:43 rwxr-xr-x
> shv supergroup
> {code}
> This is how it should look:
> {code}
> 2 174779 2008-01-29 10:00 -rw-r--r-- shv supergroup
> /CHANGES.txt
> 2 11358 2008-01-28 17:27 -rw-r--r-- shv supergroup
> /LICENSE.txt
> 2 101 2008-01-29 12:06 -rw-r--r-- shv supergroup
> /NOTICE.txt
> 2008-01-23 17:43 drwxr-xr-x shv supergroup
> /Work
> 2008-01-23 17:43 drwxr-xr-x shv supergroup
> /Work/hadoop-data
> {code}
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
