Can I change the order of the output fields from the Linux cut command?

Question

I am using cut command in command line and seems I can't get the output I like. Do you have any idea why I am getting this? Is it something that I do wrong?

This is the normal output and I would like to output in different order:

[root@upbvm500 root]# ls -al IDS_DIR/a | tr -s " "
-rw-r--r-- 1 root root 0 Jan 1 17:18 IDS_DIR/a
[root@upbvm500 root]#

[root@upbvm500 root]# ls -al IDS_DIR/a | tr -s " " | cut -d" " -f5,6,7,8,3,4,1
-rw-r--r-- root root 0 Jan 1 17:18

But as you can see, this is not working like expected. Any idea why they are switching places?

Guessing that you want to change the order of the fields, you might want to use any number of other tools, like sed. Using sed, you can write a substitution command to solve this. — Horus, Jan 01 '13 at 16:37
Yes, that is what I want to do, but I dont know why this is not working with cut. — Alexis Cimpu, Jan 01 '13 at 16:39

score 25 · Accepted Answer · answered Jan 01 '13 at 16:56

25

From man cut:

Selected input is written in the same order that it is read, and is written exactly once.

Use awk '{print $5,$6,$7,$8,$3,$4,$1}' instead of cut.

answered Jan 01 '13 at 16:56

Chris Seymour

83,387
30
160
202

Of course that won't work if some filename has one or more space(s) in it... – Mats Petersson Jan 01 '13 at 17:11
1

@MatsPetersson yes, good point but neither would `cut`! – Chris Seymour Jan 01 '13 at 17:12

score 6 · Answer 2 · answered Jan 01 '13 at 16:42

cut does not reorder its output. It simply collects a list of which columns to print, then prints them out as they arrive.

Use a different tool such as Awk to reorder output columns.

However, in this patricular case, try with stat or find instead of ls. It is generally not recommended to try to parse the output from ls. See http://mywiki.wooledge.org/ParsingLs

glenn jackman · Answer 3 · 2013-01-01T20:47:26.003

As others have mentioned, don't parse ls. If you want file information, use stat

stat -c "%s %y %U %G %A %n" filename

You may need to do some extra work to get the timestamp formatted as you want.

$ ls -l data
-rw-r--r-- 1 glennj glennj 13 2013-01-01 11:19 data
$ LC_TIME=POSIX ls -l data
-rw-r--r-- 1 glennj glennj 13 Jan  1 11:19 data

$ stat -c "%s %y %U %G %A %n" data 
13 2013-01-01 11:19:53.670015242 -0500 glennj glennj -rw-r--r-- data
$ stat -c "%s %Y %U %G %A %n" data | awk '{$2 = strftime("%b %e %H:%M", $2)} 1'
13 Jan  1 11:19 glennj glennj -rw-r--r-- data

Can I change the order of the output fields from the Linux cut command?

3 Answers3