LinuxQuestions.org - How do I cut fields with repeated delimiters?

- Linux - Software (https://www.linuxquestions.org/questions/linux-software-2/)

- - How do I cut fields with repeated delimiters? (https://www.linuxquestions.org/questions/linux-software-2/how-do-i-cut-fields-with-repeated-delimiters-725014/)

How do I cut fields with repeated delimiters?

Given a file listing like this:

Code:

$ ls -laR

.:

total 24

drwxr-xr-x  3 auser admin  4096 2009-05-10 13:19 .

drwxr-xr-x 20 auser admin 12288 2009-05-10 13:19 ..

-rw-r--r--  1 auser admin  998 2008-04-07 13:37 file1.txt

drwxr-xr-x  2 auser admin  4096 2009-05-10 13:21 directory1



./directory1:

total 24

drwxr-xr-x 2 auser admin 4096 2009-05-10 13:21 .

drwxr-xr-x 3 auser admin 4096 2009-05-10 13:19 ..

-rwxr-xr-- 1 auser admin 2176 2008-04-06 13:30 2file.txt

-rw-r--r-- 1 auser admin  249 2008-04-06 23:23 experiment.txt

-rw-r--r-- 1 auser admin  334 2008-04-07 11:29 somefile.txt

-rw-r--r-- 1 auser admin 1604 2008-04-02 13:37 text3.out

I want a list that looks like this:

Code:

2008-04-02 13:37 text3.out

2008-04-06 13:30 2file.txt

2008-04-06 23:23 experiment.txt

2008-04-07 11:29 somefile.txt

2008-04-07 13:37 file1.txt

What I'm trying to do is list all files in all subdirectories recursively and sort by date. (I might also pipe the output through tail to list e.g., only the 100 newest files.)

Here is what I tried:

Code:

$ ls -ogARtr | grep '^-' | cut -d ' ' -f 4-

998 2008-04-07 13:37 file1.txt

2008-04-02 13:37 text3.out

2008-04-06 13:30 2file.txt

249 2008-04-06 23:23 experiment.txt

334 2008-04-07 11:29 somefile.txt

Apparently the extra space before the smaller files makes cut not work as expected. That's my main problem. The other problem is I will apparently have to send it through sort to get an overall (not the ls per-subdirectory) sorting.

How can I achieve the desired result?

I dunno, but here's what I'd do:

Code:

bash-3.1$ ls -l | awk '{ printf("%s\t%s\t%s\n",$6,$7,$8)}' | sort -k1

                

2008-11-24        22:29        var

2009-04-12        13:54        note

2009-04-13        17:08        opt

2009-05-03        16:40        chrom.odt

2009-05-03        18:35        bin

2009-05-03        18:35        lib

2009-05-05        22:14        subjects

2009-05-10        16:14        tmp

2009-05-10        18:16        bench

The '-k1' is optional as the default sorts by first column anyway.

Of course this would also work:

Code:

bash-3.1$ ls -l | sort -k6

total 120

drwxr-xr-x 15 kafox users  4096 2008-11-24 22:29 var

-rw-r--r--  1 kafox users    82 2009-04-12 13:54 note

drwxr-xr-x 12 kafox users  4096 2009-04-13 17:08 opt

-rw-r--r--  1 kafox users  9894 2009-05-03 16:40 chrom.odt

drwxr-xr-x  2 kafox users  4096 2009-05-03 18:35 bin

drwxr-xr-x 17 kafox users  4096 2009-05-03 18:35 lib

-rw-r--r--  1 kafox users  466 2009-05-05 22:14 subjects

drwxr-xr-x 48 kafox users 40960 2009-05-10 16:14 tmp

-rw-r--r--  1 kafox users  474 2009-05-10 18:16 bench

you should remove repeats:

Code:

$ ls -ogARtr | grep '^-' | sed "s/ */ /g" | cut -d ' ' -f 4-

that will do the trick.

Quote:

Originally Posted by Mountain (Post 3536249)

Given a file listing like this:

Code:

$ ls -laR

.:

total 24

drwxr-xr-x  3 auser admin  4096 2009-05-10 13:19 .

drwxr-xr-x 20 auser admin 12288 2009-05-10 13:19 ..

-rw-r--r--  1 auser admin  998 2008-04-07 13:37 file1.txt

drwxr-xr-x  2 auser admin  4096 2009-05-10 13:21 directory1



./directory1:

total 24

drwxr-xr-x 2 auser admin 4096 2009-05-10 13:21 .

drwxr-xr-x 3 auser admin 4096 2009-05-10 13:19 ..

-rwxr-xr-- 1 auser admin 2176 2008-04-06 13:30 2file.txt

-rw-r--r-- 1 auser admin  249 2008-04-06 23:23 experiment.txt

-rw-r--r-- 1 auser admin  334 2008-04-07 11:29 somefile.txt

-rw-r--r-- 1 auser admin 1604 2008-04-02 13:37 text3.out

I want a list that looks like this:

Code:

2008-04-02 13:37 text3.out

2008-04-06 13:30 2file.txt

2008-04-06 23:23 experiment.txt

2008-04-07 11:29 somefile.txt

2008-04-07 13:37 file1.txt

What I'm trying to do is list all files in all subdirectories recursively and sort by date. (I might also pipe the output through tail to list e.g., only the 100 newest files.)

Here is what I tried:

Code:

$ ls -ogARtr | grep '^-' | cut -d ' ' -f 4-

998 2008-04-07 13:37 file1.txt

2008-04-02 13:37 text3.out

2008-04-06 13:30 2file.txt

249 2008-04-06 23:23 experiment.txt

334 2008-04-07 11:29 somefile.txt

Quote:

Originally Posted by H_TeXMeX_H (Post 3536297)

Code:

bash-3.1$ ls -l | awk '{ printf("%s\t%s\t%s\n",$6,$7,$8)}' | sort -k1

I would do awk fields $(NF-2),$(NF-1),$NF rather than fixed numbers 6,7,8.

Quote:

Originally Posted by linosaurusroot (Post 4832458)

I would do awk fields $(NF-2),$(NF-1),$NF rather than fixed numbers 6,7,8.

Why ?

P.S. This thread is old, I'm pretty sure the OP will not answer.