Using sed, awk or grep to split a string with numbers

Question

I have this file:

mmD_154Lbb_e_dxk_83233.orc
154L_bbe_Bddxk_3259.txt
14Lbe_3233.orc
m2_154Lbbe_dxk_67233.op
mZZ_1A4Lbbe_dxk_32823.op
mmD_154Lbbe_dxk_99333.orc
mmD_oS154be_dxk_12338.txt

I'm trying to use sed or awk to split the numbers and I don't have solution:

I need out put:

How can I get it to split on each delimiter?

Thanks

Where does the 2597 come from? – Wintermute Jun 09 '15 at 20:48 — Wintermute, Jun 09 '15 at 20:48
Have you tried looking up regular expressions? – rghome Jun 09 '15 at 21:23 — rghome, Jun 09 '15 at 21:23

shellter · Answer 1 · 2015-06-09T20:56:47.277

0

sed 's/\..*$//;s/^.*_//' file
       |       | 
       |       > remove all chars from beginning until last '_' char
       > remove last '.' char and all chars after that

output

IHTH

edited Jun 09 '15 at 20:56

answered Jun 09 '15 at 20:51

shellter

36,525
7
83
90

josifoski · Answer 2 · 2015-06-09T21:23:45.183

0

with GNU sed

sed 's/^.*_\([[:digit:]]\+\)\..*$/\1/' file

edited Jun 09 '15 at 21:23

answered Jun 09 '15 at 20:53

josifoski

1,696
1
14
19

Not bad but this return the full line when no digits are matched – Cyrbil Jun 09 '15 at 21:08

score 0 · Accepted Answer · answered Jun 09 '15 at 20:56

0

This awk can get you that:

awk -F '[_.]' '{print $(NF-1)}' file
83233
3259
3233
67233
32823
99333
12338

answered Jun 09 '15 at 20:56

anubhava

761,203
64
569
643

This is also returning the parsed line when no digits are found: (eg: `echo 'test.txt | awk -F '[_.]' '{print $(NF-1)}'` return `test.txt`) – Cyrbil Jun 09 '15 at 21:11
Do you know the requirement for such input? – anubhava Jun 09 '15 at 21:19

Using sed, awk or grep to split a string with numbers

3 Answers3