English Deutsch Français Italiano Español Português 繁體中文 Bahasa Indonesia Tiếng Việt ภาษาไทย
All categories

I use grep and egrep countless times daily but once in awhile I come up with a problem that grep can do part of and egrep can do the other part but neither can do both.

Maybe I just need the syntax?

I have a file with 3 digit codes something like this:
001003009000000000
002004008997999000
... etc etc

I want to find lines with the same code repeated so:

grep '00\([0-9]\).*00\1'

will find it but not necessarily in the correct format. E.g. it will incorrectly identify '001030010' even though the codes are 1, 30 and 10.

I would like to use: grep '00\([0-9]\)\(...\)*00\1'
but the wildcard does not apply to groups in grep. And egrep cannot backreference a match in the form \n.

How can it be done?

2007-02-22 07:43:26 · 1 answers · asked by scruffy 5 in Computers & Internet Programming & Design

I've read the manuals. They seem to indicate that it can't be done. I'd like to know if there is another way.

2007-02-22 08:27:58 · update #1

1 answers

Here is the manual page to grep maybe this can help you http://www.linuxcommand.org/man_pages/grep1.html

and this one is to egrep
http://www.nada.kth.se/cgi-bin/man?p=egrep&s=1&ss=&M=

2007-02-22 08:21:48 · answer #1 · answered by theguy 2 · 0 0

fedest.com, questions and answers