Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!rpi!fitz
From: fitz@mml0.meche.rpi.edu (Brian Fitzgerald)
Newsgroups: comp.unix.questions
Subject: Re: Word-oriented GREP
Message-ID: <sgjgd2f@rpi.edu>
Date: 15 Apr 91 04:25:20 GMT
References: <1991Apr15.014626.28903@berlioz.nsc.com>
Distribution: na
Organization: Rensselaer Polytechnic Institute, Troy NY
Lines: 29
Nntp-Posting-Host: mml0.meche.rpi.edu

Taed Nelson writes:
>
>When I use the command "grep V\[0-9\]\[0-9\]\[0-9\] fred.c" it returns
>	#define VERSION "V002"
>  or somesuch.  What I would really like is just the string of characters
>  which matched:
>	V002

Try:

sed -n -e 's/.*\(V[0-9][0-9][0-9]\).*/\1/p'

(use \[ and \] in csh)

This just gets the first occurrence on a line, though.  Wizards know
how to print more than one occurrence with shell commands, even if they
are not separated by white space. (i.e. V001,V002,...)

Since I'm not a wizard, I tried lex:

%%
V[0-9][0-9][0-9]	printf("%s\n",yytext);
.	|
\n	;

Save as foo.l and type "make LDLIBS=-ll foo" (or "lex -t foo.l >
foo.c ; cc -o foo foo.c -ll")

Brian