Question

I have 9000 PHP files, most of which contain the text foo. I know that the file that I need has the text foo in a HEREDOC, but it is likely tens of lines away from either the beginning or end of the HEREDOC so grep's -A or -B options won’t help.

Is there any way to perform this search? Is there a better tool to use than grep?

This is on a CentOS 5.2 box, if relevant. Yes, I know that the OS is outdated!

Answer 1

This is pretty trivial with both awk and sed but only awk has the ability to print both the filename and the line number

 awk '/<<<HDOC/{f=1;next}f{if($0 ~ /\<foo\>/)print NR,FILENAME,": "$0}/HDOC;/{f=0}' *.php

Input

$ cat heredoc.php
blah
blah
foo
bar
foo
bash

$str = <<<HDOC
this is a
heredoc which contains
the word foo
more
stuff
here
HDOC;
other lines
with foo
here

Output

$ awk '/<<<HDOC/{f=1;next}f{if($0 ~ /\<foo\>/)print NR,FILENAME,": "$0}/HDOC;/{f=0}' *.php
11 heredoc.php : the word foo