The bits you need to get started are File::Find to scan for the files and something like the following code to build a regex used in the search:

use strict;
use warnings;

my @constants = qw(foo bar baz);
my $match = '\b' . join ('\b|\b', @constants) . '\b';

while (<DATA>) {
    print if /$match/;
}

__DATA__
A foo line
no hit
bar line
barren line
No food here
foo_bar_baz doesn't hit too
[download]

Prints:

A foo line
bar line
[download]

$ find . -name '*.[ch]' | xargs egrep -w 'foo|bar|baz|boz'
[download]

Dave.

Ok, removed the h, and now it works...hmmm...well I can write a 2 line script, one for the *.c's and one for the *.h's !!!

So it worked, with a tweak or 2...i like it because it's simple...but I still want to 'Perlize' it...

Just for the experience!

Thanks for your advice and help!

ki6jux

The -w option was unknown, so I removed that.

Then it ran without finding the file, test_code.c in the current directory. So I looked up the commands, to better understand what we were doing...it should work, but it just returns empty...no printout at all.

ki6jux

my rep now stands @ -3...

egrep --recursive --include=*.[ch] 'foo|bar|baz|boz' *
[download]

But you need an advanced grep, such as gnu grep.

A word spoken in Mind will reach its own level, in the objective world, by its own weight

All the find, grep, egrep and File::Find answers are relevant but I'm really starting to like ack (or App::Ack) for this type of thing.

ack --type=cc 'foo|bar|baz'
[download]

use strict;
use File::Find;

# Perl script to search files for particular strings.
# Recurses into subdirectories.
# If you don't specify a directory to search on the command line,
# it will default to the current directory.

# define strings to look for here:
my @Constants = (
  "foo",
  "bar",
  "baz",
);
  
my $path = $ARGV[0] || ".";
my $count=0;

find( sub {
  my $n = $File::Find::name;
  # search for files ending in .h or .c
  if ($n =~ /\.h$/ || $n =~ /\.c$/ ) {
    ++$count;
    # might want to use die instead of warn
    open(IN, "$n") or warn "can't open file $n : $! ";  
    while (<IN>) {
      my $line = $_;
      for my $search_string (@Constants) {
        # the "i" makes the match case insensitive
        if ($line =~ /$search_string/i) {      
          # print filename and line number and matching string
          print "$n : $. : $search_string\n";           
        }
      }
    }
    close (IN);
  }
}, $path);


print "Found matches in $count files.\n";
[download]

update: Fixed typo in original post - the path should default to the current working directory. Also, looking at Grandfather's reply, if you want a pattern like "foo" to NOT match something like "food" or "foobar", change the pattern match line to this:

if ($line =~ /\b$search_string\b/i) {
[download]

This adds "end of word boundary" around the pattern.

Yes!

I understand this code and I need case sensitivity, so I can remove the i switch.

This looks very cool, and I will type it in right away...!!! I was here late last night looking at the Find::File CPAN entries...but they just didn't make any sense to me...maybe it was too late... :D

Thanks scorpio17!

I will spend some time playing with this script!

Thanks to all the Monks for your feedback, it was very instructive. I will post the final version out on the Code Catacombs for others to share!

Cheers! 8)

ki6jux