Try something like this instead, to read and parse lines from a file:

use FileHandle;
my $fh = new FileHandle;
open($fh, "<", $file) or die "Unable to open file '$file' ($!)\n";
while (my $line = <$fh>) {
    chomp $line;
    if ($line =~ /(<IMG.*?BR>)...(Figure\d*)/i) {
        # You got a match
    }
}
[download]

Hello

i think you forgot to open the file in order to search inside it

open the file

open(FILE,"$html_dir/$FigFile/$DirFile") or die $!;
while (<FILE>){
...
}
close(FILE)
[download]

HTH

bye

You've got lots of problems, some critical, some minor.

A quick rundown:

• You've got use diagnistics; which helps you figure out how you shot yourself in the foot, but not use strict; which help you not shoot yourself in the foot in the first place. Highly recommended for new (and not so new) users.
• Use lexical rather than global file and directory handles. The scoping issues are much easier to track.
• Use "or" rather than || when you aren't actually doing boolean logic, the precedence is much lower and it avoids needing to use lots of parenthesis.
• The system/library error variable is $! not $1.
• Don't use capturing parenthesis in regular expressions if you aren't capturing anything.
• If you need to reuse open directory/file handles, you need close them first.
• Factor out long complex expressions when you can, especially if you find yourself using them over and over.
• Full Stops are significant inside regular expressions. Escape them if you want the literal character.
• You need to open a file before you can read from it. You can't just read from the file name.
• You can't match a multiline expression with a regex if you are reading the file in line by line.
• Don't be afraid to use whitespace in your code. It can really make it easier to follow the logic.

• for all files in a given directory,
  
  • make a backup copy
    
  • change permissions
    
  • extract patterns matching /IMG SRC=.../ and /Figure.../
    
• print out each image URL found

this pretty much outlines the structure of the program....so writing it is a fairly simple process of following the above structure.

personally, i would make the backup copy of the files and change the permissions as a separate task (just on the general principle that a tool like this should only do one thing so that it can be re-used easily - and also so that you can run it while testing it WITHOUT making any changes to the files/directories on disk), but you can do it within the perl script if you want.

once you have the IMG SRC urls in a hash, you can do whatever you want with them, including printing them out as an <A HREF="..."> HTML link.

#! /usr/bin/perl -w

use strict;
use File::Copy;

# pass directory to scan as arg1 (default to current dir)
my $dir = shift || "./" ;

# get list of non-hidden files in directory
opendir(DIR, $dir) || die "can't opendir $dir $!";
my @files = grep { /^[^.]/ && -f "$dir/$_" } readdir(DIR);
closedir DIR;

my %images = ();

# process each file
foreach my $file (@files) {
    next if ($file =~ /\.pl/) ; # skip perl program files
    copy($file, "$file.bak");
    chmod 0600, $file;

    my $img = '';
    my $fig = '';

    open(FH,"<$file") || die "couldn't open $file for read: $!\n";
    while (<FH>) {
        chomp ;
        s/^\s*|\s*$//g;  # strip leading and trailing spaces
        if (/<IMG SRC/) {
            $img = $_ ;
        } elsif (/Figure\s+\d+/) {
            $fig = $_ ;
        } ;
    } ;
    close(FH);

    # if we found an IMG SRC line *AND* a Figure line, then
    # add it to the images hash.
    if ($img && $fig) { $images{$fig} = $img } ;
};

foreach (sort keys %images) {
    print "$_ : $images{$_}\n" ;
};
[download]

This is for work. I'm a UNIX shell scripter(Sys Admin) who was more or less thrown into this position.

I have done a smattering of other languages, and lots of classes, but nothing prepared me for actual production coding. My company is in a jam and I am trying to help out until they can hire a more proficient Perl Programmer.

I have found that I enjoy Perl, and plan to continue to work with it full time. I am enrolled in several classes; unfortunately, they don't start until April.

Edited by planetscape - removed br tags, replaced with <p></p> tags

Okay, now I understand. The problem, I think, for your situation is this: if you do a great job with this script with mostly borrowed code you will likely be asked to do a bunch more (and starting in April means you won't be done with those classes until what, June?).

You need help now, but it isn't really code snippets that you need. I think you need a few good books that will help you basically do what Sys Admins need to do often. Here is a quick reading list:

• Effective Perl Programming
• Perl for System Administration
• The Perl Cookbook
There are others, but if I were you I would read Effective Perl Programming cover-to-cover (it is a thin book), scan through the Perl Cookbook for script problems you are having right now, and then read the System Administration book. I used to have this Perl Black Book on my desk for a while until I read the above books and realized that the Black Book (at least my edition) was full of bad programming examples.

BTW, Perl programming can be addictive. Before I started learning Perl I was a professional VB programmer. Nowadays I struggle to remember how to do simple things in VB because I use it so rarely. Welcome to the world of Perl.

Celebrate Intellectual Diversity

This probably doesn't help now, but you could have gotten a part-time proficient perl programmer from http://jobs.perl.org/ to get you through the crisis.

duff

Don't forget about PM's excellent Tutorials section. I am thinking of Getting Started with Perl and in particular, woolfy's Where and how to start learning Perl. (Don't forget about Super Search either. ;-) )

Good luck!