(Dermot) RE: RE: RE: RE: Re: Stripping page headers

[Dermot]

Ok, I've made two modifications to what I originally posted and now it works ok with your data. My original script would never have worked properly with your report file, it worked with the trivial example that I tested it on but how and ever. Here is a working version:

#!/usr/bin/perl -w
use strict;

my ($REPFILE, $report);

undef $/;

open REPFILE, "report.rpt" or die "Cant open $REPFILE: $!\n";

$report = <REPFILE>;

$report =~ s/^User Report.*//mg;
$report =~ s/^All Users.*//mg;
$report =~ s/^User Name.*//mg;
$report =~ s/^-> Token.*//mg;

print $report;

close REPFILE;

Addition of m modifier to the substitution.

Because we are dealing with the whole report file in one scalar it is effectively one string and the rule for ^ and $ is that they match at the start and end of a string, not a line. To get ^ and $ matching at the start and end of a line instead of a string we have to add the m modifier. Now it sees the string in $report as a series of lines delimited by \n characters.

Addition of .* to the regex to deal with the rest of the line.

By adding .* to the regex we cause it to match (i) start of line, (ii) piece of text that we're using as a tag on the line, (iii) rest of the line up to the next \n. A dot character in a regex matches any character except a newline (\n). If you want it to match a newline you can specify this using the s regex modifier. Just to top off the confusion you can actually use both the s and m modifiers on the same regex. Most people assume that they mean single-line vs multi-line but actually they mean match newlines with dot and match ^ and $ in lines not in the whole string.

[Anonymous Monk]

You rock! :) It works flawlessly. BIG Thanks!