#! /opt/bin/perl
#	________________________________________________________________
#	  /\ /\		The Web Developer's Virtual Library
#	-{-<*>-}-	World-Wide Web	   <URL:http://WDVL.com/>
#	__\/_\/_________________________________________________________
#	Author	:	Lucy Richmond
#	Purpose	:	Read the access log and create a listing of the 404s
#			with a referring page within our site
#	Usage	:	find_error.pl
#	Comment	:	The input file (access_log) is a standard Unix log file.
#			The output file (cur_404.log) is of the form:
#				bad URL	(tab)	calling URL
#	Disclaimer:	This software is provided freely on the understanding
#			that the Author will not be held responsible for any
#		problems arising from it's use, and that there is no support.
#	________________________________________________________________


	open	(IN, "<access_log")||die$!;
	open	(OUT, ">cur_404.log")||die$!;

	while   (<IN>)    {
		($who, $j, $j, $datime, $TZ, $method, $URL, $protoc, $status,
		$bytes, $refer, $browser)
					= split ;

		next if !($status =~ /404/);
		if (($refer =~ /wdvl.internet.com/i) or
			($refer =~ /stars.com/i) or
			($refer =~ /wdvl.com/i)) {

			print OUT "$URL	$refer\n" if ($URL =~ /html/) ;
			}
		}