#!/usr/bin/perl -lw

use strict;
use warnings;

while(<DATA>){
  chomp;
  my @data = split /\t/;
  my $ddrul = $data[7];
  $ddrul =~ s/\"//g;
  print "'$ddrul'";
}

__DATA__
EST-NY    234    5-Oct    Springfield    MA    Springfield College    
+Townhouse Conference Room    http://www.spfldcol.edu/home.nsf/welcome
+/visit/directionsc
EST-NY    923    18-Oct    Salisbury    MD    Wor Wic Community Colleg
+e        http://www.worwic.edu/campus/directions.pdf
EST-NY    886    19-Oct    Frederick    MD    Hood College        http
+://www.hood.edu/welcome_to_hood/index.cfm?pid=_maps.htm#a1
SW    328    19-Oct    Houston    TX    University of Houston - Clear 
+Lake        "http://prtl.uhcl.edu/portal/page?_pageid=328,217631,328_
+217645&_dad=portal&_schema=PORTALP"
[download]

'http://www.spfldcol.edu/home.nsf/welcome/visit/directionsc'
'http://www.worwic.edu/campus/directions.pdf'
'http://www.hood.edu/welcome_to_hood/index.cfm?pid=_maps.htm#a1'
'http://prtl.uhcl.edu/portal/page?_pageid=328,217631,328_217645&_dad=p
+ortal&_schema=PORTALP'
[download]

I don't know what else I can think my data is. The example data I supplied is what it is.

The crux of my code is



    open(INFILE,"$ew_sites_file") || die "Cant open $ew_sites_file for
+ reading $!\n";
    while($line = <INFILE>) {
        chomp $line;
        ($unused, $site, $date, $city, $state, $facility, $unused, $dd
+url) = split(/\t/,$line); #tab-delimited file
        $ddrul =~ s/"//g;
        $ddurl =~ s/\r//;  #lose that bad newline break
        
        #build the registration site lookup flatfile : state and javas
+cript URL    
        open(OUTFILE,">>$ew_regist_parse_file") || die "cant open $ew_
+regist_parse_file, $!\n"; 
        print OUTFILE "$state|$start_tag$site$inbetween$city$inbetween
+$state$inbetween$date$inbetween$ddurl$semicolon$city, $state : $date$
+endtag\n";
        close OUTFILE;        
        
        #build the locations flatfile            
        open(OUT2FILE,">>$ew_locate_parse_file") || die "cant open $ew
+_locate_parse_file, $!\n"; 
        print OUT2FILE "$state|$city|$facility|$date|$ddurl\n";
        close OUT2FILE;
    }
    close INFILE;
[download]

The warnings say that the $ddrul =~ s/"//g; is using an initalized value, which I don't understand because it is. The output is two files, an example of one (the simpler one) is:

TX|Austin|University of Texas at Austin|11-Oct| http://www.utexas.edu/
+cee/tcc/forms/tcclargemap.pdf
TX|Houston|University of Houston - Clear Lake|19-Oct|"http://prtl.uhcl
+.edu/portal/page?_pageid=328,217631,328_217645&_dad=portal&_schema=PO
+RTALP"
TX|El Paso|University of Texas - El Paso|25-Oct| http://www.utep.edu/s
+earch/campusmaplarge.html
[download]

Still with the extra quotes.

a use strict would have caught the mispelled variable "ddrul" as opposed to what it should be - "ddurl"

I'd probably have written it this way.

while (<INFILE>) {
    my (undef, $site, $date, $city, $state, $facility, undef, $ddurl) 
+=
        /\G"?([^\t]*?)"?(?:\t|[\r\n]+$)/g;
}
[download]

   $ddrul =~ s/"//g;

You don't need to escape double-quotes in a regex. If that doesn't work you'll need to show us more of your code.

Alternately you could use Text::CSV_XS or one of the other CSV modules on CPAN. They usually have a tab-delimited mode and should deal with quoted values fine.

-sam

$ddrul =~ s/"//g;
[download]

I'm parsing the data and putting the URL's in an href based on there being no quotes around the URL I'm reading. <a href=""...some url..""> doesn't work.