Well, this is a VERY localized script - but I thought it was cool, seems to be working quite well, and I am a palm user that does not surf with it. This script will grab the daily headlines from www2.alberta.com/news and (using cron) update my palm pilot with the file before I leave. I am ready to be critiqued :-)

#!/usr/bin/perl -w
###########################################
# This will hopefully become a full       #
# fledged perl script that downloads      #
# news from alberta.com and converts      #
# to txt or some other format readable by #
# palms the idea is to run this nightly   #
# using cron, and schedule syncs with the #
# Palm Pilot                              #
# - Slycer 10/19/2000                    #
###########################################

use strict;
use LWP::Simple;

my $content;                       #the content of the alberta.com ind
+ex
my $workingdir = "/home/slycer";          #the dir to write all the fi
+les to
my $url = "www2.alberta.com/news"; #the url of alberta.com news
my $outlist= "abcom.htm";          #the outfile for the above
my $outhtm= "news.htm";            #the outfile for all the urls below
my (@urls,                         #the list of good urls
    @end_ref,                      #the ending line of the stories
    @start_ref,);                  #the starting line of the stories
my %storieshash;                   #the hash that contains the text
my $nfile="news.txt";              #the news file that we read/write t
+o (plain
                                   #text)
my $lines="83";                    #the amount of CRAP between stories
my $conprog="/usr/bin/txttopdb";  #the script to convert to pdb format
my $final="stripped.txt";          #the final file to convert
my $pdbfile="news.pdb";            #the name of the pdb file
my $title="DailyNews";             #the Title (as displayed in the Pal
+m)


chdir $workingdir;


#####First go get the index file from alberta.com######

open (OUT,">$outlist")||warn "Could not create alberta.com news";
unless (defined ($content = get "http:\/\/$url")) {
    die "Unable to access abcom website : $!";
}
print OUT $content;
close OUT;


#####Make the URLS usable#####

open(CLGR,"<$outlist");
while (<CLGR>){
    if (/.*fs\.*/){
    s/.+href=\"//i;
    s/\".+//;
    chomp ($_);
    $_= "\"http:\/\/$url\/$_\"";
    print "$_\n";
    push (@urls => $_);
    }
}
close (CLGR);


###Done creating our urls - now go get the stuff - uses the subroutine
+###
&fetch;


###Convert it all to text - don't bother reinventing the wheel###
my $text= `lynx -dump -nolist $outhtm`;
open (NEWS => ">$nfile");
print NEWS $text;
close (NEWS);


###Now that we have a plain text file - strip the CRAP out
&strip;


### call the txttopdb script (from freshmeat) to convert to Palm forma
+t
#for some reason having troubles finding the files!!
#so we will joing the dir and file name...

$final=join '/',$workingdir,$final;
$pdbfile=join '/',$workingdir,$pdbfile;
`$conprog -t $title $final $pdbfile\n`;

###lastly - do some cleanup
unlink ($outlist,$outhtm,$nfile,$final)|| warn "Could not erase the fi
+les $!";

###Subs from here on down###

#######
#FETCH#
#######
sub fetch{
    open (HTM => ">$outhtm")||warn "wtf.. $!";
    chomp (@urls);
    foreach (@urls){
    print "getting $_\n";
        unless (defined ($content = get $_)) {
            warn "could not get $_\n";
        }
    print HTM $_;
        print HTM $content;
    }
    close (HTM);
}

#######
#STRIP#
#######
sub strip{

###This part finds out the ENDS of each story###

    $end_ref[0]=0;               #set first story to begin @ line 0
    open (NEWS,"<$nfile")            #the news file in plain text..
    ||warn "Could not open news file $!";
    my $i=1;
    while (<NEWS>){
    $storieshash{$i}=$_;
    if ($_ =~ /.*opyright*/){    #all of them end in "copyright"
        
        push (@end_ref=>$i);     #a reference to the hash that marks t
+he end
        #of the story
    }
    ++$i;
    }
    
###This next part needs to figure out how many lines between the END o
+f
###story (marked by "copyright") and the beggining of the next ($lines
###l8r)  Yes, this is aimed strictly at alberta.com
    
    foreach (@end_ref){push (@start_ref => ($_ + $lines))}
    shift @end_ref;                  #get rid of the zero
    pop @start_ref;                  #get rid of the last story
    
    open (FINAL,">$final");
    foreach (@end_ref){
    print FINAL "\n                       Story starts here\n";
    for ($i = shift(@start_ref);$i <= $_;$i++){
        print FINAL "$storieshash{$i}";
    }
    }
    close FINAL;
}
[download]