I learned much from Juerd's vandale.pl, that checks words in the dutch dictionairy.

Too bad they (vandale) changed their HTML. Although it's less broken now, this means my original script no longer works. I have an updated version available, and will update the PM node soon. Update (200306081720+0200) - done.

Juerd # { site => 'juerd.nl', plp_site => 'plp.juerd.nl', do_not_use => 'spamtrap' }

use LWP::UserAgent;
[download]

I should have used LWP::Simple and so should you :)

This is copy/paste-programming, I think. I used LWP::UA because some earlier version did some extra work, but I guess you just copied my code. My bad code.

sub Jammer_hoor {
[download]

Subs should be declared before they're used, in my opinion. Especially when the name of the sub is useless. Why the capital J, by the way?

if ( $nummer !~ m/\b\d{3}-\d{7}\b/ ) {
[download]

So !!!!!!!!!!!!!!hallo wereld!!!!!!@#$@#$%^#$%^#$%^#"$%#$%#"$%"#$%#$%"#$%123-4567890!!!!!!!!!! is a valid Dutch phone number? Besides that, it's possible to have a three-digit area number and a 6-digit subscriber number. With 06 number's it's 2-8, even.

if ( $info !~ m/Naam:/g ) {
[download]

I think explicit m with // as delimiters is misleading. As misleading as any of q'' qq"" qx`` m//. Explicit code isn't always clearer.

$info =~ s/\s+/ /gi;
[download]

Why /i? Afraid there will be uppercased whitespace? For all your regexes: only use modifiers that are useful for that regex.

#filter out all the adds and links
$info =~ s/$nummer//gi;
$info =~ s/<.*?>//gi;
$info =~ s/\.\.\..*?\.\.\.//gi;
$info =~ s/\{.*?\}//gi;
$info =~ s/\(.*?\)//gi;
$info =~ s/table.*?://gi;
$info =~ s/1\ item.*?sp\;//gi;
$info =~ s/2\ item.*?sp\;//gi;
$info =~ s/\s+/ /gi;

#format for easy cut 'n paste
$info =~ s/Telnr: /\n/g;
$info =~ s/Naam: /\n/g;
$info =~ s/Adres: /\n/g;
$info =~ s/Plaats: /\n/g;
$info =~ s/Postcode: /\n/g;
$info =~ s/Fax: /\nfax: /g;
[download]

You can topicalize using for:

for ($info) {
    #filter out all the adds and links
    s/$nummer//gi;
    s/<.*?>//g;
    ...
}
[download]

print "$info";
[download]

Get rid of those quotes. You're stringifying a string, copying the value in memory more often than necessary.

Juerd # { site => 'juerd.nl', plp_site => 'plp.juerd.nl', do_not_use => 'spamtrap' }

>I should have used LWP::Simple and so should you :)

>This is copy/paste-programming, I think. I used LWP::UA
>because some earlier version did some extra work, but I guess you
>just copied my code. My bad code.

Well, the script I wrote before this one used lynx to dump the file. I copied your use of modules to make it more platform-independant.

sub runit {
# first filter out the relevant telinfo:
open TEL, "lynx -dump zoekopnummer.ath.cx/index.php?nummer=$nummer|";
open (LOG, ">$log");
while(<TEL>){
print LOG $_;
}
[download]

was part of my first version. I just changed the fetching of the page and used the, "in you own words", bad method you used.

Just calling it copy/paste is not entirely fair I think, but hey, I gotta admit, you're a far better coder than me.

>So !!!!!!!!!!!!!!hallo wereld!!!!!!@#$@#$%^#$%^#$%^#"$%#$%#"$%"#$%#$%"#$%123-4567890!!!!!!!!!! is a valid Dutch phone number?
>Besides that, it's possible to have a three-digit area number and a 6-digit subscriber number. With 06 number's it's 2-8, even.

Nope, it isn't. And according to testing on my windows machine and my linux machine this tiny utility doesn't think it is either.

As for 06 numbers, they are not listed on this site , so it's pretty useless to check for them, and the same goes for servicenumbers, I guess that's what you're aiming at with the three-digit area number and a 6-digit subscriber number.

You're totally right about the sloppy matches and about the the unnecessary "ignore case" substitution. So here's a rewitten version that fixes (most of) those mistakes and remarks.

Oh, and using LWP::Simple ;)

#!/usr/bin/perl -w
#gettel.pl - poor man's foongrep for dutch telephone numbers 

use strict;
use LWP::Simple;

my $nummer;

sub jammer_hoor {
    print
      "Usage: Specify a dutch telephone number like\ngettel.pl 020-123
+4567\n";
    exit;
}

if ( $ARGV[0] ) {
    $nummer = $ARGV[0];
}

else {
    jammer_hoor();
}

if ( $nummer !~ m/\b\d{3}-\d{7}\b/ ) {
    jammer_hoor();
}

#get page
my $url  = "http://zoekopnummer.ath.cx/index.php?nummer=$nummer";
my $info = get($url);

#no more newlines!
$info =~ s/\n//g;

#substitute number
$nummer =~ s/-//g;

if ( $info !~ m/Naam:/g ) {
    print "\nNo info found on that number\n";
    exit;
}

for ($info) {

    #filter out all the adds and links
    s/$nummer//g;
    s/<.*?>//g;
    s/\.\.\..*?\.\.\.//g;
    s/\{.*?\}//g;
    s/\(.*?\)//g;
    s/table.*?://gi;
    s/1\ item.*?sp\;//gi;
    s/2\ item.*?sp\;//gi;
    s/\s+/ /g;
    s/Telnr: /\n/g;
    s/Naam: /\n/g;
    s/Adres: /\n/g;
    s/Plaats: /\n/g;
    s/Postcode: /\n/g;
    s/Fax: /\nfax: /g;
}

print $info;
[download]

Teabag
Sure there's more than one way, but one just needs one anyway - Teabag

Nope, it isn't. And according to testing on my windows machine and my linux machine this tiny utility doesn't think it is either.

2;0 juerd@ouranos:~$ perl gettel.pl '!!!!!!!!!!!!!!hallo wereld!!!!!!@
+#$@#$%^#$%^#$%^#"$%#
$%#"$%"#$%#$%"#$%123-4567890!!!!!!!!!!'

No info found on that number
[download]

Get rid of the \bs and get real anchors instead :) Also, the 4-digit area code issue is about area codes. Has nothing to do with the 090[069] service/sex/game numbers.

112          People with sirens :)
00 800 x+    International toll free
00 x+        International anything
067[0-5] x+  Data services
06760 x+     Data: internet dial-up
0800 x+      Toll free
0900 x+      Information
0906 x+      Erotic
0909 x+      Entertainment
16xx .*      Carrier select
0xx xxxxxxx  Normal (084 and 087 are for PDA-ish services)
0xxx xxxxxx  Normal
[download]

Detailed information: http://www.ez.nl/beleid/home_ond/dgtp/beleidwetgeving/nummers_namen/documenten/numberingplan_netherlands.pdf

Juerd # { site => 'juerd.nl', plp_site => 'plp.juerd.nl', do_not_use => 'spamtrap' }