Skeeve has asked for the wisdom of the Perl Monks concerning the following question:

Yes!, I found 159645 and while I find it very good, it is, I guess, too much for my purpose.

I just want to let (in the end) my computer dial phone numbers from my addressbook and maybe from some text.

For this I need to identify potential phone numbers in some text. Here implemented as the DATA section.

I crafted a lengthy regular expression that satisfies my needs and, I hope, most needs of other people. I'd like to make it a bit more flexibel and configurable. Especially in identifying and changing the phone numbers.

The idea is to change the phone numbers found to the shortest string of digits I have to dieal to reach that phone. So no need to verify that a number is correct but, assuming that it is correct, show what I would have to type in.

The rules here in Germany are:

  1. International calls start with "00"
  2. Calls to other areas start with "0"
  3. Calls in my area start with any digit but "0"

I think I allowed most of internationally used notations, but I'm not sure whether or not I interpret them correctly. E.g.: What about the x1234 x-tensions in 001-555-6666 x1234? What do they effectivly mean? What will I have to type in?

But now: Here is my code, open for discussion... 

use strict;
use warnings;

my @numbers;
while (<DATA>) {
    push @numbers, $1 while m#(\+?\d+(?:\s*[/\-()]?\s*\d)*(?:\s*x\s*\d
++)?)#g;
    # \+?           -> optional +
    # \d+           -> at least one digit
    # (?:           -> followed by
    #     \s*       ->   optional whitespace
    #     [/\-()]?  ->   followed by a seperator: / - ( )
    #     \s*       ->   optional whitespace
    #     \d        ->   one digit
    # )*            -> this is optional and might appeat more often
    # (?:           -> followed by maybe an x-xension
    #     \s*       ->   optional whitespace
    #     x         ->   'x'
    #     \s*       ->   optional whitespace
    #     \d+       ->   digits
    # )?            -> that's it...
}
exit unless @numbers;

# Now we got some numbers

# try to change them to what I would dial
my $minlen = 3;
my %replace= (
    '+' => '00', # How to dial to other countries
    '0049' => '0', # What would be my country code and what Do I use i
+nstead?
    '0241' => '', # What is my local area code and what Do I use inste
+ad
);
my $replace= join '|',map quotemeta,keys %replace;
$replace= qr/^($replace)/;

# parse each number
foreach (@numbers) {
    s/(?<=\d)\s*\(0\)//; # change the naughty (e.g.) +49(0) to +49
    tr/+0-9//cd; # Remove everything but "+" and digits
    1 while s/$replace/$replace{$1}/; # make the replacements as often
+ as possible
}

# print the resulting numbers line by line
print join qq(\n),grep {
    length($_) >= $minlen; # if they are long enough
} @numbers;
__DATA__
my area international a) +49241234567-1
my area international b) 0049241234567-2
another area +49211234567-3
same area but without international 0211234567-4
my area 234567-5
international number 001555234567x6

[download]
s$$([},&%#}/&/]+}%&{})*;#$&&s&&$^X.($'^"%]=\&(|?*{%
+.+=%;.#_}\&"^"-+%*).}%:##%}={~=~:.")&e&&s""`$''`"e

Replies are listed 'Best First'.
Re: Request for Comments: Phone Number Finder
by Anonymous Monk on Aug 30, 2007 at 08:52 UTC
    can always learn from "too much"
    http://search.cpan.org/dist/phone/

    E.g.: What about the x1234 x-tensions in 001-555-6666 x1234? What do they effectivly mean? What will I have to type in?
    1 2 3 4, duh :)

[reply]
      1 2 3 4, duh :)

      Sure? I thought so to but read somewhere that 1 2 3 4 is to be pressed after waiting for the connection. So yes, I have to type that, but not at once. But: Is my information correct? Or does it depend (on whatever)?

      s$$([},&%#}/&/]+}%&{})*;#$&&s&&$^X.($'^"%]=\&(|?*{%
      +.+=%;.#_}\&"^"-+%*).}%:##%}={~=~:.")&e&&s""`$''`"e
[reply]
[d/l]
[select]
        It depends on whatever, sometimes you have to tell the operator "extension 1234 please"
[reply]

Back to Seekers of Perl Wisdom