Hi there,

I want to validate UK phone numbers which are always 11 chars starting with a zero. That part is simple however, I would like to permit the user to also enter whitespace and/or hyphens. Opinion differs on where those characters should go so I'm happy to allow the user to do whatever they feel like and simply strip those chars out later. All the following should pass...

07816924120

078 1692 4120, 078-1692-4120

0781 692 4120, 0781-692-4120

07816 924 120, 07816-924-120

07816 924120, 07816-924120 even...

0 7 8 1 6 9 2 4 1 2 0 or...

078-1692 4120

Essentially as long as there's 11 digits (no more, no less), no consecutive whitespaces or hyphens, the first digit is a zero and the second digit is a non-zero digit it should match, everything else should fail. Happily I can ignore country codes and suchlike.

I'm would like the same regex to work in both python and javascript.

The regex I have does work but it's very ugly/ repetitive...

 ^[ -]{0,1}0[ -]{0,1}[1-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9][ -]{0,1}[0-9]$

Told you!

I'm assuming there's got to be a neater / more concise way of writing this, is that a fair assumption? and if so what's the trick?

Cheers,

Roger Heathcote

Matching on your given specs

Raw Match Pattern:
0[ -]?[1-9](?:[ -]?\d){9}\b

Javascript Code Example:
<script type="text/javascript">
  var re = /0[ -]?[1-9](?:[ -]?\d){9}\b/;
  var sourcestring = "source string to match with pattern";
  var results = [];
  var i = 0;
  for (var matches = re.exec(sourcestring); matches != null; matches = re.exec(sourcestring)) {
    results[i] = matches;
    for (var j=0; j<matches.length; j++) {
      alert("results["+i+"]["+j+"] = " + results[i][j]);
    }
    i++;
  }
</script>

$matches Array:
(
    [0] => Array
        (
            [0] => 07816924120
            [1] => 078 1692 4120
            [2] => 078-1692-4120
            [3] => 0781 692 4120
            [4] => 0781-692-4120
            [5] => 07816 924 120
            [6] => 07816-924-120
            [7] => 07816 924120
            [8] => 07816-924120
            [9] => 0 7 8 1 6 9 2 4 1 2 0
            [10] => 078-1692 4120
        )

)

---

Michael

"In theory, theory and practice are the same. In practice, they are not."
Albert Einstein

That's great mash, thank you very much. I've also now understand grouping, repetition and backreferences a lot better having been initially puzzled by your reply and having subsequently googled around for an explaination of what the "?:" part does :)

Cheers,

Roger.

The problem with your specification is that UK phone numbers are not all 10 digits after the 0 trunk prefix or +44 country code.

Some 41 area codes also contain at least some 9 digit numbers. Additionally, all 0500 and some 0800 numbers have only 9 digits after the leading 0 or +44 prefix.

If you're parsing input, don't allow only a leading 0. You should also allow +44, +44 (0), 00 44, 011 44, and so on.

Let the user enter the number any way they like. Parse it to extract the NSN part, then validate that part has the right number of digits and is in a valid range.

There's detailed listings at:

http://www.aa-asterisk.org.uk/index.php/Number_format

http://www.aa-asterisk.org.uk/index.php/Regular_Expressions_for_Validating_and_Formatting_GB_Telephone_Numbers

 

The simplest form for validating "'00 or 011 or + followed by 44 and optional 0' OR a '0'  - followed by '9 digits and other stuff intermixed'" limiting "other stuff" to things like spaces, brackets, periods, hypens, etc, is:

    ^\(?(?:(?:0(?:0|11)\)?[\s-]?\(?|\+)44\)?[\s-]?\(?(?:0\)?[\s-]?\(?)?|0)(([\s\(\).-]+\d){9,10})$

The NSN part is in $1.