UK Phone Number regex validation?

[ozzyman]

Can anyone good at regex show me how to put all of these regex validations into 1 please?

//don't allow country code
String regex1 = "/^(\+)[\s]*(.*)$/";

//check all chars are digits
String regex2 = "/^[0-9]{10,11}$/";

//check first digit starts with 0
String regex3 = "/^0[0-9]{9,10}$/";

//check that the number is appropriate
String regex4 = "/^(01|02|03|05|070|071|072|073|074|075|07624|077|078|079)[0-9]+$/";

thanks

[Eranga]

Something like this?

^\s*\(?(020[7,8]{1}\)?[ ]?[1-9]{1}[0-9{2}[ ]?[0-9]{4})|(0[1-8]{1}[0-9]{3}\)?[ ]?[1-9]{1}[0-9]{2}[ ]?[0-9]{3})\s*$

[Eranga]

If the number is start with zero then anyway no need to worry about the + sign at the beginning. Because zero at the first place means + sign omitted anyway.

Only the digits can validate with a simple rage [0-9]

[ozzyman]

hi Eranga, thanks for the reply. i was hoping you'd also explain it a bit to me! i've broken up your regex pattern to try to understand it but i don't understand parts of it. this is what i've got:

^\s*
begins with a whitespace character, zero or several times

\(?
no or exactly 1 open-bracket

Section A:
-------------------------------
(020[7,8]{1}\)?
contains '020'
followed by -- 7 or 8 (what does comma do??) { ONE OCCURENCE }
followed by zero or one close-bracket

[]?
i dont understand this part

[1-9]{1}
any digit 1-9 once

[0-9]{2}
any digit 0-9 twice

[]?
again, don't understand

[0-9]{4})
any digit 4 times
------------------------------------

|
must match either Section A OR Section B


Section B
-----------------------------------------
(0[1-8]{1}
zero followed by any digit 1-8 ONCE

[0-9]{3}
any digit, thrice

\)?
optional close-bracket, why??

[]?
lol, really dont get it

[1-9]{1}
any digit 1-9 once

[0-9]{2}
any digit 0-9 twice

[]?
someone please tell me what this is!

[0-9]{3})
any digit, thrice

\s*$
ends with any number of white-spaces
-----------------------------------------

[ozzyman]

i've also come up with this:

^(0[12357])[0-9]{9}$

starts with 0 followed by one of the following: 1,2,3,5,7
the rest is any digit 0-9, total 11 digits only

(apparently 10 digits is extremely rare so i'm not that bothered about 10 digit numbers now)

[Eranga]

Range and number define with [0-9]{3}, any digit from zero to nine and only 3 digits can be used.

If you define []? collectively it has no meaning. Separate them into two [] and ?, then only it gives a meaning. ? sign matches the preceding character 0 or 1 times only. [] matches anything inside the brackets for 1 character position once and only once.

| is the pipe. As you said pipe matches either side. But you can use it powerfully with grouping. For example abcd(ef|gh) matches both abcdef and abcdgh.

No need of really use white-space validations. But this regex I've written for a selection on a paragraph. So there could be white-spaces, leading and trailing.

Have I missed anything?

[Eranga]

i've also come up with this:

^(0[12357])[0-9]{9}$

starts with 0 followed by one of the following: 1,2,3,5,7
the rest is any digit 0-9, total 11 digits only

(apparently 10 digits is extremely rare so i'm not that bothered about 10 digit numbers now)

Seems this is fine. As I said no need to worry about the country code if you are validating initial pattern.

Is this validate with all your numbers you've. Since I'm not in UK I don't have a better understanding about the number patterns.

[ozzyman]

yeah its okay really because in my system the user is not inputting the data its the staff so i dont really need validation but i thought it'd be better to add some anyway. they made us think like that when i was doing Computing at college.

[Eranga]

Building up those kins of regex, better to try as much as generalize. So you can use them in latter case with minimum effort.