正则表达式：标识不同格式的电话号码

Question

I have a website that people write jokes there. users can send jokes they like to their (or their friends) phones as sms. and the sender of the joke (who has added the joke to the site) displayed below it:

Joke #12234

this is the body
of the joke

sender: John

some times people use their phone numbers as the sender name, that is not allowed in public. I want to determine if there is a phone number in the sender name to be able to censor it. I assume any number bigger than 6 digits as a phone number. but the problem is that user might separate the numbers like:

1234567890 should become 1234XXX7890
123 456 7890 should become 123 XXX 7890
123-456-7890
123456-7890

and so on. any of the forms similar to above formats should be censored. I tried removing non numeric characters and then use regular expressions but the problem is then it also fetches:

john23 peterson12345

can anyone offer a better way?

Answer 1

To keep all the formatting, replace

(\d{3}[-\s()]*)\d{3}([-\s()]*\d{4})

with

$1XXX$2

To further constrain for 10-digit numbers ( ie disallow additional numbers immediately before and after), use negative lookaround assertions:

(?<!\d)(\d{3}[-\s()]*)\d{3}([-\s()]*\d{4})(?!\d)
^^^^^^^                                   ^^^^^^

Finally, what if typos lead users to insert space or symbols between groups, eg (123)45 6-7890 ? To catch these too, do the following:

(?<!\d)((?:\d[-\s()]*){3})(?:\d[-\s()]*){3}((?:\d[-\s()]*){4})(?!\d)

This may, however, catch "too much," eg 1-2-3-4-5-6-7-8-9-0 . You will have to determine what balance you want to strike.

Answer 2

because of a lot of phone number format in this world, you can use this regex to identify any phone number. ^[0-9-+s()]*$