[SOLVED] phone matching U.S. format

[2levelsabove]

Hello why is this not working ? all i want is to be able to match numbers in format of

 

xxx-xxx-xxxx  or (xxx) xxx-xxxx

 

 

 

$phone_number = 'sjkadkjsahsahdsjkjdjhdshkjdk jsakhkjhk 469-767-7695 sadsadsadsadsad';



$pattern3='/^([0-1]([\s-./\\])?)?(\(?[2-9]\d{2}\)?|[2-9]\d{3})([\s-./\\])?(\d{3}([\s-./\\])?\d{4}|[a-zA-Z0-9]{7})$/';

if (preg_match($pattern3, $phone_number, $matches)) 
{
    // we have a match, dump sub-patterns to $matches
    $phone_number = $matches[0]; // original number
    $area_code = $matches[1];    // 3-digit area code
    $exchange = $matches[2];     // 3-digit exchange
    $number = $matches[3];       // 4-digit number
    $extension = $matches[4];    // extension


print_r($matches);

}

[Daniel0]

Try this: ^(?:\((\d{3})\) |(\d{3})-)(\d{3})-(\d{4})$

It'll match both 123-456-7890 and (123) 456-7890

[nrg_alpha]

I have used contionals:

 

$str = '(123) 345-6789';
if(preg_match('#^(\()?\d{3}(?(1)\)\x20|-)\d{3}-\d{4}$#', $str, $match)){
echo 'The number format is good...';
} else {
echo 'The number format has errors...';
}

 

You can alter $str to not have brackets (but you will then need to add a dash as in the xxx-xxx-xxxx example in the OP.

If you only remove one bracket, you should have an error regardless to whether there is a space or dash after the first 3 digits..

if both brackets are there, then there must be a space following the area code's closing bracket (as in (xxx) xxx-xxxx in the OP).

 

So the basic breakdown is this:

 

^(\))?  from the start, find a '(' character as optional

\d{3}   then find three consecutive characters

(?(1)\)\x20|-)   this is the contional part.. if there is a capture from the first set of parethesis, then look for the closing ')' character, followed by a space, otherwise, look for a dash

\d{3}-\d{4}$  then, finally look for 3 digits, a dash then 4 digits.

 

[2levelsabove]

Daniel,

 

ok i tried this. but still nothing is outputted.

 

$phone_number = 'sjkadkjsahsahdsjkjdjhdshkjdk jsakhkjhk 469-767-7695 sadsadsadsadsad';



$pattern3="/^(?:\((\d{3})\) |(\d{3})-)(\d{3})-(\d{4})$/";

if (preg_match_all($pattern3, $phone_number, $matches)) 
{
    // we have a match, dump sub-patterns to $matches
    $phone_number = $matches[0]; // original number
    $area_code = $matches[1];    // 3-digit area code
    $exchange = $matches[2];     // 3-digit exchange
    $number = $matches[3];       // 4-digit number
    $extension = $matches[4];    // extension


print_r($matches);

}

[Daniel0]

Well, nrg_alpha's pattern is better because it won't result in empty backreferences.

 

Also, the ^ and $ means "match start of string" and "match end of string", respectively, so you cannot have that. You could remove those meta-characters.

[nrg_alpha]

Oops.. I was so fixated on the xxx-xxx-xxxx or (xxx) xxx-xxxx part I forgot that it was in a string with other characters in it.. (in my example, $str contains only the phone number).. should have included the sample sentence in the OP..so yeah, remove the ^ and $ characters from the pattern. My mistake 

 

[2levelsabove]

Thanks a lot guys !!!

 

you guys are awesome!!

 

 

Im just so new with regex!

[nrg_alpha]

Hold the phone guys... now that I test my pattern against:

$str = 'sjkadkjsahsahdsjkjdjhdshkjdk jsakhkjhk 469-767-7695 sadsadsadsadsad';

 

My regex doesn't hold I don't think (not when I start playing with brackets surrounding the area code).. I'll revise my pattern.

 

 

[nrg_alpha]

ok.. here is a quick revised version.. this version assumes only word characters surround the phone number:

 

$str = 'sjkadkjsahsahdsjkjdjhdshkjdk jsakhkjhk (469) 767-7695 sadsadsadsadsad';
if(preg_match('#(?:\w+)?(\()?\d{3}(?(1)\)\x20|-)\d{3}-\d{4}(?:\w+)?#', $str, $match)){
echo 'The number format is good...';
} else {
echo 'The number format has errors...';
}

 

So I added (?:\w+)? at each end of the patter.. this does not capture if it does find your typical [a-zA-Z0-9_] characters one or more times (all of these optional of course).

Running this preg expression against the updated $str seems to hold up.

 

 

[2levelsabove]

Thanks alpha. I appreciate your help on this matter