preg_match not matching strings across multiple lines

[ursvmg]

Hi all,

I'm trying to pattern match street address from a paragraph. My RexEx code is able to match the address if it is found in a single line. But I'm not able to make it work for an address that spreads over multiple lines.

 

Following is the code:

 

$address = "I need a Regular expression to find a complete postal address for example Jack John 
1100 Glendon, Los Angeles, CA, 90024 United States . Thanx in advance i need it very urgently ";

$pattern1 = "/((\d+)[\-, ]*(\w+[\- ]*\w+[\- ]*\w+))[\-, ]*(\w+[\- ]?\w+)[\-, ]*([A-Z]{2})[\-, ]*(\d{5})/";

if ( preg_match( $pattern1, $address, $matches ) == 1 ) {
echo "Match found";
}

 

The above gives out "Match found" as output.

 

But I need to modify this RegEX to work for the paragraph below.

"I need a Regular expression to find a complete postal address for example

Jack John

1100 Glendon,

Los Angeles,

CA, 90024

United States

Thanx in advance i need it very urgently "

 

I tried the following RegEx. But it is not working

$pattern1 = "/((\d+)[\-, ]*(\w+[\- ]*\w+[\- ]*\w+))\s(\w+[\- ]?\w+)[\-, ]*([A-Z]{2})[\-, ]*(\d{5})/s";

 

Kindly help me on this.

 

P.S. I'm not very much familiar with advanced topics of RegEx. I arrive at this RegEx only with the help of google.

 

Regards,

ursvmg

[spfoonnewb]

<?php
$address = "I need a Regular expression to find a complete postal address for example Jack John 
1100 Glendon, Los Angeles, CA, 90024 United States . Thanx in advance i need it very urgently ";

$pattern1 = "/((\d+)[\-, ]*(\w+[\- ]*\w+[\- ]*\w+))[\-, ]*(\w+[\- ]?\w+)[\-, ]*([A-Z]{2})[\-, ]*(\d{5})/";

if ( preg_match( $pattern1, $address, $matches ) == 1 ) {

echo "<pre>";
print_r($matches);
echo "</pre>";
}

?>

[cags]

Your pattern makes extensive use of the \w backslash character. This stands for 'non-vertical whitespace' characters, therefore they will never match vertical whitespace. The sections in between these you only allow for '\', '-' and ' '. You will need to decide where you wish to allow line breaks and add in the appropriate characters as required ('\r' and '\n').

[cags]

Salathe just pointed out that I obviously miss-represented the definition of \w in that last post, \w of course represents word characters, not non-vertical whitespace, but the point still stands that it will not match \r and \n characters.

[ursvmg]

When i viewed the html source code, it had something like below.

"I need a Regular expression to find a complete postal address for example

<br>

Jack John

<br>

1100 Glendon,

<br>

Los Angeles,

<br>

CA, 90024

<br>

United States

<br>

Thanx in advance i need it very urgently "

 

So I did a work around by removing all the html tags using strip_tags() function. Now i have continuous piece of data to pattern match and its working fine.

 

Please let me know if there is any other easy way to do it.

 

Regards,

ursvmg

[cags]

Ahh, if it is <br> tags that are splitting the data then strip_tags is probably as good as anything. If they are in particular places you could add them to the pattern if you wanted, but if you don't need the tags I don't see a downside to using strip_tags.