[End User]

Originally Posted by carters-site

My major was in Computer Science

Lol, that explains it. (Just kidding!)

Seriously though, as programmers, we think differently and we're often very literal. My friends will ask me if I know what time it is, and I'll say, "Yes."

They'll roll their eyes and ask me, okay, would I tell them what time it is, and I'll think for a moment, and I'll say, "Yes". They'll roll their eyes again, and yell, "Well, what &#$%! time is it??" And I'll say, "It's 2:30. Why didn't you ask?"

It's not my problem if they don't ask the right question, is it?

When my wife asks me if I want burgers or chicken for dinner, I say, "Yes!". Then she mutters a whole lot of words in Khmer that I don't recognize, but I'm pretty sure it's something about how much she loves me like crazy, or words to that effect.

[wirehopper]

I hope this helps.

This is an expanded version of the earlier code. It has two regular expressions. The first surrounds any digit strings with spaces and punctuation, the second removes any extra spaces from the entered string.

The $aStrings array simulates search entries from users, 'Potterton Puma 80e' is the the 'target'.

PHP Code:

<?php

$link = mysql_connect('localhost', 'user', 'password');

if (!$link)

    die('Could not connect: ' . mysql_error());



$aStrings=array('potterton','potterton     puma','80e','puma80e','pott puma','potterton 80e','potterton puma','puma 80 e','potterton

 80 e');

foreach ($aStrings as $k => $v)

{

        $sQuery='SELECT "Potterton Puma 80e" REGEXP \''.make_regexp($v).'\' AS \'found\'';

        $rResult=mysql_fetch_assoc(mysql_query($sQuery));

        echo (($rResult['found']==1)?'Match':'No match').' on '.$v.' [ '.$sQuery.' ]'."\n";

}

mysql_close($link);



function make_regexp($sString)

{

        return preg_replace(array('/(\d+)/','/[\s]+/'),array('[[:print:][:punct:]]*${1}[[:print:][:punct:]]*','[[:space:]]*'),$sStri

ng);

}

?>

Output (matches code above)

Code:

Match on potterton [ SELECT "Potterton Puma 80e" REGEXP 'potterton' AS 'found' ]
Match on potterton     puma [ SELECT "Potterton Puma 80e" REGEXP 'potterton[[:space:]]*puma' AS 'found' ]
Match on 80e [ SELECT "Potterton Puma 80e" REGEXP '[[:print:][:punct:]]*80[[:print:][:punct:]]*e' AS 'found' ]
Match on puma80e [ SELECT "Potterton Puma 80e" REGEXP 'puma[[:print:][:punct:]]*80[[:print:][:punct:]]*e' AS 'found' ]
No match on pott puma [ SELECT "Potterton Puma 80e" REGEXP 'pott[[:space:]]*puma' AS 'found' ]
Match on potterton 80e [ SELECT "Potterton Puma 80e" REGEXP 'potterton[[:space:]]*[[:print:][:punct:]]*80[[:print:][:punct:]]*e' AS 'found' ]
Match on potterton puma [ SELECT "Potterton Puma 80e" REGEXP 'potterton[[:space:]]*puma' AS 'found' ]
Match on puma 80 e [ SELECT "Potterton Puma 80e" REGEXP 'puma[[:space:]]*[[:print:][:punct:]]*80[[:print:][:punct:]]*[[:space:]]*e' AS 'found' ]
Match on potterton 80 e [ SELECT "Potterton Puma 80e" REGEXP 'potterton[[:space:]]*[[:print:][:punct:]]*80[[:print:][:punct:]]*[[:space:]]*e' AS 'found' ]

The REPLACE approach works, but there may be a performance hit, and, if any new characters are introduced, it would have to be updated.