Question

My current MSSQL table has a “phone” column which is a varchar. Unfortunately, the phone numbers that are already in the database are not in standard format. For example, 888-888-8888 OR 888/888/8888 OR (888)8888888 OR 8888888888.

I want to get all the rows that are equivalent to 88888888, i.e it should match with 888-888-8888, (888)888888 etc.

I have tried using REPLACE() but there are certain rows where entries have other alphabetic characters like “e”, “ex”, “ext”, etc. So I would like to replace all non-numeric characters.

What would be the best way to get the “matching” rows using MSSQL query?

Answer 1

A simple version using MySQL:

SELECT * FROM `phones` WHERE `phone` LIKE '%8%8%8%8%8%8%8%8%8%8%'

Using PHP:

// Get all your table rows into $rows using SELECT ..
foreach ($rows as $row) {
    $row['phone'] = preg_replace('/\D/', '', $row['phone'];
    // Save the row using UPDATE ..
}

The regular expression \D matches any non-numeric character. See php.net/preg_replace for more information.

If you just want to find a row that matches “8888888888”, then you could use:

if (preg_match('/\D*8\D*8\D*8\D*8\D*8\D*8\D*8\D*8\D*8\D*8\D*/', $row['phone'])) {
    ..
}

Which could simplify/abstract to:

$match = '8888888888';
if (preg_match('/' . preg_replace('/(\d)/', '\D*$1', $match) . '\D*/', $row['phone'])) {
    ..
}