Question

I am making a PHP utility that imports and analyzes a CSV file into $data, and whether or not to INSERT “new” rows into the database ($saveNew).

Right now, I have a bit of an ugly mess: (in generalized pseudo-PHP)

function synchronize($data,$saveNew) {
    $existing_ids = $table->find_all('ID'); //array of ID's in the table
    $incoming_ids = get_key('ID',$data); //grabs the ID field from each record
    $new_ids = array_diff($incoming_ids,$existing_ids);

    foreach ($data as $record) {
        if (in_array($record['ID'],$new_ids)) { //it's new
            if ($saveNew) $table->insert($record);
            else continue;
        } else {
            $table->update($record);
        }
     }
 }

To me, this just has a smell, and I think I could accomplish this in just a single query, except that I’m not that familiar with SQL.

I’m using a simple ORM in my app, but I can easily just use straight SQL.

Oh, and I’m using MySQL.

Can this be done with just one query? It seems like this would be a common problem with a simple solution, but I just can’t figure it out.

Answer 1

Have a look at MySQL’s INSERT ... ON DUPLICATE KEY ... syntax, which allows you to do just that: http://dev.mysql.com/doc/refman/5.1/en/insert-on-duplicate.html

Whether you want to implement that in your ORM or in that specific subroutine of yours is up to you…