Question

I currently have the following regex to parse data. And an array of “exclusions”

$userNameArray = (userName1, user Name2, User Name 3);

$re = '/^(?<timeMined>[0-9]{2}:[0-9]{2}:[0-9]{2}) # timeMined 
     \s+
     (?<userName>[\w\s]+)        # user name
     \s+(?:has\s+looted)\s+    # garbage text between name and amount
     (?<amount>\d+)              # amount
     \s+x\s+                     # multiplication symbol
     (?<item>.*?)\s*$            # item name (to end of line)
   /xmu';
preg_match_all($re, $sample, $matches, PREG_SET_ORDER);
foreach ($matches as $value){
    code
}

My code currently has an if statement that if $value['userName'] is in $userNameArray to execute part of a code, and if not, do a different part. However this would be significantly easier if I could just parse the bad users out in the regex.

Answer 1

While you could use a negative lookahead as in

$re = '/^(?<timeMined>[0-9]{2}:[0-9]{2}:[0-9]{2}) # timeMined 
     \s+
     (?!user1|user2|user3)       # exclude users <--
     (?<userName>[\w\s]+)        # user name
     \s+(?:has\s+looted)\s+    # garbage text between name and amount
     (?<amount>\d+)              # amount
     \s+x\s+                     # multiplication symbol
     (?<item>.*?)\s*$            # item name (to end of line)
   /xmu';

that would encode significant application logic into a regular expression. Most likely, your current solution is easier to understand, more readable, and easier to change.