PHP count and seperate text from text file

Let's say inside the text file is this information:

<Amanda> Hi there, how are you?
<Jack> Hi, im fine 
.
.
.
.
<Jack> see you later

I want to count the words each user have said the output should be for example like this

Amanda: 50
Jack: 40

First I want to not count the <Amanda> or <Jack> and next I want to count every words they said and insert it to variables Amanda and Jack

This is what I have done

    $usercount1 = 0;
    $usercount2 = 0;  

    //Opens a file in read mode  
    $file = fopen("logfile.txt", "r");  
    //Gets each line till end of file is reached  
    while (($line = fgets($file)) !== false) {  
        //Splits each line into words
        $words = explode(" ", $line);  
        $words = explode("<Amanda>", $line);  
        //Counts each word  
        $usercount1 = $usercount1 + count($words);  
    }

    while (($line = fgets($file)) !== false) {  
        //Splits each line into words  
        $words = explode(" ", $line);
        //Counts each word  
        $usercount2 = $usercount2 + count($words);  
    } 

As per my understanding this could be a possible solution.

// Input
$input = "<Amanda> Hi there, how are you?\n<Jack> Hi, im fine \n <Jack> see you later";

// Initialize counters
$amandaCount = 0;
$jackCount = 0;

// Split input by lines
$lines = explode("\n", $input);

// Loop over lines
foreach ($lines as $line) {
  // Remove user tags
  $cleanLine = preg_replace("/<.+?>/", "", $line);
  
  // Split line into words
  $words = str_word_count($cleanLine, 1);
  
  // Count words per user
  if (strpos($line, "<Amanda>") !== false) {
    $amandaCount += count($words);
  } elseif (strpos($line, "<Jack>") !== false) {
    $jackCount += count($words);
  }
}

// Output
echo "Amanda: $amandaCount\n";
echo "Jack: $jackCount\n";

I would go a more general approach. This way you can analyze all users. Using a blacklist, just exclude them.

• First go through all the lines and match for username and text.
• Rebuild data structure by iterating and counting up using a blacklist.

The blacklist is formatted like this, because finding keys is faster than finding values.

$input = <<<'_TEXT'
<Amanda> Hi there, how are you?
<Jack> Hi, im fine
<Jack> see you later
<John> Hello World, my friends!
<Daniel> Foo!
_TEXT;
preg_match_all('/^<([^>]+)>(.*?)$/m', $input, $matches);

$blacklist = ['Amanda' => 1, 'Jack' => 1];
$words = [];
foreach ($matches[2] as $index => $match) {
    $user = $matches[1][$index];
    if (isset($blacklist[$user])) {
        continue;
    }
    $words[$user] = ($words[$user] ?? 0) + str_word_count($match);
}
print_r($words);

Array
(
    [John] => 4
    [Daniel] => 1
)

I would implement the blacklisted names in the regex to filter them out as early as possible.

A negated lookahead ensures that Amanda and Jack are excluded. (?!Amanda>|Jack>)

The m pattern modifier changes the meaning of the ^ ("start of string" anchor) to be the "start of a line" anchor.

Parentheses around the name subpattern will create capture group 1 (accessible as element [1]). \K will restart the fullstring match, so the space-delimited words substring will be accessible via [0].

Use destructuring syntax in the foreach() for convenient variables.

Code: (Demo)

preg_match_all(
    '/^<((?!Amanda>|Jack>)[^>]+)> \K.+/m',
    $chat,
    $matches,
    PREG_SET_ORDER
);
$result = [];
foreach ($matches as [$words, $name]) {
    $result[$name] = ($result[$name] ?? 0) + str_word_count($words);
}
var_export($result);