Find all strings of a text that have XXX at the EOL - php

Question

I have a text file with 700,000 lines of login attempts some of which are successful but mainly - not. Here is an example:
login attempt 2 to server IP as user_name - password failed
login attempt 3 to server IP as user_name - password failed
login attempt 4 to server IP as user_name - **successful**
login attempt 5 to server IP as user_name - password failed
login attempt 6 to server IP as user_name - **successful**
and so on. How can I delete all lines that do not end with 'successful" word?
pseud-ocode:
convert all lines to separate arrays with each word as an array element
to write something like this (pseudocode):
while(line_number <= 700000) {
$all_occurrences .= (end($array) == 'successful') ? whole_line : '';
}
so only these will remain:
> login attempt 4 to server IP as user_name - **successful** login
> attempt 6 to server IP as user_name - **successful**
Any thoughts?
Thank you!

Answer 1

The best thing is to read the files line by line so you don't run into memory constraints.
Solution 1:
<?php
$if = '/tmp/login.attempts';
$of = '/tmp/login.attempts.purged';
$ifh = fopen($if, 'r');
$ofh = fopen($of, 'w');
while(($line = fgets($ifh)) !== false) {
if(preg_match('/successful/', $line)) {
fwrite($ofh, $line);
}
}
fclose($ifh);
fclose($ofh);
?>
Or if you want to use arrays like your sudo code.
Solution 2:
<?php
$if = '/tmp/login.attempts';
$contents = file_get_contents($if);
$all_attempts = explode("\n", $contents);
$successful_attempts = array();
foreach($all_attempts as $attempt) {
if(preg_match('/successful/', $attempt)) {
$successful_attempts[] = sprintf("%s", $attempt);
}
}
// The successful_attempts array contains all your entries
//print_r($successful_attempts);
?>