Rotate text files based on size - php

Question

I am trying to record hits to my website to a text file and I want to limit the size of the text file to some value. Once the limit is crossed I want a new file to be created dynamically. With my current code, a new file does get created after the set limit is passed, but the rest of the data is only stored on that new file.
public function storeActivityHitCount(Request $request)
{
if($request->ajax() && isset($request->data)){
$clientIP = request()->ip(); $i= 1;
$date = date("Y-m-d h:i:sa");
$data = $clientIP.', '.$date.', '.$request->data;
$file = public_path().'/adminpanel/hits/activity/activity'.$i.'.txt';
if(!file_exists($file)){
fopen($file,"w+");
}
$filesize = filesize($file);
if($filesize >= 76){ $i++;
$file1 = public_path().'/adminpanel/hits/activity/activity'.$i.'.txt';
if(!file_exists($file1)){
fopen($file1,"w+");
}
$content = file_get_contents($file1);
$content .= $data. PHP_EOL;
$upload_success = file_put_contents($file1, $content);
}else{
$content = file_get_contents($file);
$content .= $data. PHP_EOL;
$upload_success = file_put_contents($file, $content);
}
if($upload_success){
return Response::json(['status' => 'success'], 200);
}
return Response::json(['status' => 'failed'], 400);
}
abort(403);
}

Answer 1

Your original code of course was only trying one other file and then writing to it regardless of its size. You want to put that logic into a repeating structure. Or in other words, you want to look for a different file while you keep finding full ones.
public function storeActivityHitCount(Request $request)
{
if ($request->ajax() && isset($request->data)) {
$clientIP = request()->ip();
$date = date('Y-m-d h:i:sa');
$data = $clientIP . ', ' . $date . ', ' . $request->data;
$i = 1;
$file = public_path() . '/adminpanel/hits/activity/activity' . $i . '.txt';
while (filesize($file) >= 76 && $i <= 20) {
$i++;
$file = public_path() . '/adminpanel/hits/activity/activity' . $i . '.txt';
}
if (filesize($file) >= 76) {
// the loop got away from us
// do something?
$upload_success = false;
} else {
$upload_success = file_put_contents($file, $content, \FILE_APPEND);
}
if ($upload_success) {
return Response::json(['status' => 'success'], 200);
}
return Response::json(['status' => 'failed'], 400);
}
abort(403);
}
I put an upper limit of 20 iterations on the loop; you usually don't want a while loop without some kind of escape mechanism.
file_put_contents will always create a file that doesn't exist, so you didn't need to use fopen (and, you weren't using fclose.) Furthermore, if you pass the FILE_APPEND flag, it will append to the existing file; no need for getting contents and appending stuff to it.
Try to keep your code consistent and readable. Multiple commands on one line, inconsistent whitespace around control structures and operators, inconsistent indentation: all these things end up making you work harder than you have to.
And, of course, this would all be better handled by standard system tools like logrotate which is probably running on your server.

Answer 2

I would isolate the action of getting the activity log file. Like using a function like this:
function getAvailableActivityLogFile(){
$looking=true;
$i=0;
while($looking){
$i++;
$file_path = public_path() . "/adminpanel/hits/activity/activity{$i}.txt";
// If file does not exist or file is less than 76bytes you have the right candiate
if(!file_exists($file_path) || filesize($file_path) < 76){
$looking=false;
return fopen($file_path, 'a+');
}
// Otherwise keep looking on next iteration. You can also write some logic to have a max of loop iterations or max files to look for also.
}
}
Then you can use this function to get the next available file and don't bother with too much logic about what file is available. In my version about you should use fwrite() to write to the file using the file pointer returned by the function.
With the a+ option, you get a pointer there that appends content to the file on every new fwrite to the pointer.
You can also write the function to retrieve a path instead of a pointer.