File Downloads

[robin@19.5degs.com]

If a user is is downloading a file from the server via a php
link...there are 2 ways I can do it..

1. Create a symlink to the file and .. redirect user to it and delete
the symlink later

OR

2. Just send appropriate headers to the browser and read the file to
it.

I was wondering which of these 2 approaches will take more resources.
Is there a way to benchmark this?

Regards,
Robin

[Phil]

Re: File Downloads

robin@19.5degs. com wrote, On 22/09/06 6.38 a:

1. Create a symlink to the file and .. redirect user to it and delete
the symlink later

I suspect that this would not only be more effort, but prone to a lot
more failure.

For example, how would you detect when the user has finished downloading
the file and you need to clean up the symlink? How would you stop other
people from grabbing this symlink and also downloading the file? How
would you create unique symlinks? What if you run into filesystem
limitations?

-Phil