What I would like to do is for example after all the php code is finished (so the finall html page has been rendered and ready to be send to browser) I would like to parse the page for links. For example I would like to change all the email links to images or add a pdf pic in front of all pdf links...
The thing is that I don't know how is this actually done, what is the right aproach. At the moment I have a index page where I include other php (sub)pages (which may also have some php code). And in the index.php page is also a lot of php code. I don't want to parse the php files before the code was executed.
So What I would like to do is, that after all the php code to render the page was executed, I would like to parse it and serch for links and than do something with them.
Given the filename, your function will include the file. It will process any and all PHP. And what gets outputted will get "caught" by the output buffering. This is then returned from the function. So what you get out of your function is the output of your PHP file, which in your case should be pure HTML.
Once you have this HTML in the variable "string", you can further parse/manipulate it or output it to the user.
Since you want to add PDF images to the front of PDF links, you can then add code to loop through the HMTL content looking for PDF links and inserting the image code into the original string. This will require some minimal thinking to setup the coding, but is relatively simple.