So, dose it mean that the "middle" page can't be redirected automatically?

It probably means that you will have to write to write the code to automate it. (Of course, someone could still pop up and say, "no, wait; there's a module for that on CPAN." I don't know of one though.)

The final page has a different URL than the "middle" page. The only way to fetch the result page is parse out of the "middle" page,right? Any other methods??

If the URL of the final page is subject to change and the only place to get it is from the "middle" page, then yes, you'll have to parse it out. But at least it shouldn't be particularly hard to do. You might want to be sure the URL actually does change before going to the trouble though. If it is the same across requests, then there's probably no need for you to look at the middle page at all.

-sauoq
"My two cents aren't worth a dime.";