That page - apart from being marked-up in a rather old-fashioned way - isn't too bad at all. If you look at the page source code, you can easily see a table structure that you can use to parse it.
You will want to use a module like WWW::Mechanize
to interact with the website. This moduel allows you to interact with web content like a user would in a browser. You can make your script "click" on links, to get to the text files. Use the table structure of the "browse" page to iterate over all the molecules, each time following the link through to the text data files.
Have a go with a simple example first. There are a few here
. If you are getting stuck, post the script you have so far and what's happening so we can help you along. Good luck!