Hi, I'm an Australian theoretical physicist who loves data processing and programming. I do a lot of web scraping work also.
Your project looks interesting, and I believe is easier than you think. The processing of the PDF text is almost trivial; here's the output from today's report
[login to view URL]
which I can process with text manipulators (sed/awk/grep, etc...) to provide you with an array of the data you require. I can also write a script (bash/PHP/R, etc...) to handle the downloading, text extracting, text processing, and file dump that you require daily.
I look forward to hearing from you.