EDGAR 10 K Filings - parsing the html

Anyone done something like this for their company? I know its standard read text, parse text, etc… but 10K’s have lots of gobblygook I was hoping to avoid.

Thanks