HI Joost, I’m looking to download Exhibit 21 (which is a static/invariable SEC reference) from 10-K filings stored on EDGAR.  Instead of retrieving the entire 10-K text file, EDGAR stores this exhibit separately as “ex21.htm”  For example, see 2014 10-K filing for CIK 1141807. One problem that “I think” lies in wait is that the SEC index files only refer to text files, is that right?  If so, would you have any suggestions other than retrieving the entire 10-K text file? Thanks for any insight you may have!

The full 10-Ks don’t take up that much space, and exhibit 21 is clearly marked within the full filing (very easy to extract); I have no experience with pulling out separate/additional files, but if it is named consistently, pulling them in like that (since files) would work as well.

Hope this helps,

Joost

Hi Joost, how would you extract the exhibit, REGEX? Finding the beginning and end might be dicey, then strip the white space?

hi,

I forgot the exact details, but I remember identifying the beginning and ending is super-easy/trivial. Take a look, it is something with <DOCUMENT> and </DOCUMENT>, and with some identifier for exhibit 22. You won’t need a regular expression, just match against a string.

Best,

Joost