An experimental parser for PDF Campaign Finance Reports from the Texas Ethics Commission.
Right now you have to edit the source to modify which pages it scrapes. Currently it's hardcoded to scrape all the contributor-containing pages in data/test.pdf (pages 4-46).
$ git clone [email protected]:brandonrobertz/parse-tx-cfr.git
$ cd parse-tx-cfr
$ lein run
Running it will give you a list of vectors, each of which contains the information from a contributor cell in data/test.pdf. I don't clean up the output or convert it to a delimited format. As you can see, it's fairly accurate as-is, with the exception of some line breaks and formatting oddities.
There are incompatabilities between GPL and Snowtide's PDFTextStream license (proprietary). This makes it impossible to distribute it as-is. Eventually I will port this to a free PDF parsing library, but until then you're on your own for installing PDFTextStream.
Copyright © 2013 Brandon Robertz GPLv3+ (w/ considering PDFTextStream a system lib) RIP EWOK BATES
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent