5 ms·

Doesn't seem to handle pdfs properly. For the mtgox link it comes up with > stream rotate type/page font structparents endobj obj endstream

• doppenhe 12y ago
  the demo version if we cant scrape the text from the HTML we cant really run the topic analysis against it. pdfs, images, etc wont work.
  • r00fus 12y ago
    PDF is easy to support if you use pdftk. You can even simply get the first page for large docs. http://www.pdflabs.com/tools/pdftk-server/ http://www.pdflabs.com/tools/pdftk-server/
    • doppenhe 12y ago
      thanks for the tip.