Edward Ross wrote: > We want to use podofo to extract text from pdf files. Has anyone used > podofo for this purpose? Does anyone have any example code which does this?
PoDoFo has, AFAIK, no built in support for text extraction. Doing so requires that the library parse content streams, which it can not yet do. Even with a content stream parser, that only gets you half way. Handling columns is hard unless the PDF is properly annotated (I've never seen one that was) and requires quite a bit of analysis. A naieve extraction approach will read text like: A black dog where it was seen terrorized at the market shoppers. this week as: a black dog where it was seen terrorized at the market shoppers. this week ... which isn't exactly what you wanted. Other complications exist, especially when handling PDF from applications that try to do their own character positioning instead of relying on PDF's rules about font metrics and drawing (perhaps to work around older broken viewers/RIPs?). I think the main tool I've seen for this is pstotext, which uses GhostScript. This may or may not fit your requirements. -- Craig Ringer ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Podofo-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/podofo-users
