Image enabling gnucash?

Dave Peticolas
Thu, 22 Feb 2001 16:56:14 -0800

Bakki Kudva writes:
> Hello everyone,
> I am new on this list and gnucash. I do have it on my Red Hat 6.2 box 
> and am very impressed with what I see.
> I am a consultant/integrator in the business document management 
> solutions. One of the major application of 'imaging' is to image enable 
> legacy accounting apps, usually running on some host via some sort of 
> terminal emulator. Using screen scraping  host based accounting apps can 
> be image enabled so that scanned images associated with particular 
> transactions  can be indexed(with additional indexes if necessary), 
> stored, retrieved and displayed. Most of the time these supporting 
> documents are scanned tiff images but could also be any type of file. 
> Most businesses spend a  lot of time handling supporting paper documents 
> and imaging can really improve productivity.
> I browsed through the project goals at but didn't see anything 
> related to image (document) enabling of gnucash. I wanted to get some 
> opinions on this feature. I am a programmer with reasonable skills and 
> would be interested in adding these features if there is interest.
> The only modifications to gnucash would be that register windows have 
> additional buttons called 'scan', 'attach', 'documents'. These buttons 
> would be tied to method calls into the imaging client with index info. 
> Scan and attach commands would invoke a scan program via 'SANE' or allow 
> attachments much like email. Additional index fields could be defined 
> and filled in (vendor id#, account# or whatever) in addition to the 
> automatic fields such as date, transaction id etc. Seperate window would 
> display the images and/or other apps launched based on mime types. Back 
> end could be mysql and the imaging engine can have its own API so that 
> other apps can share the functionality.
> So what do you folks think?!

To be honest I had never even heard of this application, but it sounds
very interesting. And anything that uses the term 'screen scraping' is
ok in my book.

It sounds like this would be a fairly specialized application, so it
would need to be something that was turned off by default so as not
to clutter up the interface for personal users. But as long as the
code is modular I don't see why it couldn't go into CVS.

A good place to start would be the HACKING document in the gnucash
CVS repository. Please do all development using the HEAD branch.