QIF Importer

[Tom Brown]

On Thu, Jan 03, 2008 at 07:13:40PM -0800, Charles Day wrote:
> On Jan 3, 2008 6:03 PM, Ian Lewis <ianmlewis at gmail.com> wrote:
> 
> > Charles,
> >
> > You mentioned in some other e-mails that you were working on fixing some
> > problems with the QIF importer. What exactly are you working on? I'm looking
> > at trying to improve the importer myself and would like to coordinate any
> > changes in an effective way. That said, I'm new to coding Gnucash haven't
> > really thought much about the scheme part yet and was more looking at fixing
> > bugs in the import druid first, but eventually turn to the import itself
> > with the eventual (and lofty) goal of rewriting the QIF import to use the
> > generic import code that the OFX and other imports are using.
> 
> The only things I'm looking at so far are 100% in Scheme, so anything you do
> with the GUI or C shouldn't clash with what I'm doing. Even if you dig into
> the Scheme code, you probably won't conflict with me unless you are working
> on qif-to-gnc.scm and more specifically on the split matching/marking bits.
> At the moment I'm looking at bug
> 114724<http://bugzilla.gnome.org/show_bug.cgi?id=114724>,
> bug 506810 <http://bugzilla.gnome.org/show_bug.cgi?id=506810>, and another
> bug in the Scheme code (again with split matching/marking) that I haven't
> reported yet. I know absolutely nothing about the generic import code and I
> don't predict working on it at all in the immediate future, but who knows...
> 
> I've aimed for small fixes so far. If a big change is needed then maybe it
> would be more productive to work on the generic importer and make it happen
> there. I don't really have a sense for this at present.
> 
> So go for it!


I'm (very slowly) making progress on a unit test for
import-export/import-backend.c and will use the test to improve how it
finds matches for reconciling. For example /* Memo heuristics */ leaves a lot to
be desired.


-- 
28 70 20 71 2C 65 29 61 9C B1 36 3D D4 69 CE 62 4A 22 8B 0E DC 3E