QIF Import fails out of CVS

Derek Atkins warlord@MIT.EDU
13 Feb 2001 17:18:17 -0500


grib@gnumatic.com (Bill Gribble) writes:

> On Tue, Feb 13, 2001 at 03:54:37PM -0500, Derek Atkins wrote:
> > > Save your whee until it works :)  
> > 
> > True enough.  That patch fixed the SEGV.  But it's not done yet. ;)
> 
> Did you ever have One Of Those Days?  

Ayup.  For example, I completely misinterpretted the C code earlier..
Those arguments to "ash" should really be -16, -32, and -48, instead
of whatever schlock I had.

> After all my "it's absolutely a GC problem... probably related to
> g-wrap" yang-talk, the truth comes out: in the SIGFIGS gnc-numeric
> code I added yesterday or day before, I was taking the log10 of a
> number without ensuring that it was >0 .  I can't be 100% sure that
> this was the problem, but it *is* a problem and I think it's the one
> at the bottom of all this.

I still don't think that the scm_to_gint64 has any prayer of
working...  However, your most recent patch does actually work.

> I think the code in gnc-helpers.c can go too, but I want to try this
> first.  If this patch fixes the problem, could you try it again with
> the straight CVS version of gnc-helpers.c?

I'm doing that now.  I'll send another message shortly.  The only
thing I _can_ tell you is that the importer wound up creating a new
(duplicate) account instead of using an existing account.

> Thanks
> b.g.

No, thank YOU! :)

-derek 
-- 
       Derek Atkins, SB '93 MIT EE, SM '95 MIT Media Laboratory
       Member, MIT Student Information Processing Board  (SIPB)
       URL: http://web.mit.edu/warlord/    PP-ASEL-IA     N1NWH
       warlord@MIT.EDU                        PGP key available