Transaction balancing
Greg Stark
gsstark at mit.edu
Fri Jan 24 11:07:31 CST 2003
Derek Atkins <warlord at MIT.EDU> writes:
> Greg Stark <gsstark at MIT.EDU> writes:
>
> > Derek Atkins <warlord at MIT.EDU> writes:
> >
> > > I think it's a bit more subtle that that. Each import mechanism has
> > > different 'strings' to match on. QIF, for example, has a PAYEE, MEMO,
> > > DESCRIPTION, Category, and I think "Account". OFX has different
> > > strings to search on. HBCI is different from both the above.
> > >
> > > I do not think there is a one-size-fits-all system.
> >
> > You left out the amount field. Lots of my transactions come through with the
> > same textual description. The only way I have of determining what they were is
> > that they're always for the same amount.
>
> Well, there is the question of "account" matching vs. "duplicate"
> matching. For duplication matching I agree -- you definitely can (and
> should) use the amount. For "account" matching (i.e., what is the
> destination account for this transaction), using the amount doesn't
> help you.
No really, it helps me. Honest. I'm not making this up.
My paycheck is the same amount week after week for months. I had automatic
deposits set up previously, but now I receive checks and deposit them at an
ATM it just says "Deposit" with the ATM number.
There's at least one other check I receive and two that I write once a month
that are always for the precisely the same amount month after month after
month. I deposit these at an ATM every month and it shows up as an ATM deposit
or withdrawal with no other identifying information.
I write a check to someone else once a month for exactly the same amount every
time. My statement says "Check #nnn" which is useful for me but not useful for
Gnucash to guess what expense account it should be transferred to. If I
haven't entered the check manually it won't have a clue. But if it's for
precisely the same amount as last month...
Consider rent. The same amount month after month...
In fact I would add the day of the month as a useful field. It even lends
itself to a simple percentage calculation for how close two days are within
the month. A lot of checks are cashed or arrive at more or less the same time
of every month.
> > I think it would be possible to throw a general purpose pattern matching
> > system at this. Even a simple system like the bayesian filters people are
> > using to pre-sort their mail should work great.
>
> You willing to help write it? ;)
Well sure. But, uh, not today.
--
greg
More information about the gnucash-devel
mailing list