[GNC] Best match probability calc is too pessimistic

David Reiser dbreiser at icloud.com
Sun Jul 5 16:28:21 EDT 2020


Done. Bug 797847 <https://bugs.gnucash.org/show_bug.cgi?id=797847>

I left out the moribund straggling check issue, since it’s a much lower priority to me. I write about 5 checks a year these days. The credit card import issue pretty much happens every time I import.

Dave
--
Dave Reiser
dbreiser at icloud.com





> On Jul 5, 2020, at 3:11 PM, Jim DeLaHunt <list+gnucash at jdlh.com> wrote:
> 
> On 2020-07-05 11:49, David Reiser via gnucash-user wrote:
>> …From my perspective, transactions posting within 3 days of the transaction date should still get a match score of 6. One day after transaction ought to be 8. One of the grocery stores I go to frequently has never posted the charge to the credit card company the next day. Sometimes there’s only a 2-day gap, but it is frequently 3 days. I suspect the difference between 2 and 3 days is related to how late in the evening I go shopping.
>> 
>> And on the extreme end, for a checking account transaction, if the imported check number (complicated by bank-prepended zeroes) and amount match exactly, and the date in gnucash is within 180 days of the bank’s date, then the transaction should still have some usable match score. In the U.S., the bank will still pay that check. I have had friends sit on checks for excessive lengths of time.
> 
> David, I completely agree with you. I have been doing a lot of transaction importing from OFX recently, and it's a continuing problem that I can't trust the matcher to handle dates the way you describe.
> 
> Would you care to add a bug report for this into the bug tracker, <https://bugs.gnucash.org/> ? If so, I'll second it.
> 
> Best regards,
>     —Jim DeLaHunt, software engineer, Vancouver, Canada
> 



More information about the gnucash-user mailing list