[GNC] Matching error in OFX import

[Simon Roberts]

I've built 4.13 and the problem seems to have been fixed :)


On Sat, Dec 31, 2022 at 9:14 AM Simon Roberts <
simon at dancingcloudservices.com> wrote:

> Gach, yes, sorry meant to reply all, will pay attention in future,
> hopefully not do that again :(
>
> But FWIW, these transactions have different FITIDs and are in a single
> import from a single file (and do not duplicate anything pre-existing).
> But, I will get back when I'm not using 3.8. It's sad (but clearly nobody
> here is in control of this!) that Ubuntu is packaging such an old version.
>
> On Fri, Dec 30, 2022 at 7:08 PM Jean L <ripngo at gmail.com> wrote:
>
>> Don't forget to reply to the list.
>>
>> Bear in mind: if you import an OFX file, match its transactions to some
>> of your register transactions, the matched transactions will be silently
>> ignored when you re-import the same OFX. That's by design. So if you match
>> everything then re-import the same OFX, you'll have nothing to match.
>>
>> GC uses the FITID to verify that an OFX transaction was already imported
>> and matched (it compares the FITID value of the transaction in the register
>> to that in the OFX file). When GC imports an OFX transaction, it copies the
>> OFX FITID to the register transaction.
>>
>> But yes, update to a recent version as there were bugs in this mechanism
>> that were recently fixed.
>>
>> J
>> On 12/30/2022 6:03 PM, Simon Roberts wrote:
>>
>> Yes, different FITIDs.
>>
>> I'm now finding that actually silently dumping some of these similar
>> transactions. (But I've also discovered I'm not using the latest
>> version--I'm on 3.8, so I think I should upgrade it and try again.)
>>
>>
>>
>> On Fri, Dec 30, 2022 at 6:57 PM Jean L <ripngo at gmail.com> wrote:
>>
>>> We might need a bit more info to give a useful answer. But just to be
>>> sure:
>>> - The FITID of the transactions in the OFX file are all different,
>>> right? That's an absolute requirement, not just within a single OFX
>>> file, but from ofx to ofx, the FITID is supposed to identify one and
>>> only 1 transaction. If you re-download the same transaction the FITID is
>>> supposed to be the same as the first time you downloaded it. GC uses
>>> this to know which transactions have already been matched when it looks
>>> at an imported OFX file.
>>>
>>> - Is the problem that you have similar transactions in your register
>>> (that you've entered manually) with the same amount and close dates?
>>> It's not impossible that GC is a bit confused because it finds 2 or more
>>> transactions in your register than seem to match 1 or more transactions
>>> in your OFX. In other words, GC does not have a way to figure out which
>>> OFX transaction should be matched with which transaction in your
>>> register if everything is very similar between them.
>>> I must say I've never run into this issue, so I don't know for a fact
>>> what GC does in that case.
>>>
>>> J
>>>
>>> On 12/30/2022 5:36 PM, Simon Roberts wrote:
>>> > With the same caveat about my being new to this...
>>> >
>>> > My transaction records show regular, very similar, transactions. The
>>> same
>>> > institution, identical amounts, usually close but not identical dates.
>>> >
>>> > The importer's matcher is confused by these similar transactions. It
>>> marks
>>> > them in red and refuses to do anything with them directly. I can unmark
>>> > them, and then get them to import, but there are other fields
>>> (specifically
>>> > there is a field "<FITID>" in the OFX file that seems like it should
>>> > distinguish them, even if everything else is identical (I'm assuming
>>> FITID
>>> > is "Financial Transaction ID", but I could be wrong since that's just a
>>> > guess!)
>>> >
>>> > In the interest of being clear, these "conflicting" transactions are
>>> in the
>>> > *same* OFX file, it's not trying to match against something already in
>>> > place.
>>> >
>>> > Can I do anything to get this to behave more helpfully?
>>> >
>>> > Cheers,
>>> > Simon
>>> >
>>> _______________________________________________
>>> gnucash-user mailing list
>>> gnucash-user at gnucash.org
>>> To update your subscription preferences or to unsubscribe:
>>> https://lists.gnucash.org/mailman/listinfo/gnucash-user
>>> -----
>>> Please remember to CC this list on all your replies.
>>> You can do this by using Reply-To-List or Reply-All.
>>>
>>
>>
>> --
>> Simon Roberts
>> (303) 249 3613
>>
>>
>
> --
> Simon Roberts
> (303) 249 3613
>
>

-- 
Simon Roberts
(303) 249 3613