[GNC] Example of multi-split feature of CSV importer?

[Geert Janssens]

Op donderdag 28 maart 2019 20:18:57 CET schreef David Carlson:
> I would like to add that the generic importer shows the memos of matched
> existing transactions rather than the descriptions on the page where
> incoming transactions are matched to existing since it better fits the
> notion of matching to the base 'split' account.  I presume that the
> improved CSV importer follows that behavior.
> 
That part of the importer hasn't been touched. So it should still be as it was in GnuCash 2.6.

Regards,

Geert

> David Carlson
> 
> On Mon, Mar 25, 2019 at 8:35 PM Patrick <plafratt at gmail.com> wrote:
> > David, Geert,
> > 
> > Thank you for this helpful guidance. I hope to try this soon.
> > 
> > Regards,
> > Patrick
> > 
> > On Mon, Mar 25, 2019 at 5:18 AM Geert Janssens <geert.gnucash at kobaltwit.be
> > 
> > wrote:
> > > Op zondag 24 maart 2019 02:47:37 CET schreef David Cousens:
> > > > Patrick,
> > > > 
> > > > 
> > > > 
> > > > When the Transaction is opened with the split button, the Description
> > > 
> > > field
> > > 
> > > > name should change to Memo. There are other changes in the headings as
> > > > 
> > > > described in the documentation (
> > > > 
> > > > https://www.gnucash.org/docs/v3/C/gnucash-help/trans-multi-enter.html
> > 
> > ).
> > 
> > > If
> > > 
> > > > my understanding is correct, each split of a transaction has its own
> > 
> > Memo
> > 
> > > > field while the Description field is attached to the transaction which
> > > > 
> > > > consists of 2 or more splits. The same description should also come up
> > 
> > if
> > 
> > > > you open the transaction in another register that there is a split to.
> > > 
> > > You
> > > 
> > > > can enter different information in the Memo field if you wish.
> > > > 
> > > > 
> > > > 
> > > > I think the importer matches on the Description field for the
> > > 
> > > transaction by
> > > 
> > > > tokenizing its content and matching to the tokenized forms of the
> > > > 
> > > > description field of existing transactions which is weighted along
> > > > with
> > > 
> > > the
> > > 
> > > > matching of amounts, dates, etc. I don't think the Memo field is
> > 
> > involved
> > 
> > > > in matching AFAIK but I haven't checked the code out to see. Geert may
> > 
> > be
> > 
> > > > able to comment on that as he is more familiar with the code than I
> > > > am.
> > > > 
> > > > 
> > > > 
> > > > I think you can specify the Memo field in the multisplit format as
> > > > well
> > > 
> > > as
> > > 
> > > > the Description field. I am away from home at the moment so I can't
> > 
> > check
> > 
> > > > for sure.
> > > > 
> > > > 
> > > > 
> > > > David
> > > 
> > > That explains it very well indeed.
> > > 
> > > 
> > > 
> > > In short: transactions have a description, splits have a memo. If you
> > 
> > want
> > 
> > > each individual split to have some descriptive information, you can add
> > > this in the memo field (which is also supported by the importer). If you
> > > want the whole transaction to have some descriptive information, you can
> > > add this in the transaction's description field.
> > > 
> > > 
> > > 
> > > And as David says, the csv importer will use the description field (and