command line QIF import

[David Carlson]

I am not a developer, but I can point out that the QIF importer works in
several discrete steps where (roughly) it

1. identifies the 'source' account type and name and tries to associate it
to an existing account in the data file.  If there is more than one
'source' account in the QIF file it repeats for each.

2.  It presents it's results to the user for verification and (if
necessary) correction.

3.  Identifies all of the 'destination' accounts or categories and tries to
match them to existing accounts.

4.  It presents it's results to the user for verification and (if
necessary) correction.

5.  It parses each transaction and compares it to existing transactions
between the particular accounts previously identified.

6. It prepares lists of incoming transactions with suggested matches to
existing transactions where found.  These lists allow user editing to
accept or correct suggested matches.  This is where it is more efficient to
import only one account at a time so the 'duplicates' can be found here.

There are many checks during the import to test the text at each step and
prevent possible crashes and gross errors during matching and acceptance.

As an experienced user I would actually prefer that the developers stay out
of this discussion as they are very busy working 'under the hood' right now
toward getting the release 2.8 series into shape.  That release should get
rid of the deprecated software that is now threatening the continued
inclusion of GnuCash in (and compatibility with) upcoming Linux packages.

David C

On Sun, Oct 15, 2017 at 6:05 PM, Adam Jensen <hanzer at riseup.net> wrote:

> On 10/15/2017 04:49 PM, Nelson wrote:
> > Thank you for your answer David.
> >
> > I am aware that QIF is a text file and I can open it in text editor.
> >
> > I have a 235000 lines QIF file created by quicken. Most of the
> transactions
> > are being imported properly but there are thousands failing as well. I
> have
> > no way to identify those failed transactions other than going line by
> line
> > in each account. What shall I "edit in quicken first" to insure a
> successful
> > import?
>
> Hi. I am very new to the list [1] and *very* new to GnuCash [2].
>
> Strangely, Nelson, I might get what you are asking far more than the
> other respondents seem to. I hope you can endure this problem, at least
> while it is being explored and discussed; rational, pragmatic people are
> vital to technical projects. ...Sometimes volunteer-based communities
> get a little cult-like in their public interface...
>
> So, I am a mature engineer [3] with some experience in
> instrumentation-sourced data wrangling (as part of large data
> organization and analysis systems); maybe I can try to steer the topic
> of file conversion/import towards the "hackers" on the list - and those
> otherwise interested in engaging such a conversation. To that end, could
> some of the software developers and maintainers chime into a technical
> discussion of the data structure issues involved in building a
> translation utility? It might be fruitful if not at least interesting :)
>
> Anyway, a belated Greetings and Salutations to you all :)
>
>
> [1]: Subscribed for about two days.
> [2]: I've only just started the semi-serious exploration, assessment and
> evaulation of the GnuCash software (and its representation of the
> historical bookkeeping/accounting processes and their
> traditional/societal basis) about three days ago - installation and one
> launch of the application on Ubuntu, one chapter of the Guide read and
> the introductory chapter of a financial Accounting book mused at.
> [3]: LinkedIn, Google::hanzer
> _______________________________________________
> gnucash-user mailing list
> gnucash-user at gnucash.org
> 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.
>