I was just going to fix it as part of my test pull request.

It really does not matter to me what the next steps are.  It is pretty
trivial.   I you think I need practice making a PR, and then fixing it with
a pull request.  Then we can do that as well.

I also have no problem with you just fixing it.

Stefan

On Sun, Jan 25, 2026 at 12:08 PM John Ralls <[email protected]> wrote:

> Stefan,
>
> Good catch. Obviously a copy-paste oversight and nearly 20 years old. I
> don’t see any access to idata via the property, only through the explicit
> setter and getter, so no real impact. That would explain why nobody noticed.
>
> Would you like to submit a PR or shall I fix it?
>
> Regards,
> John Ralls
>
> > On Jan 25, 2026, at 08:46, Stefan Koch <[email protected]>
> wrote:
> >
> > While trying to learn some of the systems in the source code through
> writing unit tests, I think I found a bug in the way the qofinstance
> properties are setup.  (I don't see much usage of this so it may not be an
> issue in practice.)
> >
> > I think the following change fixes the issue:
> >
> > modified   libgnucash/engine/qofinstance.cpp
> > @@ -244,7 +244,7 @@ static void qof_instance_class_init(QofInstanceClass
> *klass)
> >
> >      g_object_class_install_property
> >      (object_class,
> > -     PROP_EDITLEVEL,
> > +     PROP_IDATA,
> >       g_param_spec_uint ("idata",
> >                          "Object IData",
> >                          "Per instance backend private data.",
> >
> > Since I am new, I would like to confirm before going on.
> >
> > Stefan
> > _______________________________________________
> > gnucash-devel mailing list
> > [email protected]
> > https://lists.gnucash.org/mailman/listinfo/gnucash-devel
>
>
_______________________________________________
gnucash-devel mailing list
[email protected]
https://lists.gnucash.org/mailman/listinfo/gnucash-devel

Reply via email to