Active Save Button on start up

[Robert Fewell]

John,

Drat, I did not see it there, should of looked more closely. I will
investigate and see what I come up with.

Bob

On 22 April 2017 at 15:31, John Ralls <jralls at ceridwen.us> wrote:

>
> On Apr 22, 2017, at 4:26 AM, Robert Fewell <14ubobit at gmail.com> wrote:
>
> Hi,
>
> Over the last few weeks I have noticed that my test file always starts with
> the save button active. If I immediately save and close, then load the file
> again it comes back active.
>
> Created a new empty file, all is well, added one simple transaction, all is
> well, scheduled that transaction to be weekly I get the active save all the
> time.
>
> Looking at the commits in master, this one seems to be likely,
>
> https://github.com/Gnucash/gnucash/commit/3367e191c8072fe4e49b06c1e99d43
> f7fde0304f
>
>
> So the first question is, this should not happen right, if the scrub
> function fixes any thing then the save should save these changes and so it
> should not find any problems on the next load ?
>
> The other thing that I do not understand is when you look at that commit
> and compare it to the source file they are different but I can not find any
> subsequent changes to make it so.
>
>
> Bob,
>
> Thanks for noticing and pursuing this.
>
> The scrub shouldn't get called at all if there aren't any SXes in the
> book, so clearly the template_root check isn't limiting the scrub to only
> SX splits.
> Since it's recurring it would also seem that the "changed" check is
> succeeding when it shouldn't.
>
> As for the commit, it's on maint and required some changes in the merge
> commit [1] to master because of private-kvp.
>
> Regards,
> John Ralls
>
> [1] https://github.com/Gnucash/gnucash/commit/
> baef4029b20a52c6690744d52034ef9283a9383b
>