r22465 - gnucash/trunk/src/engine - Allow SchedXaction to safely handle having g_object_property_set called on

[John Ralls]

On Oct 28, 2012, at 4:44 PM, Phil Longstaff <phil.longstaff at yahoo.ca> wrote:

> Does the caller have any indication that nothing was done?

Well, it gets an empty GValue. It's pretty common for SXes to have empty (and therefore invalid) end-dates and last-dates, so it's not an error, critical or otherwise.

I think that my commit message is a bit misleading, though, since it implies that the error is from the SQL backend loading the date. It's not, it happens when the SQL backend is saving the SX and is getting the date -- as should be obvious since the change is in gnc_schedxaction_get_property().

Regards,
John Ralls


> From: John Ralls <jralls at ceridwen.us>
> To: Phil Longstaff <phil.longstaff at yahoo.ca> 
> Cc: "gnucash-devel at gnucash.org devel" <gnucash-devel at gnucash.org> 
> Sent: Sunday, October 28, 2012 7:02:53 PM
> Subject: Re: r22465 - gnucash/trunk/src/engine - Allow SchedXaction to safely handle having g_object_property_set called on
> 
> 
> On Oct 28, 2012, at 11:51 AM, Phil Longstaff <phil.longstaff at yahoo.ca> wrote:
> 
> > How about an else clause in case the value is invalid?
> 
> If it's invalid, it does nothing. What would you like it to do?
> 
> Regards,
> John Ralls
> 
> > 
> > 
> > ________________________________
> > From: John Ralls <jralls at code.gnucash.org>
> > To: gnucash-patches at gnucash.org 
> > Sent: Saturday, October 27, 2012 7:30:57 PM
> > Subject: r22465 - gnucash/trunk/src/engine - Allow SchedXaction to safely handle having g_object_property_set called on
> > 
> > Author: jralls
> > Date: 2012-10-27 19:30:56 -0400 (Sat, 27 Oct 2012)
> > New Revision: 22465
> > Trac: http://svn.gnucash.org/trac/changeset/22465
> > 
> > Modified:
> >    gnucash/trunk/src/engine/SchedXaction.c
> > Log:
> > Allow SchedXaction to safely handle having g_object_property_set called on
> > end_date and last_date with invalid GDates.
> > 
> > 
> > This happens routinely when those values are empty and are restored from
> > SQL. Thanks to all warnings being fatal when testing, this caused the dbi
> > tests to fail.
> > 
> > [BP]
> > 
> > _______________________________________________
> > gnucash-patches mailing list
> > gnucash-patches at gnucash.org
> > https://lists.gnucash.org/mailman/listinfo/gnucash-patches
> > _______________________________________________
> > gnucash-devel mailing list
> > gnucash-devel at gnucash.org
> > https://lists.gnucash.org/mailman/listinfo/gnucash-devel
> 
> 
>