Option for saving workspace

[Pierre de Villemereuil]

Hi both!

Thanks for your responses! I got the hint, here's the wish report:
https://bugs.kde.org/show_bug.cgi?id=383571

As stated in the report, I think the current behaviour makes sense as a default, as RKWard doesn't know how valuable the workspace is to the user and thus needs to ask. 

Of course, saving the workspace to a default location and restoring it, as you offer Thomas, would also be sensible. Possibly even better, as it would certainly make sense for a "session restore". The user can still discard the restored workspace if he/she doesn't care about it.

Also, I'm realising this doesn't only pertain to activities, but also more generally to session exit/restore when the session is set to restore the open applications. In that case as well, RKWard is stopping the disconnection to ask the user what to do with the workspace.

Thanks for your interest in this!

Cheers,
Pierre

Le mardi 15 août 2017, 20:05:31 CEST Thomas Friedrichsmeier a écrit :
> Hi!
> 
> On Tue, 15 Aug 2017 15:40:37 +0200
> meik michalke <meik.michalke at uni-duesseldorf.de> wrote:
> > RKWard replaces q() with a custom function, which calls
> > 
> >   .rk.do.plain.call("quit")
> > 
> > since the above call currently ignores all arguments given to q(),
> > calling it with 'save="no"' has no effect, which in turn invalidates
> > workarounds like changing the default formals of q(). i also didn't
> > manage to start RKWard with R interpreting the "--no-save" option
> > (e.g., by using an alias).
> > 
> > i think the best approach here would be to actually evaluate q()'s
> > "save" argument.
> > 
> > @thomas: does RKWard's closing dialog already support further
> > arguments, so that replacing the R functions q()/quit() would be
> > sufficient?
> 
> That should be possible. I'll try to remember looking into that (hint:
> reporting a wish might help: https://rkward.kde.org/bugs).
> 
> I'm a bit unsure, what would be the expected default behavior, though,
> and whether we need anything special to support it for the activities
> usecase. The general expectation would be that RKWard restores the
> running session as close as possible, right? So what we'd need to do is
> not simply quit without saving, but save to some standard location,
> without asking, and restore that on the next run?
> 
> Regards
> Thomas