[Gmsh] Save options file upon exit automatically?

[Christophe Geuzaine]

On 07 Aug 2011, at 04:49, Geordie McBain wrote:

> 2011/8/6 Christophe Geuzaine <cgeuzaine at ulg.ac.be>:
>> 
>> Hi guys,
>> 
>> We should indeed clarify/enhance the "save options" behavior. At the moment we can choose between three "levels" when saving options:
>> 
>> - "session" (gets saved by default in .gmshrc on exit)
>> - "option" (can get saved in .gmsh-options on exit, or is saved there manually when choosing "Save Options->As default")
>> - "full" (used when explicitly saving the options)
>> 
>> Maybe we could provide a way to save the options for the current file each time we exit? This could make Gmsh be completely "persistent" per model. Or do you have other suggestions?
>> 
>> Let me know what you think.
> 
> Personally I'm quite happy with the current behaviour. The new
> suggested option sounds O.K. too, though I can't imagine using it
> myself, and can see it being annoying if ever activated accidentally.
> 
> I think it might be a good idea to keep the standard default option
> settings as the default behaviour, because when newcomers are learning
> a program they can sometimes get themselves and the state of the
> program into a tangle from which the best step is to quit and restart.
> There is the "Restore default options" recourse , but a beginner
> mighn't not find that without help, as happened the other day
> <http://www.geuz.org/pipermail/gmsh/2011/006588.html>.
> 
> I guess considering the worst possibility in each case:
> 
> 1. If persistence is the default, a clumsy user can get the state in a mess.
> 
> 2. If tabula rasa is the default, a clumsy user might spend hours
> settting all their favourite options and then lose them.
> 
> I'm not sure, but I think the second user would be more likely to
> recover and find the remedy, since if they've already found how to set
> options, they've had a good chance of finding the option which
> controls the saving of state.
> 
> But Gmsh is already cleverer than this simple dichotomy and realizes
> that different answers to it are appropriate to different variables,
> and even has three answers rather than two.  This is good.
> 
> Anyway (sorry, I know this is already too long), the only concrete
> suggestions I'd make are:
> 
> * Expand the little paragraph beginning "Appendix B Options"
> <http://geuz.org/gmsh/doc/texinfo/gmsh.html#Options>, as currently it
> only refers to two possibilities rather than three, e.g. by cutting
> and pasting from your e-mail.
> 

Excellent idea: done.

> * Mention in "3 Running Gmsh on Your System"
> <http://geuz.org/gmsh/doc/texinfo/gmsh.html#Running-Gmsh-on-your-system>
> that the state is affected by saved options, refer to Appendix B, and
> succinctly tell how to get the system back to the pristine state.
> 

Also done.

Thanks for the feedback,

Christophe

> Geordie McBain

-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science 
http://www.montefiore.ulg.ac.be/~geuzaine