After r6510, autocombine.py can combine the composition (and pressure, stress)
field. The combined result is stored in an opt-file, not in the cap-file.

The OpenDX scripts (visFull.net and visRegional.net) do not know how to import
the opt-file yet.

>
> Eh Tan <tan2@geodynamics.org> added the comment:
>
>> the output of viscosity is not working properly.
>
> Can you explain more about this?

This comment was just based on what you had shown me the other
day, that you put the composition field in place of
viscosity. Maybe I misinterpreted what you had said.

>
> I agree we need to provide a way to combine and visualize the composition
> field.
> I need to think about how to do it in a least amount of trouble to the
> users.
>
> If the composition field is added to the combined cap files. All user
> scripts
> working on the cap files, regardless whether they are interested in the
> composition field or not, must be modified. This will be a big hassle to
> the users.

Maybe there is a way around this. Perhaps, composition could be added as a
new collumn to the *.velo files. The number of collumns read could be
different depending on if a composition flag is set in the *.cfg files.

Since the flag won't be set in the old files, the autocombine, etc., files
would continue to read the old number of collumns.

>
> I am more inclined to combine the compostion field to a seperated file.
> Then, if
> the users are interested in the composition field, they can add new
> functionality to their scripts. If not, their scripts continue to work.
>
> ----------
> assignedto:  -> tan2
> status: unread -> chatting
>
> _____________________________________________________
> Roundup issue tracker <issue_tracker@geodynamics.org>
> <http://geodynamics.org/roundup/issues/issue99>
> _____________________________________________________
>

> the output of viscosity is not working properly.

Can you explain more about this? 

I agree we need to provide a way to combine and visualize the composition field.
I need to think about how to do it in a least amount of trouble to the users. 

If the composition field is added to the combined cap files. All user scripts
working on the cap files, regardless whether they are interested in the
composition field or not, must be modified. This will be a big hassle to the users.

I am more inclined to combine the compostion field to a seperated file. Then, if
the users are interested in the composition field, they can add new
functionality to their scripts. If not, their scripts continue to work.

I think that the output of composition in Version 2.2 will have to be clarified,
not least because the output of viscosity is not working properly.

If temperature is output in the velo files, perhaps composition should as well.
However, we may need a composition output flag that is written out into the
pid*.cfg file and interpretatble by autocombine.