Also it would be interesting to have a way to include all the default observables (currently with observables="all" in addProcess tag) and also use the observables included in the processesFile as there could be observables defined in the file that are not included with the “all” option for example those who reference volumes in the geometry e.g. gasVolumeEDep.
The problem is that the list of observables may become very large. The idea behind xxVolumeEDep observables is that we may replace xx by any volume on the geometry. Then, if my geometry integrates many volumes I will end up with the same number of observables. This will happen in similar way with observables connected with particle names, imagine if we include ions, etc.
Finally I believe this kind of custom observable should not be a default. Since we will end up I think with 100’s of observables.
Finally, I think the user should take care to read the documentation of the process and find out which observables are available. The fact of using observables=“all” should be used for testing and debugging, and we should chose the most relevant observables for tidyness and efficiency of a particular analysis.