[yt-users] new feature in EnzoSimulation class

[Jean-Claude Passy]

  Thanks, Britton.

Best,

Jean-Claude Passy

On 20/08/10 15:38, Britton Smith wrote:
> Greetings,
>
> I've just added a new feature to the EnzoSimulation class that a 
> number of people have been asking for.  If you don't know what 
> EnzoSimulation does, see here:
> http://yt.enzotools.org/doc/extensions/analyzing_an_entire_simulation.html
>
> For people who have been setting Enzo to output every constant number 
> of cycles, rather than a constant amount of time, the EnzoSimulation 
> class did not work, as there is no way to know from the parameter file 
> exactly how many datadumps will get made during the simulation and at 
> what times.  Also, if you do collapse simulations where you are 
> outputting data every time a new AMR level is reached, it doesn't work 
> for you either.  To fix that, I've added a new keyword, 
> get_data_by_force, that if set to True, will search the working 
> directory for all directories that match the data dir name 
> convention.  It will then open them all and get times and redshifts by 
> hand.  It takes slightly longer, but only a second or two compared to 
> previously less than a second.
>
> So to use this, you simply need to do, for example:
> import yt.extensions.enzo_simulation as ES
> my_run = ES.EnzoSimulation(parameter_file, get_data_by_force=True, 
> initial_time=0)
>
> And my_run.allOutputs will have your list of datasets.
>
> Britton