[soln-plugin-ascent]

Uses Alpine Ascent to plot on-the-fly . The following parameters can then be set:

1. nsteps — produce the plots every nsteps time steps:

   int

2. division — the level of linear subdivison to use

   int

3. region — region to visualise, specified as either the entire domain using * or a combination of the geometric shapes specified in Regions:

   * | shape(args, ...)

There are then three components that can be used to build plots. Scenes which define the render, Pipelines that can be used to apply filers and build sequenceces of data manipulations, and Fields which are used to define field expresions.

1. field-{name} — this is an extension to the Ascent library where users define expressions for the fields used. This can either be a scalar or a vector, where the latter is defined by a comma separted list of expressions.

string | string, string (, string)

2. scene-{name} — a scene to plot with Ascent options passed in a dictionary. Each scene needs a field, and the expression for that field must have been set either via a field command or a pipeline. Additionally, one or multiple render-{name} dictionaries must be defined to configure the rendering of the scene. Multiple render dictionaries give multiple views of the same scene.

dict

3. pipeline-{name} — a pipeline of data manipulations that can be used within a scene. The value is a dictionary containing the valid configuration options. Pipeline objects can be stacked together to form a pipeline of filters by making a list of dictionaries. Finally, the q-criterion and vorticity filters require that a field called velocity is defined.

dict | [dict]

Example:

[soln-plugin-ascent]
nsteps = 200
division = 5

field-kenergy = 0.5*rho*(u*u + v*v)
scene-ke = {'render-1': {'image-name': 'ke-{t:.1f}'}, 'field': 'kenergy', 'type': 'pseudocolor'}

field-mom = rho*u, rho*v
pipeline-amom = {'type': 'vector_magnitude', 'field': 'mom', 'output-name': 'mag'}
scene-va = {'type': 'pseudocolor', 'pipeline': 'amom', 'field': 'mag', 'render-1': {'image-width': 128, 'image-name': 'm1-{t:4.2f}'}, 'render-2': {'image-width': 256, 'image-name': 'm2-{t:4.2f}'}}

Note that setting nsteps to be too small can have a significant impact on performance as generating each image has overhead and may require some MPI communication to occur.

This plugin also exposes functionality via a CLI. The following functions are available

1. pyfr ascent render — render an image from a pre-existing mesh and solution file. It must be run with the same number of ranks as partitions in the mesh. By default it will use settings from the first section of the settings file that it is passed. Alternatively, a specific section name can be provided. In both cases all other sections are ignored.

Example:

pyfr ascent render mesh.pyfrm solution.pyfrs settings.ini