Python Bindings Known Issues

[imikejackson]

Actions need to be documented

@imikejackson

• [x] Completed There is currently NO documentation for each of the actions.

Filter Parameters Documentation

@imikejackson

• [ ] Completed

Each Parameter should have a screen shot from the User Interface to help the user identify the parameter

IDE Filter Documentation

@imikejackson

• [ ] Completed

Update the .pyi files to include first paragraph from filters documentation

[imikejackson]

• [x] https://github.com/BlueQuartzSoftware/simplnx/pull/925 Hopefully included in this PR

• Preflight is tough. Need examples and thorough docs Preflight Actions need explanation

• Generate Python Skeleton Needs some more docs to fully explain the options

• DataArray API Tuple/Comp dims functions tDims?

• Filter Writing: when to use certain parameters in what situation

• Update the docs more with this kind of information

[imikejackson]

• [x] COMPLETED @JDuffeyBQ

Pipeline Objects

import simplnx as nx
import nxutility
import numpy as np

# Create the DataStructure instance
data_structure = nx.DataStructure()
# Read the pipeline file
pipeline = nx.Pipeline().from_file( 'lesson_2.d3dpipeline')

# Append a new filter onto the end of the pipeline
create_geom_args : dict = {
        "array_handling": 0,
        "cell_attribute_matrix_name": "Cell Data",
        "dimensions": [ 100, 100, 100 ],
        "geometry_name": nx.DataPath("Output Geometry"),
        "geometry_type": 0,
        "length_unit_type": 7,
        "origin": [ 0.0, 0.0, 0.0 ],
        "spacing": [ 1.0, 1.0, 1.0 ],
        "warnings_as_errors": False
}

# Execute the modified pipeline
pipeline.append(nx.CreateGeometryFilter(), create_geom_args)
result = pipeline.execute(data_structure)
nxutility.check_pipeline_result(result=result)

pipeline.to_file( "Modified Pipeline", "Pipelines/lesson_2_modified_pipeline.d3dpipeline")

The last line in this python will fail because not all of the arguments are present in the create_geom_args dictionary. We should just write it out using the default arguments at that point

[imikejackson]

• [x] Completed

Pipeline Filter

There should be easier access to the "human_name" and the "name" attributes for the Pipeline filter, i.e., a developer should not have to jump through pipeline_filter.get_filter().human_name()

pipeline_filter.human_name()
pipeine_filter.name()

[imikejackson]

• [x] Completed

C++ Filter Naming

All filters should end with "Filter"

[imikejackson]

PreflightUpdatedValues

• [x] Completed

In the Python Filter, one cannot return preflight updated values.

The following does not work, throwing an error about supported types.

return nx.IFilter.PreflightResult(output_actions=output_actions, errors=None, warnings=None, preflight_values=preflight_values)

Progress Messages:

• [x] Completed

This does not work because ProgressMessage is not exposed

    message_handler(nx.IFilter.Message(nx.IFilter.Message.Type.Progress, f'Progress Message:', 50))

DataStructure API Additions

• [x] Completed

path_exists(nx.DataPath)
path_exists(str)

nx.filter.XXX
nx.parameter.XXX
nx.geometry.XXX

AttributeMatrix API Additions

• [x] Completed

tuple_shape()
component_shape()
__size__() 

[imikejackson]

Breakdown the API into more namespaces

• [ ] Completed

This is Non-Trivial. Putting a pin in it for now. The real problem is generating the .pyi files with the namespace