compass.pipeline.coordinator.COMPASSCollection

class COMPASSCollection(runtime)

Bases: BaseRunMode

Concrete Strategy for document collection mode

Parameters:

runtime (compass.pipeline.runtime.PipelineRuntime) – The runtime object containing all dependencies, configurations, and settings for the pipeline run. This object should be initialized with the user’s request and any necessary setup (e.g., folder creation, model registry construction) before being passed to the workflow. The workflow will use the runtime to access configurations such as the mode to run in, the tech being processed, model configurations, known sources, and any other relevant settings needed to execute the workflow for the specified mode.

Methods

run(jurisdictions_df)

Run process mode over all requested jurisdictions

async run(jurisdictions_df)

Run process mode over all requested jurisdictions

Parameters:

jurisdictions_df (pandas.DataFrame) – A DataFrame containing information about the jurisdictions to process. This DataFrame should include all necessary information for each jurisdiction, such as its code, full name, and any other relevant metadata needed for processing. The workflow will iterate over each jurisdiction in the DataFrame and execute the collection step for each one, using the information provided in the DataFrame to guide the processing steps.

Returns:

str – A summary message of the collection run, including key information such as the number of jurisdictions processed, documents found, total cost, and output locations. The exact content of the message may vary depending on the results of the processing.