Pipeline

This module contains the Pipeline class and related components for the EZStitcher pipeline architecture.

For conceptual explanation, see Pipeline. For information about directory structure, see Directory Structure.

Pipeline

class ezstitcher.core.pipeline.Pipeline(steps=None, name=None)

A sequence of processing steps that are executed in order.

For detailed information on pipeline construction, including best practices and different approaches, see Pipeline.

Parameters:

steps (list of Step) – Initial list of steps
name (str) – Human-readable name for the pipeline

add_step(step)

Add a step to the pipeline. See Pipeline for best practices on when to use this method versus providing all steps during initialization.

Parameters:: step (Step) – The step to add
Returns:: Self, for method chaining
Return type:: Pipeline

run(input_dir=None, output_dir=None, well_filter=None, microscope_handler=None, orchestrator=None, positions_file=None)

Execute the pipeline.

This method can either:

Take individual parameters and create a ProcessingContext internally, or
Take a pre-configured ProcessingContext object (when called from PipelineOrchestrator)

The orchestrator parameter is required as it provides access to the microscope handler and other components.

Parameters:

input_dir (str or Path) – Optional input directory override
output_dir (str or Path) – Optional output directory override
well_filter (list) – Optional well filter override
microscope_handler (MicroscopeHandler) – Optional microscope handler override
orchestrator (PipelineOrchestrator) – PipelineOrchestrator instance (required)
positions_file (str or Path) – Optional positions file to use for stitching

Returns:

The results of the pipeline execution

Return type:

dict

input_dir

Get or set the input directory for the pipeline.

Type:: Path or None

output_dir

Get or set the output directory for the pipeline.

Type:: Path or None

ProcessingContext

class ezstitcher.core.pipeline.ProcessingContext(input_dir=None, output_dir=None, well_filter=None, config=None, **kwargs)

Maintains state during pipeline execution.

The ProcessingContext holds input/output directories, well filter, configuration, and results during pipeline execution. It serves as a communication mechanism between steps in a pipeline, allowing each step to access and modify shared state.

For detailed information about how the context is used for communication between steps, see Pipeline Context in the Pipeline documentation.

Parameters:

input_dir (str or Path) – The input directory
output_dir (str or Path) – The output directory
well_filter (list) – Wells to process
config (dict) – Configuration parameters
**kwargs –
Additional context attributes that will be added to the context

input_dir

The input directory for processing.

Type:: Path or None

output_dir

The output directory for processing results.

Type:: Path or None

well_filter

List of wells to process.

Type:: list or None

config

Configuration parameters.

Type:: dict

results

Processing results.

Type:: dict

Step Classes

For documentation on step classes like ImageStitchingStep, see Steps.