Parallel

The parallel module provides background task execution capabilities for long-running operations in the Mosaic GUI, ensuring responsive user interface during compute-intensive tasks. It supports both single task submission and batch processing with concurrency limits.

BackgroundTaskManager

The core class managing task queuing, execution, and result handling. Uses Qt signals for inter-component communication.

BackgroundTaskManager

Signals

The BackgroundTaskManager emits the following Qt signals:

task_started(task_id: str, task_name: str)

Emitted when a task begins execution.

task_completed(task_id: str, task_name: str, result: object)

Emitted when a task completes successfully. The result contains the function’s return value.

task_failed(task_id: str, task_name: str, error: str)

Emitted when a task fails with an exception.

task_warning(task_id: str, task_name: str, warning: str)

Emitted when a task completes but produced warnings during execution.

running_tasks(count: int)

Emitted periodically with the current number of running tasks.

Convenience Functions

Module-level functions that provide easy access to the singleton BackgroundTaskManager instance.

submit_task(name, func[, callback])
submit_task_batch(tasks[, max_concurrent, ...])

Usage Example

from mosaic.parallel import submit_task, submit_task_batch

# Single task
def expensive_computation(data):
    # Process data...
    return result

task_id = submit_task(
    "compute_mesh",
    expensive_computation,
    callback=on_result,
    data=my_data
)

# Batch of tasks with concurrency limit
tasks = [
    {"name": "process_1", "func": func, "args": (data1,)},
    {"name": "process_2", "func": func, "args": (data2,)},
    {"name": "process_3", "func": func, "args": (data3,)},
]
batch_id = submit_task_batch(tasks, max_concurrent=2)