Detailed description

GUI applications that want to perform a long-running task in response to user interactions (for example, running a time-consuming calculation, or submitting a complex search query to a remote database) face two major issues:

• If the task is executed directly on the main thread, it blocks the GUI event loop, making the entire application appear unresponsive to the user.

• It’s not generally safe to update a GUI directly from a worker thread, so a task run on a worker thread needs to find a way to safely communicate events back to the main GUI thread.

For TraitsUI-based applications, Traits Futures provides a solution to these issues, similar in principle to the Python standard library concurrent.futures package. Tasks are submitted to an executor, and on task submission the executor immediately returns a “future” object. That “future” object has observable attributes (“traits”) representing the application’s view of the state of the background task. Rather than waiting on the future’s state, an interested observer can listen to updates to those traits and update the GUI state as necessary when changes occur. The Traits Futures machinery ensures that updates to the future’s traits always occur on the main thread, freeing observers from thread-safety concerns.

For further information, see the documentation pages at https://docs.enthought.com/traits-futures/.