Cancellation handling: six recorded examples
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
Show the cancellation handler and relevant exception behavior. Compare cleanup followed by propagation with a handler that suppresses cancellation or leaves the path unclear.
Logging a cancellation is different from propagating it. The review follows the control flow after the catch: cleanup and rethrow, continued work, or a normal return.
Runtime exception hierarchies matter. A generic catch clause may or may not catch cancellation; supply that fact instead of expecting the model to infer an unstated language version.
The source linter checks swallowed cancellation. This example flags the shown behavior for review; it does not claim that every deliberately contained cancellation is a bug in every framework.
This is an independently authored text demonstration inspired by the linked X discussion. Its rules and synthetic inputs are ours. It does not reproduce the author’s application, verify reported outcomes, or perform an external action.

Use the sample to understand the rule, then test your own input for cancellation handling.

Show the cancellation handler and relevant exception behavior.

Open the decision rules and keep the intended distinction between Propagated, Suppressed, No path shown, Review. Edit the choices only when testing your own rule.

Use a sample first, then try your own input. Compare the selected label with the other returned probabilities.

Review Review before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Cleanup then rethrow | Propagated | 97.0% |
| Logged then continued | Suppressed | 98.0% |
| Success fallback | Suppressed | 97.0% |
| Ordinary validation recovery | No path shown | 78.0% |
| Unknown catch-all semantics | Review | 71.0% |
| Missing branch | Review | 71.0% |
These are six original synthetic inputs, tested once per request revision. The percentage is the leading option probability, not measured accuracy. Authored expectations and any disagreements remain in the downloadable evidence. 6 of 6 results matched the authored fixture expectations.Verified
The downloads contain six original inputs, their expected labels and one recorded API observation per input revision. Disagreements stay visible.
The single-request SDK runner uses your private TYPESAFE_API_KEY. The offline consumer reads saved observations, checks request hashes and prints a pending output. It does not call the model or change any external system.
# Offline: inspect the downloaded saved observations. No API key is needed.
python preview_choice_workflow.py samples.json saved-api-results.json
# Optional live request: install the SDK and set TYPESAFE_API_KEY privately.
npm install @typesafe-ai/sdk@0.6.0
node run-choice-sample.mjs samples.json case-1
# The live response prints to the terminal; it does not update saved-api-results.json.The input contract and these limits define what the result means.
No. Cleanup followed by propagation is one of the explicit examples. The question is what happens to the cancellation signal afterward.
Yes, framework or application contracts can define containment. This demo reports the supplied path; a maintainer must judge whether that behavior is intended.
It describes the distribution over the supplied choices for this request. These six synthetic examples are functional tests with authored expectations, not a production accuracy benchmark.
No. The demo displays a label and distribution. The downloadable consumer produces a pending preview packet. Application code and a separate authorization policy would control any real action.
Edit the example above to see how a changed fact affects the decision, or open the Playground for another task.