Related-write atomicity: 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 related writes and their transaction failure contract. Distinguish all-or-nothing execution from a possible partial commit, independent writes and missing wrapper behavior.
The question starts with the invariant: what must remain true if the operation fails? Two writes can be independent, or they can form a single state change that is invalid when only half completes.
A named wrapper is insufficient evidence. This demo requires its transaction and rollback behavior to be supplied. It does not connect to a database, inspect isolation levels, or prove that a real driver implements the stated contract.
This original code and behavior review adapts a review-only rule in the taste-lint repository linked by the source post. Synthetic inputs and labels are ours. Review-only source status does not establish an executed upstream check; this page does not run code or certify an application.

Use the sample to understand the rule, then test your own input for related-write atomicity.

Show the related writes and their transaction failure contract.

Open the decision rules and keep the intended distinction between Atomic, Partial, Independent, Unclear. 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 Unclear before using a result. The label is a decision about the supplied information, not an executed action.
| Original input | Observed label | Leading probability |
|---|---|---|
| Project and owner membership | Partial | 94.0% |
| Verified transaction | Atomic | 100% |
| Unrelated records | Independent | 64.0% |
| Opaque lock wrapper | Unclear | 100% |
| Lock without rollback | Partial | 96.0% |
| Two services | Unclear | 99.0% |
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.
Not by itself. A lock can prevent concurrent access while still allowing each write to commit separately. Show the rollback contract before selecting Atomic.
No. The input must name the invariant requiring joint completion. Unrelated records can remain valid when updated separately.
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.