FA-74330 / Feature flag rollout bucketing / Member archive
Per-context evaluation memo: Memo ignores the flag version · case 05
After a flag is edited, clients keep serving results computed from the old version.
Case contract
calls are [flag, context dict of string attributes, flag version]. The toy evaluation is (sum of code points of flag + key + country) + version) % 3. Results are memoised by (flag, context contents independent of attribute order, version). Return [value, hit|miss] per call.
Why this case matters
Evaluation caches in SDKs must key on every input that can change the result.
One recorded failure
Sample boundary fixtureThis sample comes from the broken implementation of a controlled reproducer.
| Boundary fixture | Actual | Expected | Outcome |
|---|---|---|---|
| version bump misses the memo | [[1, "miss"], [1, "hit"]] | [[1, "miss"], [2, "miss"]] | Failed |
MEMBER ARCHIVE
The complete case is available to members.
This record includes three runnable implementations, regression fixtures, execution results, and source hashes.
Member access is invitation-based. Sign in with your invited account to inspect the sources.
Sign in to the archive ↗