Repository navigation
test: close OPA adapter review coverage gaps - #96
Conversation
Signed-off-by: Srinivasa Dinakar <78673153+dinakarjs@users.noreply.github.com>
|
🟡 Contributor Check: MEDIUM
Automated check by AgenTrust Contributor Check. |
rajnisht7
left a comment
There was a problem hiding this comment.
The changes looks good, this PR adds three focused tests that close the remaining OPA coverage gaps from the earlier review, which are invalid ids elements, challenge/error mapper paths, and omitted result -> not_applicable.
Signed-off-by: Srinivasa Consultants <78673153+dinakarjs@users.noreply.github.com>
|
Refreshed current main at e193283; the merge brings in the current dependency locks and does not change this PR's tests or runtime source. Installed the resulting hash-pinned test lock, then validated on Python 3.12: 151 unittest cases run, 149 passed and 2 skipped. Version, bundled-schema, OTel compatibility, TypeScript-schema parity and conformance-fixture checks pass; diff whitespace checks pass. Published tree matches the tested tree. Hosted checks and fresh independent review remain separate. |
|
@imran-siddique current head e193283 is approved by @rajnisht7, who is listed in this repository's maintainer policy, and all seven hosted checks pass. The branch is current and mergeable. I attempted a normal head-pinned squash merge under the user's instruction to close the work, but the GitHub connector returned 403 Resource not accessible by integration. Please merge through the normal maintainer path; no branch rule or approval bypass is requested. |
Follow up the nonblocking review on #73 with three test cases: reject invalid elements inside the ids list, preserve successful challenge/error mapper outputs, and map an omitted result to not_applicable. The unhashable-mapper correction and its dict/list controls are already on main; this PR preserves them. Production source, schema and contract are unchanged.
Validation on Python 3.12: full unittest suite 151 tests, 2 skipped. The runner inherited OTEL_TRACES_SAMPLER=parentbased_traceidratio with OTEL_TRACES_SAMPLER_ARG=0.01; clearing OTEL_TRACES_SAMPLER for the test process restored the default recording behavior. No tracing environment is modified persistently. The first run with inherited sampling failed three unrelated span-recording checks; these are not claimed as fixed by this PR.
Three isolated mutations were detected: removing ids element validation (4 failures), mapping challenge/error to deny (2 failures), and defaulting missing result to False (1 failure). Source restored and git diff --check passes. These controls verify adapter behavior only, not policy enforcement or OPA service execution.