Release/0.12.2#52
Merged
Merged
Conversation
Bug-fix release layered on top of 0.12.1. No wire-format change; both fixes are client-side only. * BUG #4 -- check_workflow_budget() now sends a fresh uuidv7 as the "execution_id" field on every /check call instead of reusing workflow_id. The server's gate_reserve_v3 overwrites the field on response anyway, but a client-side placeholder that collides across calls confuses the v3 reservation binding on /track when Transport.track_single() reaches the backend and the field is stale -- exact symptom is 503 RESERVATION_NOT_FOUND per CLAUDE.md section 29. * BUG #5 -- new nullrun.runtime._GATE_CACHE (5s TTL, keyed on (workflow_id, chain_id, model)) collapses consecutive /gate calls from inside `with chain(...)` to a single roundtrip, avoiding 100 /gate calls per 100-step agent loop. Single-shot (Hard mode) callers MUST bypass the cache -- Hard mode's binary allow -> block semantics would let a stale "allow" leak a budget-exhausted call through. Opt-out via NULLRUN_GATE_CACHE_DISABLE=1 for callers that want the legacy always-roundtrip behaviour (used by live smoke tests per docs/runbooks/budget-blue-green-smoke.sh). Tests: 158 new lines in tests/test_v3_wire_contract.py covering per-call execution_id uniqueness, uuidv7 format validation, and the new cache data-structure invariants + opt-out cases. Bumps __version__ + pyproject.toml to 0.12.2.
maltsev-dev
force-pushed
the
release/0.12.2
branch
from
afa283f to
9fe47c4
Compare
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
`ruff check src/` flagged that the BUG #4 line `from nullrun.uuid7 import uuid7_str # CLAUDE.md §24` landed mid-way through the first-party import block (between `nullrun.context` and `nullrun.observability`), breaking I001 import sort. Moved to the bottom of the first-party block (alphabetic order — `uuid7` sorts after `transport`). Also lets ruff auto-fix two cosmetic cleanups in tests/test_v3_wire_contract.py: * sort `_V3_ERROR_CODE_MAP` alphabetic in the existing transport import group (was below `_parse_v3_error_envelope`); * drop a stray blank-line gap between two top-level `from nullrun.transport import (...)` groups. `ruff check src/ tests/` after the fix: 8 pre-existing I001 findings remain in unrelated test files (predicate, test_circuit_breaker_branches.py, test_framework_patches.py, etc.) — out of scope for this PR. Scope above matches the CI step (`ruff check src/`). No behavioural change.
12 tasks
maltsev-dev
added a commit
that referenced
this pull request
Jul 5, 2026
…e on decisions (#53) * fix(sdk): drift.md 2026-07-04 - wire idempotency_key + preserve status code + fail-CLOSED honesty Three SDK-side fixes for drift.md (2026-07-04) P1 items + open Q4. The remaining drift items (P0-1, P0-2, P0-3, P0-4, P1-3, P1-4) are docs-only - SDK code is correct, SDK_README.md is wrong. Those need a README rewrite, not a code fix. F1 (drift.md P1-5 + open Q4): wire idempotency_key on /track v3 single-event - new contextvar get_server_minted_idempotency_key + symmetric set/reset/clear - _capture_server_minted_execution_id now also reads response["operation_id"] (which equals the /check idempotency_key, runtime.py:1260) - _enrich_event stamps it onto the wire_event for llm_call - _build_v3_track_payload propagates onto the v3 /track payload (with contextvar fallback for tests / direct callers) - why: without this, transport-level retry on the SAME event either re-runs CONSUME_SCRIPT (-> 503 RESERVATION_NOT_FOUND since reservation key was DEL-ed after first consume per CLAUDE.md sec 25) or double-bills F2 (drift.md P1-1): HTTP status_code on every decision exception - NullRunBlockedException / NullRunBudgetError / NullRunChainError / NullRunWorkflowInactiveError / NullRunConsumeOverbudgetError accept status_code parameter - _parse_v3_error_envelope populates status_code from response.status_code for every branch (402 budget, 403 workflow/chain cross-org, 422 CONSUME_OVERBUDGET, 503 RATE_LIMIT_REDIS_UNAVAILABLE, ...) - why: FastAPI exception handlers reading exc.status_code previously got None / 500 for budget blocks (the backend's 402 was lost in the NullRunBudgetError -> NullRunBlockedException constructor chain) F3 (drift.md P1-2): fail-CLOSED/OPEN honesty in module-top docstring - runtime.py docstring table now distinguishes SDK-side transport failure (network/5xx/breaker open -> fail-OPEN on /check path) from wire 4xx/5xx that names an enforcement failure (BUDGET_REDIS_UNAVAILABLE -> 402 fail-CLOSED; RATE_LIMIT_REDIS_UNAVAILABLE -> 503 fail-CLOSED) - why: SDK_README claim "Fail-OPEN na infrastructure failures" was half-wrong - conflated two different failure modes. The README fix belongs in the docs rewrite (out of scope here). Tests: tests/test_drift_fixes_2026_07_04.py - 15 tests, all pass. - F1: 5 tests pinning contextvar lifecycle + payload shape - F2: 8 tests pinning status_code on every decision exception class - F3: 2 tests pinning fail-CLOSED on Redis-unavailable wire responses Regression: 140+70 passed in targeted critical-path suites (test_v3_server_minted, test_error_envelope, test_handle, test_protect, test_capabilities, test_drift_fixes_2026_07_04, test_runtime_branches, test_transport_branches, test_integration_contract, test_high_reliability_fixes). No regression. Per scripts-commit-no-push rule: commit locally, NOT push. * release(0.13.0): drift-fixes — idempotency_key on /track + status_code on decisions + patch coverage Three SDK-side fixes per docs/drift.md (2026-07-04) P1 items + open Q4. The remaining drift items (P0-1 / P0-2 / P0-3 / P0-4 / P1-3 / P1-4) are README-only — SDK code is correct, SDK_README.md is wrong. Those go in a separate README rewrite PR; do not block this release. Plus a 4th fix that was missing on 0.12.2: this commit also closes the codecov/patch-coverage gap that dragged PR #52 below the 70% floor. 1. Idempotency-key propagation to /track v3 single-event (P1-5 + Q4) * `_capture_server_minted_execution_id` now also reads `response["operation_id"]` (which equals the /check `idempotency_key`, runtime.py:1260). * `_enrich_event` stamps the value onto `wire_event` for `llm_call`. * `_build_v3_track_payload` propagates it onto the v3 /track body with a contextvar fallback for tests + direct callers. Why: without this, transport-level retry on the same event either (a) re-runs CONSUME_SCRIPT -> 503 RESERVATION_NOT_FOUND since the reservation key was DEL-ed after the first consume per CLAUDE.md §25, or (b) double-bills the underlying budget. 2. status_code preserved on every decision exception (P1-1) * NullRunBlockedException / NullRunBudgetError / NullRunChainError / NullRunWorkflowInactiveError / NullRunConsumeOverbudgetError now accept status_code: int | None = None. * _parse_v3_error_envelope populates it from response.status_code for every branch: 402 budget, 403 workflow/chain cross-org, 422 CONSUME_OVERBUDGET, 503 RATE_LIMIT_REDIS_UNAVAILABLE, etc. Why: FastAPI exception handlers reading `exc.status_code` previously got None / 500 for budget blocks — the backend's 402 was lost in the NullRunBudgetError -> NullRunBlockedException constructor chain. 3. fail-CLOSED / fail-OPEN honesty in the runtime.py module docstring (P1-2) Distinguishes SDK-side transport failure (network / 5xx / breaker open -> fail-OPEN on the /check path) from wire 4xx/5xx that names an enforcement failure (BUDGET_REDIS_UNAVAILABLE -> 402 fail-CLOSED; RATE_LIMIT_REDIS_UNAVAILABLE -> 503 fail-CLOSED). The README had conflated the two with a single "Fail-OPEN on infra failures" claim; README rewrite is tracked separately under drift.md P0-1. 4. Patch-coverage gap (regression fix from 0.12.2) * tests/test_v3_wire_contract.py::TestGateCacheRuntimeFlow — 3 runtime-level chain-mode cache tests that drive NullRunRuntime.check_workflow_budget inside `with workflow(...) + with chain(...)`. Covers runtime.py:1287-1310 (cache_enabled predicate, cache key, cache hit/miss branches, NULLRUN_GATE_CACHE_DISABLE=1 bypass). * These cover the exact range that dragged PR #52 codecov/patch below 70%. Files in this commit * src/nullrun/__version__.py — bumped 0.12.2 -> 0.13.0 + 0.13.0 release block in the docstring. * pyproject.toml — version = "0.13.0" + drift-release comment (drift-prevention, same pattern as #50). * CHANGELOG.md — new [0.13.0] - 2026-07-04 section preceding [0.12.2]. * docs/drift.md — NEW audit document (the file referenced by the fix(sdk) commit message). * tests/test_v3_wire_contract.py — 190 lines of TestGateCacheRuntimeFlow (3 tests) appended to the existing TestGateCache class block. Tests: 140+70 critical-path tests pass (test_v3_server_minted, test_error_envelope, test_handle, test_protect, test_capabilities, test_drift_fixes_2026_07_04, test_runtime_branches, test_transport_branches, test_integration_contract, test_high_reliability_fixes). The 3 new TestGateCacheRuntimeFlow tests are confirmed green on local pytest pre-commit. No regression on test_drift_fixes_2026_07_04 (15 fix(sdk) tests). Backends on 1.0.0 keep working unchanged. Pinning unchanged: SDK_MIN_VERSION_FOR_V3 = "0.12.0". Recommended upgrade path: 0.12.2 -> 0.13.0 (no on-wire breaking change). * fix(version): close orphan docstring so 0.13.0 section parses CI on PR #53 (`test (3.10/3.11/3.12)` all FAIL, coverage 0.05%) surfaced a SyntaxError collected in `src/nullrun/__version__.py` line 86: v3.13 / 0.13.0 (2026-07-04) — drift-fixes release: closes the SDK-side ^ SyntaxError: leading zeros in decimal integer literals are not permitted; use an 0o prefix for octal integers Root cause: in the 0.13.0 release commit I closed the module docstring with `"""` after the 0.12.2 section and then started writing the 0.13.0 section as if it were still inside a docstring — Python parsed `v3.13 / 0.13.0 (2026-07-04) — drift-fixes ...` as module-level expressions, barfed on `(2026-07-04)` (the `07` is a leading-zero integer literal — illegal in Python 3), and the rest of the test collection cascade-failed with 64 collection errors. Coverage hit 0.05% because pytest couldn't even collect, not because tests regressed. Fix: drop the orphan `"""` between the 0.12.2 and 0.13.0 sections and replace it with a `---` separator (the docstring stays open all the way to the module-final `"""` above `__version__ = "0.13.0"`). Verified locally: $ python -c "import src.nullrun.__version__; print(__version__.__file__)" (no SyntaxError, imports cleanly) `grep -c '"""' src/nullrun/__version__.py` now reports 2 (open + close), as expected for a single module docstring. * fix(lint): sort server_minted_idempotency_key import per ruff I001 CI on PR #53 (`test (3.11)` failure, others matrix-cancelled at fail-fast) surfaced `ruff check src/ I001` on the import block inside `NullRunRuntime._capture_server_minted_execution_id`: I001 Import block is un-sorted or un-formatted --> src/nullrun/runtime.py:2590:5 The block (added in the `fix(sdk)` commit) imported context helpers in this order: set_server_minted_execution_id set_server_minted_reservation_at set_server_minted_idempotency_key Ruff's alphabetic sort puts `idempotency` before `reservation`, so the fourth line is sorted up to the third slot. Pure cosmetic — no behaviour change, same set of imports. `ruff check src/` after fix: All checks passed. Verified on Python 3.10 / 3.11 / 3.12 locally before commit.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Why
How
Test plan
cd backend && cargo test,cd frontend && npm test)cd frontend && npm run lint)cd frontend && npm run type-check)Risk
Checklist
CONTRIBUTING.md(if present)