[rmcdaniel]

Workflow context identity

Refs #193. Add read-only ctx.workflow_id and ctx.run_id from the original workflow task context. Exact string IDs survive suspension/cold replay; the successor gets its new run identity. No transport, history, protocol, dependency or tracing-carrier change.

Current source 005a493, whole 0c4d9523b2c505622aa1c0cf30ca1a03e40fc661, based on main 81d8686. Test-first 1b6c32c has seven expected failures on unchanged runtime. The implementation makes them pass, including literal 0 / 17 / 017, resumed activity results, read-only access and successor replay with predecessor IDs supplied separately as business input.

The first 439 candidate's hosted language, package, integration, lint and Sonar gates pass; its regression-corpus guard and aggregate fail. That original stays failed. This head adds two immutable replay fixtures and passes local corpus validation plus 68 actual corpus/identity tests, Ruff PASS. Earlier 154 replay/identity tests and mypy for 36 source files remain their own result. Fresh full CI for this head is pending. Initial incomplete-fixture and missing-tool attempts are not counted as passes.

Reference documentation is included. Website Python guidance in all eight locales follows qualified publication. No accessor availability is claimed for current PyPI 2.8.0. Compatible minor publication and actual installed first-run/replay/continue-as-new acceptance remain required before closing #193.

All current task tooling/dependency/cache state was actually removed 07:26:30 UTC, before 08:00. Borrowed tools and foreign resources were preserved. Only clean pushed source and compact readable results remain.

[rmcdaniel]

The #194 candidate passes the hosted language, package, integration and Sonar checks, but its regression-corpus guard fails. The new identity accessors touch replay-sensitive source; the first candidate did not append immutable replay evidence. I am adding two actual replay fixtures: cold activity resume with the literal workflow ID 017, and a continue-as-new successor whose new run ID stays distinct from the prior IDs in its business arguments. No runtime or policy guard is relaxed.

Local corpus validation now recognizes both new fixtures; actual corpus replay is qualifying in isolated source tooling, with dependency/raw cleanup by 08:00 UTC. The first incomplete fixture and missing-tool attempts are unsuccessful tooling results, not SDK defects. Normal fresh-head CI remains required. No PyPI release or installed artifact acceptance is claimed yet.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud