Release v1.18.0: white operating states and emitter channels

[foadshafighi]

Summary

Records can now publish white operating states for tunable-white, dim-to-warm and multichannel luminaires, plus emitter channels, alongside the tested baseline. A referenced color_tuning block, one shared resolver and consistency validation keep the baseline authoritative, while builder 0.11.0 projects both numeric range endpoints. The workbook adds color-state and emitter-channel sheets with independent spectrum groups. Shared PIM guidance maps the new authored fields and generated range fields, with links from all four platform guides.

This prepares ULC 1.18.0 and adds one public LCVA example with endpoint report values and the original measured spectra. The eight existing examples retain their grades and change only the builder marker.

Test plan

• go vet ./..., gofmt check, go test -race ./..., and CLI build.
• All nine examples pass validation and build-index --check; schema drift guard resolves all 310 references.
• CLI fixtures exercise V1-V6, static-white refusal, repeated baseline/end refusal, labeled same-CCT presets, channels-only RGBW, dim levels and settings-only records.
• C1-C9 CSV/XLSX tests, spectrum-group ownership tests, rubric counts and release version guard 1.18.0.
• Native LP claims and document-truth review converged after three documentation fixes and a fixes-only second round. Parent gates ran; the claims lens's sandboxed executable/PDF verification was unavailable in its environment.

Callouts for reviewers

1. Tunable-white and dim-to-warm records need both endpoint CCTs for Standard and endpoint CRI Ra plus melanopic DER for Full. Stored records without ends fall below Standard after rebuilding. Existing grading rows continue to read the tested baseline.

2. Channels and endpoint spectra remain enrichment. The LCVA example is schema-valid at the incomplete floor with its ordinary missing-data roadmap. Unpublished endpoint flux and melanopic DER stay absent; through-optics status is unstated. Source files are individually hashed and excluded from the repository.

3. scope_version remains 2.0.0: 71 gating rows, 73 enrichment rows, two observations and 61 gating pointer rows. This release adds schema fields and vocabulary while preserving all eight original example grades.

4. Endpoint-metric roadmap messages retain the uniform row-level /color_tuning/states contract. A missing endpoint metric can belong to the top-level tested baseline. The shared mapping guidance explains both homes; endpoint-specific actual-path messages remain a future design refinement.

5. Published states in this release carry CCT and describe white operating conditions. Colored emitters have channel identity, peak wavelength, flux and spectra; colored scene presets are outside 1.18.0.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-17T07:46:47.360975Z	f66bbc1	New commits

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[foadshafighi]

@greptileai review

[github-actions]

Codex Automated Code Review

Code Review Summary

PR: ULC 1.18.0 adds published color states, emitter channels, endpoint grading, workbook conversion, and a tunable-white example. Reviewed all 58 changed files.

P0 - Critical Issues (Must Fix)

None found.

P1 - High Priority Issues (Should Fix)

None found.

P2 - Medium Priority Issues (Consider Fixing)

None found.

P3 - Low Priority Issues (Optional)

None found.

Positive Observations

• Taxonomy enums remain separate from structural definitions; additions describe fixture capabilities and published measurements.
• A shared state resolver keeps baseline ownership, CCT selection, index generation, and grading consistent.
• Workbook spectrum routing preserves independent provenance for each state or channel.
• New tests cover duplicate identities, endpoint consistency, spectrum routing, and grading behavior.

Validation: both schemas pass Draft 2020-12 structural checks, all 310 $ref pointers resolve, and all nine examples pass schema validation. The Go test suite could not start because the read-only environment prevents creating its build cache; runtime behavior remains unverified here.

---

Automated review by OpenAI Codex

[foadshafighi]

REBUTTED (no change required): the Codex automated review reports no actionable P0, P1, P2 or P3 findings on 4593182e824da6797db3e574afdbebea118d2538. Its schema checks passed. The reviewer could not start Go tests in its read-only cache environment; the implementation gate ran go vet ./..., go test -race ./..., the CLI build, all nine example validation/index checks and the schema drift guard successfully. Validator CI is tracked separately on the same head.

REBUTTED (no change required): the Codex connector summary also completed on 4593182e824da6797db3e574afdbebea118d2538 with no findings and no unresolved findings. The three required CI checks are now green on that head, including Validator CI. No manual Codex retrigger was requested.

[greptile-apps]

Confidence Score: 5/5

[High impact] The PR appears safe to merge, with no new actionable issues found.

Summary

Adds published white states and emitter channels alongside the tested baseline, with shared CCT resolution, range fields, endpoint grading, and workbook support.

• Tunable-white records publish their range endpoints.
• Workbook authors can enter states, channels, and their spectra.

_{Reviews (3) · Last reviewed commit: "Clarify white operating states and color..." · Reviewed by Greptile}

[foadshafighi]

FIXED (G1) / REBUTTED (G2): the Greptile review at 4593182 was folded in af4b3275e12e1b36284a1b6ba3ce74bd1459e98d. G1 is fixed by the shared PIM mapping section and all four platform links. G2 is rebutted as a design residual: the existing hard-gap layer has no record-aware endpoint-message hook, so the uniform row contract remains; the new guidance explains the endpoint requirement and the baseline's actual metric homes. Both inline threads are answered and resolved. The full local gate passed for the fold; same-head CI and automatic Codex review are running. No new Greptile review has been requested.

REBUTTED (no change required): Greptile subsequently reviewed af4b3275e12e1b36284a1b6ba3ce74bd1459e98d automatically, without a new request. Its updated summary reports Reviews (2), confidence 5/5 and no new actionable issues. The existing G1 FIXED and G2 REBUTTED dispositions remain recorded above.

REBUTTED (no change required): Greptile also reviewed final head f66bbc1b5f44b23a767e23f94379075dbd269a39 automatically, without a new request. Its updated summary reports Reviews (3), confidence 5/5 and no new actionable issue. The three required CI checks are green on this head. No manual Greptile retrigger or CLI review was run.

[github-actions]

Codex Automated Code Review

Code Review Summary

PR: ULC 1.18.0: published color states, emitter channels, endpoint grading, workbook conversion, and a tunable-white example.

P0 - Critical Issues (Must Fix)

None found.

P1 - High Priority Issues (Should Fix)

• Color-mixing presets require a CCT even when CCT is undefined. schema/ulc.schema.json:1455 requires every published state to carry nominal or measured CCT. Consequently, an RGB “Red” preset with valid chromaticity coordinates or a spectrum fails validation: saturated red has no meaningful correlated color temperature. The workbook repeats this restriction at tools/validator/internal/sheet/color_tuning.go:105. This prevents the advertised color-mixing state support from representing ordinary colored presets without inventing data. Require CCT for white-range endpoints, allow other color states to use chromaticity or spectral data, and update workbook conversion and state-identity checks accordingly.

P2 - Medium Priority Issues (Consider Fixing)

None found.

P3 - Low Priority Issues (Optional)

None found.

Positive Observations

• New enums remain in the taxonomy; structural definitions remain in the record schema.
• The shared resolver consistently handles baseline ownership and CCT precedence.
• Spectrum grouping preserves independent provenance and sample grids.
• Both schemas pass Draft 2020-12 schema checks; all 310 $ref pointers resolve, and all nine examples validate.
• No workflow changes or committed research/ files were found.

Validation limitation: The Go test suite could not start because the read-only environment prevents creation of Go’s build cache.

---

Automated review by OpenAI Codex

[foadshafighi]

REBUTTED (scope): in 1.18.0, published color states are white operating states with a CCT by design. The shared identity rule uses CCT, dim level and label, and the two endpoint grading rows read those white states. Colored emitters are represented as emission channels with peak wavelength, flux and spectrum; a colored scene such as an RGB Red preset is a different concept and is outside this release. The schema and workbook refusal demonstrated here match that intended scope. f66bbc1b5f44b23a767e23f94379075dbd269a39 clarifies the boundary in the shared PIM and workbook guides. No schema, converter, identity rule or test was changed. Arbitrary colored scenes remain later scope if a manufacturer publishes them.

This answers the automatic Codex review of af4b327. The full local gate passed for the clarification commit.

[github-actions]

Codex Automated Code Review

Code Review Summary

PR: ULC 1.18.0 adds published white operating states, emitter channels, endpoint grading, generated CCT ranges, workbook conversion, and a reference example.

P0 - Critical Issues (Must Fix)

None found.

P1 - High Priority Issues (Should Fix)

None found.

P2 - Medium Priority Issues (Consider Fixing)

None found.

P3 - Low Priority Issues (Optional)

None found.

Positive Observations

• New enums remain in the taxonomy; structural definitions reuse existing measurement and provenance shapes.
• A shared resolver keeps CCT selection consistent across validation, grading, and index generation.
• Tests cover baseline ownership, duplicate identities, endpoint constraints, and spectrum routing.
• Source documents in the new example carry SHA-256 hashes.

Validation completed: both schemas pass Draft 2020-12 metaschema checks, all 310 $ref pointers resolve, and all nine examples pass schema validation.

The Go test suite could not run because the read-only environment prevents creation of Go’s build cache. Runtime behavior and index parity remain unverified here.

---

Automated review by OpenAI Codex

[foadshafighi]

REBUTTED (no change required): the automatic Codex review of final head f66bbc1 reports no actionable P0, P1, P2 or P3 findings. Its schema checks passed. Runtime tests were unavailable in the reviewer's read-only cache environment; the separate same-head Validator CI executed successfully, and the local mandatory gate passed, including race tests and all nine validation/index checks. Release notes and schema-reference CI also passed on f66bbc1b5f44b23a767e23f94379075dbd269a39.

REBUTTED (no change required): the Codex connector summary also completed on this final head with has_any_findings=false and has_unresolved_findings=false. All emitted Codex review results on the pushed heads have been answered, including the af4b327 scope request.