rtmx-ai / rtmx / 26977756499

83%

push

github

feat(verify): multi-dimensional completeness policy + configurable vocabularies (v1.3) (#117)

* feat(verify): multi-dimensional completeness policy + configurable vocabularies

Two related closed-loop verification capabilities, registered as REQ-VERIFY-009
and REQ-VERIFY-010 (specs + RTM rows + dogfooded test markers included).

REQ-VERIFY-009 - configurable completeness policy. Status was one-dimensional
(one passing test => COMPLETE). The phoenix/do178c/iso26262 schemas already
define multi-dimensional verification columns, but nothing used them to gate
completion. A new `rtmx.completeness` block adds `policy: combinations`, which
marks COMPLETE only when passing tests cover >= min_combinations distinct tuples
of the configured dimensions (scope/technique/env). Default `policy: simple`
reproduces the historical rule, so existing projects are unaffected.
`require_all_pass` (default true) is preserved. Applies to the results path
(verify --results); the native go-test path keeps the simple rule.

REQ-VERIFY-010 - configurable dimension vocabularies. `rtmx.completeness.
vocabulary` lets projects extend the accepted scope/technique/env values so
schema-specific values validate (e.g. the phoenix schema's static_field /
dynamic_field) instead of being rejected. results.Validate is retained as a
wrapper over the new ValidateWithVocabulary; built-ins preserved when unset.

Backward compatible. Tests dogfood the dimensional markers via rtmx.Req(...);
all packages pass; dependency reciprocity reconciled.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

* Potential fix for pull request finding

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

* Potential fix for pull request finding

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

* chore: set v1.3.0 release date in CHANGELOG

---------

Co-authored-by: rhino11 <ryan@iotactical.co>
Co-authored-by: Clau... (continued)

111 of 122 new or added lines in 4 files covered. (90.98%)

13192 of 16252 relevant lines covered (81.17%)

28.54 hits per line