forked from molecule-ai/molecule-core
bfbbe57610
[Molecule-Platform-Evolvement-Manager] Closes two of the 0%-coverage files surfaced by the baseline run in PR #2147 (vitest coverage instrumentation). Both files are tiny utility helpers with high-touch read paths. ## utils.cn (8 cases) Wraps `twMerge(clsx(inputs))` — every conditionally-styled component flows through here. The load-bearing case is the **last-wins Tailwind dedup**: `cn("p-2", "p-4")` → "p-4". A regression that lost twMerge would silently double-apply utilities (cosmetically broken, breaks `:where()` rules + theme overrides). Cases: - single class unchanged - multiple positional classes joined - array input flattening (clsx) - object syntax with truthy/falsy keys - last-wins dedup on conflicting Tailwind utilities (the regression-locked guarantee) - non-conflicting utilities both survive (p-2 + m-4) - mixed input shapes (string + array + object + string) - nullish / empty inputs don't throw ## runtime-names.runtimeDisplayName (4 it.each cases + 3 it()) Friendly-name lookup that surfaces the workspace runtime in the chat indicator, details tab, and a few component labels. Cases: - known runtimes map to display strings (claude-code → Claude Code, langgraph → LangGraph, etc.) - unknown runtime falls back to input string verbatim (a NEW runtime not yet in the lookup still renders something operator-debuggable rather than a generic placeholder) - empty string falls back to "agent" (final default) - case-sensitivity pinned: "Claude-Code" / "LANGGRAPH" miss the lookup. The upstream slug is already normalized lowercase, so a future refactor that lowercases input "for safety" would silently change behavior — pinning the contract here. ## Test plan - [x] All 17 cases pass locally (~129ms) - [x] No SUT changes — pure additive coverage - [ ] CI green ## #1815 progress - [x] Step 1+2: coverage instrumentation + script (#2147) - [x] 0%-file gaps utils.ts + runtime-names.ts (this PR) - [ ] More 0%/low-coverage files: lib/canvas-actions.ts (25%), store/classNames.ts (17%) — separate PRs - [ ] Step 3b: thresholds + CI gate once baseline catches up 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> |
||
|---|---|---|
| .. | ||
| __tests__ | ||
| api | ||
| validation | ||
| api.ts | ||
| auth.ts | ||
| billing.ts | ||
| canvas-actions.ts | ||
| credits.ts | ||
| deploy-preflight.ts | ||
| design-tokens.ts | ||
| hydrate.ts | ||
| runtime-names.ts | ||
| runtimeProfiles.ts | ||
| services.ts | ||
| tenant.ts | ||
| utils.ts | ||
| ws-close.ts | ||
| ws-url.ts | ||