feat(oxc): lint results inspecteor#432
Conversation
@vitejs/devtools
@vitejs/devtools-kit
@vitejs/devtools-oxc
@vitejs/devtools-rolldown
@vitejs/devtools-vite
@vitejs/devtools-vitest
commit: |
| "@clack/prompts": "catalog:inlined", | ||
| "@vitejs/devtools-kit": "workspace:*", | ||
| "ansis": "catalog:deps", | ||
| "cac": "catalog:deps", |
There was a problem hiding this comment.
I am actually ok with keeping gunshi for a bit of diversity
There was a problem hiding this comment.
cac is more lightweight than gunshi. Given that the current CLI is relatively simple, I think it's a better fit for now. We can consider switching to gunshi if the CLI grows more complex in the future.
|
The complete implementation of the lint inspector will be submitted in a separate PR. The current proposal is:
|
|
Would you like to directly implement it in this PR? Right now this PR only remove the cli which makes this devtools' feature incomplete |
# Conflicts: # packages/oxc/src/app/app.vue # packages/oxc/src/app/components/SessionCard.vue # packages/oxc/src/app/components/SummaryCard.vue # packages/oxc/src/app/pages/index.vue # packages/oxc/src/app/pages/lint/report/[id].vue # packages/oxc/src/app/pages/lint/report/index.vue
|
The new changes look awesome! Great work! A bit nitpicking for the ui, for the detailed page, do you think it's worth being a dialog instead of a page (like before)? As this is where users would spend most of the time on, I wonder if it would be better to be like a page? |
# Conflicts: # packages/oxc/src/node/rpc/index.ts # pnpm-lock.yaml
|
Yeah, A separate page would be better. Will implement it in the follow-up PR. |

Summary
oxc-devtools-play.mp4
Non-Oxc changes
packages/core/playground: updates the playground sooxc-devtoolscan be exercised and visually verified there.packages/rolldown/src/appandpackages/ui/src/utils/date-groups.ts: move the date-grouping helper into the shared UI package so Rolldown and Oxc can reuse it.packages/ui/src/utils/icon.ts: reorder package-ID matching to prevent package paths from being classified by an earlier, broader rule.Closes #433