Kingsmai/pokopiawiki.tootaio.com
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(seo): centralize route metadata and expand sitemap coverage

Browse Source 
Remove static fallback tags from Nuxt config to prevent duplication.
Auto-apply noindex to authenticated and permissioned routes.
Add home, project updates, and legal pages to sitemap.
Properly escape JSON-LD structured data.
This commit is contained in:
xiaomai
2026-05-06 11:01:19 +08:00
parent 425f2f4d5f
commit f7986ca520
5 changed files with 23 additions and 30 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
SSR_MIGRATION_TASKLIST.md
View File

@@ -94,9 +94,12 @@ Keep this file aligned with implementation progress while the SSR migration is i
- [ ] Implement SSR data loading for stable public routes in small groups, starting with low-risk public pages.
- [x] Pokemon and Event Pokemon list routes SSR-load shared options and the first public list page.
- [ ] For each SSR-enabled public route, render title, description, canonical URL, robots value, Open Graph, Twitter card, and structured data from public business data and system wording only.
- [x] Route-level SEO output now owns dynamic title, description, canonical, robots, Open Graph, Twitter card, and valid inline JSON-LD without duplicate static Nuxt head metadata.
- [ ] For detail pages, use entity names, public images, localized public fields, and canonical detail URLs after public API data loads server-side.
- [ ] Preserve `noindex` on auth, admin, new, edit, and in-development routes.
- [x] SEO resolver defaults authenticated, verified, and permissioned routes to `noindex`, while existing route metadata continues to mark auth, edit/create modal, and in-development pages as `noindex`.
- [ ] Keep `robots.txt` and `sitemap.xml` generated from the same stable public route set documented in `DESIGN.md`.
- [x] Sitemap includes Home, public index sections, Project Updates, and legal pages; robots keeps auth, admin, edit/create, and in-development routes disallowed.
- [ ] Avoid serializing private auth state, raw permissions, internal audit payloads, or unneeded API payload fields into Nuxt payloads.
- [ ] Confirm localized reads follow the fallback order in `DESIGN.md`: requested locale, default-language translation, base field.
@@ -105,6 +108,7 @@ Keep this file aligned with implementation progress while the SSR migration is i
- Pokemon and Event Pokemon list routes now SSR-load the shared options payload and first public list page through `useAsyncData`; filter changes, infinite loading, and route-backed create modals continue to use the existing client behavior.
- Pokemon list SSR API failures are contained to null initial data so rendered HTML falls back to the existing skeleton/empty behavior without exposing backend stack traces, raw errors, or internal fields.
- Public Pokemon list SSR data does not request `api.me()` or forward cookies; create actions remain client-hydrated from the current user after mount.
- The static fallback SEO tags in Nuxt config were reduced to non-route-specific defaults so route-level SSR SEO is the single source for canonical, robots, social metadata, and JSON-LD.
## Phase 6: Browser-Only UI Isolation
@@ -148,6 +152,7 @@ Keep this file aligned with implementation progress while the SSR migration is i
### Phase 8 Validation Notes
- 2026-05-06: After SSR auth cookie forwarding and Pokemon/Event Pokemon first-page SSR data, `pnpm --filter @pokopia/frontend typecheck`, `pnpm --filter @pokopia/frontend lint`, and `pnpm --filter @pokopia/frontend build` passed. The current `lint` script runs `nuxt typecheck`.
- 2026-05-06: After SEO foundation updates, `pnpm --filter @pokopia/frontend typecheck`, `pnpm --filter @pokopia/frontend lint`, and `pnpm --filter @pokopia/frontend build` passed. Local built-server smoke on port `20116` verified `/pokemon` route-level canonical/meta/JSON-LD, `sitemap.xml`, and `robots.txt`.
## Phase 9: Cleanup