22. Head tags by page type — Medium (PV, DH)
Added 11 Nov 2026. One URL of each page type fetched on both sites (checks/page_type_check/page_type_check.py, output page_type_check.json / .txt).
Which page types output a canonical (from @quintype/seo getSeoData + live check):
| Page type | Canonical | Title / description | Issue? |
|---|---|---|---|
| Story | story["canonical-url"] or story URL |
Story SEO fields | OK. AMP pages canonical to the HTML story. |
| Section (real section) | Section URL; ?page=2 / /india/2 also → page 1 |
Section SEO | OK (see #23). |
PV short-slug collection (/cartoons, /horoscope, /hasana…) |
None | Collection name / home description | Yes, #19. DH: own generator, OK. |
PV /<story-slug> fallback |
Self (wrong URL) | Home title | Yes, #21. DH: 301. |
| Tag | canonical-slug or self |
Tag meta, else name and "Read stories listed under on X" | PV: 4 tags show the home title and description (/tags/cartoons, /tags/horoscope, /tags/op-ed, /tags/weekly-horoscope, 1.9M impressions together). Both: default description is broken English. |
| Author | Self | Name / bio | OK (H1 count, #20). |
Static (/about-us) |
Self | Empty <title> and description on both sites |
Yes: add meta title/description to static pages in the CMS. |
App pages with no SEO handler: PV authors-page, cartoon-page, horoscope-page, premium, instagram, bookmark, subscribe, search; DH horoscope-page, premium, news-shots, newsletters, subscribe, search |
Self | Home page title and description (the library falls back to the home page) | Yes: duplicate titles. Give each a customTags entry in isomorphicRoutes (PV app.js ~line 610 already does this for home-page; DH passes customTags: {}). |
Search (/search?q=…) |
/search |
Home title | Indexable (index, follow). Should be noindex, follow. DH has 5 search URLs with impressions (1,303). |
| Bookmark, subscribe, profile-type pages | Self | Home title | Should be noindex. |
| 404 | None | "ಪೇಜ್ ಸಿಕ್ಕಿಲ್ಲ" / "Page Not Found", status 404 | OK for unknown paths (both sites return real 404s). Exception (12 Nov 2026): unknown DH tag slugs return 200 with the home title, see #28. DH /authors-page and /cartoon-page routes exist in code but return 404. |
| Shell (PWA) | Skipped by design | — | OK. |
Answer to "which page types don't output a canonical": only PV's short-slug collection pages (#19), 404s and the PWA shell. The 404 and shell cases are correct. The bigger page-type problems are the wrong-URL canonical on PV's story fallback (#21) and the duplicate home-page titles on app, search and some tag pages.
Where each wrong value comes from: code or CMS (checked 19 Nov 2026; PV tpml 0b3f7227, @quintype/seo 1.50.0 in node_modules; Quintype tags API on the origin host):
| Problem | Source | Mechanism | Who fixes |
|---|---|---|---|
| App pages (search, bookmark, cartoon-page, horoscope-page, premium…) show the home title | Code | getSeoData in @quintype/seo sends any page type it doesn't know (default:) to the home page's SEO data. PV app/server/app.js gives customTags only for home-page; DH passes customTags: {}. |
Tech: a customTags entry per page type |
| Search, bookmark, subscribe pages indexable | Code | Nothing sets noindex for these page types. |
Tech |
/about-us (and /app-support, /sitemap) empty title and description |
Code | These are hand-coded routes in PV app/server/routes.js:8-21 (STATIC_ROUTES, contentTemplate: "./about-us"), not CMS static pages. The library reads the title from CMS page data, which these routes don't have, so it is empty. |
Tech: give each route its own title and description |
4 PV tags (/tags/cartoons, /tags/horoscope, /tags/op-ed, /tags/weekly-horoscope) show the home title |
Code, triggered by data | PV data-loaders/tag-page-data.js finds the tag through stories (works), then looks it up with client.getTags(slug); for these 4 slugs the tags API returns 0 tags (checked 19 Nov 2026; bengaluru returns 1). The loader passes an empty tag, and the library falls back to the home page. Why the tags API has no entry for these slugs is not known. |
Tech: build the SEO tag from the story's tag when the lookup is empty (the loader already has the name); ask Quintype why these slugs are missing |
| Tag description "Read stories listed under on X" | Code default + empty CMS field | buildTagsFromTopic uses this hard-coded sentence when the tag's meta-description is empty (e.g. bengaluru: no meta-title or meta-description in the CMS). |
CMS: fill in descriptions for important tags; Tech: better default sentence |
So every item except the tag descriptions needs a code change; filling in CMS fields alone fixes only the tag-description part. DH static and app pages were not traced in code (live result only).