rega's gates:post run reported this gate failing its mapbox half — "NO map constructed", 21 api.mapbox.com requests, 0 bars. Two theories followed and both were wrong, mine included. rega then MEASURED the leaf on both token rungs: 1 Mapbox map, 2 GL canvases, 3 visible search inputs, 0 page errors, via the host global AND via ?tk=. The product was never at fault. This gate's detection was. THE MAP PROBE WALKED THE DOM FOR A CANVAS. That is a chain of assumptions — every shadow root open, the canvas carrying the class I expected, existing by the time I looked — and when any link breaks the probe reports ABSENCE, which is indistinguishable from a broken product. It now hooks the engine's own Map.prototype.setStyle through both globals, which cannot be wrong about the DOM because it never looks at it. Both globals, because the leaf loads only the engine its provider asks for; the property is defined with a setter so the hook survives the engine script loading afterwards, and the whole thing is wrapped so a hook can never be the reason a page fails. The canvas walk stays as a second opinion, so an engine that stops routing through setStyle degrades to the old behaviour rather than accusing the product again. THE BAR SELECTOR MATCHED ONE CLASS. input.mapboxgl-ctrl-geocoder--input, and nothing else — so it reported 0 bars on a page carrying three visible search inputs. Whether the miss was the class or the shadow walk cannot be told from here, and that is exactly the problem: a selector narrow enough that one wrong assumption yields ZERO will accuse the product of being broken when it is only wrong about a name. It now counts what a reader would call a search box. Both probes are borrowed from rega's tokrung.mjs, which measured this leaf correctly while this gate did not. Where a positive control exists that measures correctly, copy the control. IT STILL MEASURES BOTH TOKEN RUNGS, per studio. The ?tk= theory turned out to be wrong, but the two cases are worth keeping on their own merit: the host global is the rung an embedding partner reaches for, and if the two ever diverge the gate now says which one failed instead of leaving a channel switch to look like a fix. Rejected api.mapbox.com responses are recorded too, so "requests but no map" can be told apart from a refused key — a 401 says the key, not the code. THE LESSON, recorded in the header because this is the second time in two days that a probe of mine accused a working product: a measurement that yields ZERO when its own assumption is wrong will always blame the thing it is measuring. Source only; this commit edits one gate and changes no product code. I cannot run it — Playwright is rega's. |
||
|---|---|---|
| .. | ||
| all-people-groups | ||
| big-picture | ||
| cities-internet | ||
| country-posters | ||
| dashboard-plus-grid | ||
| demo-map | ||
| doxa-maps-spa | ||
| doxa-research-map | ||
| doxa-simple-map | ||
| doxa-vision-card | ||
| golden-path-check | ||
| gospel-resources | ||
| key-cities | ||
| my-upg-100-list | ||
| people-groups-grid | ||
| plugin-studio | ||
| pplr-data-maps | ||
| prayer-walk-planner | ||
| regions-dashboard | ||
| research-dashboard | ||
| research-map-plus-grid | ||
| stack-demo | ||
| template-bundle | ||
| upg-neighbor | ||
| wall-of-people-groups | ||
| .gitkeep | ||
| README.md | ||
plugin-designs/ — where YOUR maps live
The shared template lives in ../library/ (aliased @map) — you never edit it per-map.
You build your map by making a folder here, plugin-designs/<your-map>/, and composing the
template's building blocks. Your folder is yours (colors, API, i18n); the template is shared
(Apache-2.0, propagated to every map on the next build).
Shared vs. yours — the balance
library/ (@map) ── SHARED, reusable, parameterized ──► legend · map · map-controls
│ (every map reuses these)
│ imported by
▼
plugin-designs/<your-map>/ ── YOURS ──► the map's "main" class (the .vue profile = the Vue
│ instance for that map)
└── src/ ── YOURS ──► this map's OWN non-reusable components, colors, API, utils
One bundle can hold as many maps as you want — add another .vue profile and rebuild.
- Reusable →
library/. The legend, the map, the map-controls are shared, parameterized components every map reuses. Fix once, every map benefits. Never edit it to customize one map. - This map's own →
plugin-designs/<your-map>/. It holds two things: the map's "main" class (each.vueprofile is the Vue instance for that map) and that map's own non-reusable components (insrc/). Nothing here leaks to another map. - Many maps per bundle. The bundler bundles one map or many — add another flat
.vueand rebuild.
Production or experimental — every profile declares itself
Every app profile is one of two things, and experimental profiles are unlimited and encouraged — create as many as you need:
- Experimental, not built: prefix the folder with
_— the build skips it; it just sits in code. Zero cost, zero risk. - Experimental, built: builds normally, but it belongs in the Experimental section of the power app's layout (nav section) — never embedded on partner surfaces.
- Production: a real product — main sections of the power app + individually embeddable.
(The power app = the SPA layout that composes all profiles into one application.)
The pattern in one picture
plugin-designs/<your-map>/ ← one FOLDER = one BUNDLE, named after the folder
├── index.js ← entry: registers the custom element, finds the profiles
├── overview.vue ← one .vue = one PROFILE (flat in the folder — no profiles/ subfolder)
└── detail.vue ← another profile in the SAME bundle
builds to ─────────────► public/dx-maps/<your-map>/<your-map>.js (one file)
- Folder name = bundle name. Folders starting with
_or.are skipped (handy for WIP). - Each
.vue= one profile (one map screen). One bundle can hold many. - Pick the profile at embed time with the
profilefield — every profile shares one bundle:
<script src="/dx-maps/<your-map>/<your-map>.js"></script>
<your-map profile-config='{"profile":"overview","tk":"pk.eyJ..."}'></your-map>
<your-map profile-config='{"profile":"detail","tk":"pk.eyJ..."}'></your-map>
Add a map — three steps
# 1. create the folder + one profile
mkdir -p plugin-designs/<your-map>
# index.js ← copy doxa-simple-map/index.js, rename the tag
# <name>.vue ← your screen, built from @map/* blocks (flat in the folder)
# 2. build
npm run build # → public/dx-maps/<your-map>/<your-map>.js
# (just this one: BUNDLE=<your-map> npm run build:bundle)
# 3. embed the script tag + element (dev staging at `npm run dev` shows a paste-ready snippet)
Your three things (everything else is shared)
- Colors — how pins are colored. Pick one the template ships (
getColorStrategy('language-family')), or drop your own inplugin-designs/<your-map>/src/colors/(start fromtemplate-bundle/src/colors/example-mode.js). Never edit@map/colors/to recolor your map. - API — your map reads its data from the
profile-configon the page, so the same bundle points at different data with no rebuild.DataSourceManageracceptscsv(bundled file),api(one JSON URL), orrest-api(a live REST endpoint, the default). Prefer the real API over inlined JSON. - i18n — the template ships 11 base locales in
@map/i18n/locales/. UseuseI18n()for strings andRTL_LOCALESfor right-to-left; add or override strings from your own profile.
The one hosting variable
The only thing that changes between hosting modes is the <script src>: a relative path when
your own site serves the file, or an absolute CDN URL when a partner site serves it. The bundle
is identical either way. See ../documentation/CONTRIBUTING.md for the two CDN modes and partner-site embedding.
Profile patterns (mix freely)
- Individual — one
.vueper bundle (simplest). - Parameterized — one profile, behavior driven by
profile-configprops (e.g.tabs[]). - Nested — one profile that renders others (workbench-style, like
doxa-research-map).
See also
doxa-simple-map/index.js— fully commented reference bundle entrytemplate-bundle/— minimal two-profile skeleton to copy../library/colors/README.md— the color-strategy contract../README.md— repo overview & build pipeline ·../documentation/CONTRIBUTING.md— build vs dev, CDN modes