The map's packet built a jsPDF file and downloaded packet.pdf while its poster already
printed. It now goes through usePosterPrints.printPacket, the same path as Simple
Research (8457b5dc): the live map rides in as an image data URL, and an explicit
adapter maps the rows, reading religionName and countryName so a card never prints a
religion code ("MSN") under a people group's name.
check:composite-freshness found stale on main: doxa-research-map. Rebuilt, tree + SRI
regenerated; the landing precondition (fresh, one folder, SRI) passed in main-capture scan.
Both halves are rega's measurements on main 373b7979b, now that the WAGF selection holds and it is worth
optimising rather than optimising a bug.
NINE WRITES PER CLICK -> THREE. A single "Asia (Region)" click produced nine regions-* paint writes:
fill-opacity, line-color and line-width, three times each inside ~110 ms. Two paths painted the same
highlight from DIFFERENT sources — onSemanticTreeSelect from node.isoCodes, applyDimFilter from
mapLayers.regionIsoMap — each triggering the other through mapStore. Two sources for one visual is not
just wasted work: it is how they drift, and it already cost a debugging cycle, because when the highlight
was being cleared we first had to establish WHICH writer had done it. All four paths (click, watcher,
clear, and the layer-add below) now go through paintRegionScope, which writes a property only when the
value actually differs — asked of the MAP rather than a remembered key, so a layer torn down and re-added
cannot leave a skip holding a paint that is no longer there.
5.5 s AND 11.1 s OF SHADER COMPILE, MOVED OFF THE GESTURE. rega measured the row click compiling a new
variant — 5.5 s MapLibre, 11.1 s Mapbox on SwiftShader — because that click was the first moment these
three properties stopped being constants. Identical to the pins defect and fixed the same way: the idle
value is now ['case', ['has', 'iso_3166_1_alpha_3'], x, x], both branches the same so not one pixel
changes, and it is installed the moment addRegionsLayer returns. The variant therefore compiles on the
tab visit that is already loading a 6.4 MB GeoJSON, not in the middle of a gesture.
That placement is the whole point, and it is what I got wrong on the pins the first time: an idle
expression that is never installed pre-warms nothing. Here it is painted at layer-add, where the layers
first exist.
LEFT ALONE DELIBERATELY: highlightCountryByIso (0.30/0.02) and applyDimFilter's dim path (0.20) are
different states of the same layers, not duplicates of this one, so collapsing them into this writer
would have merged three features into one function to save nothing.
Verified: gates:pre 12/12, check:composite-interop PASS. Needs eyes: rega — expect 3 regions-* writes per
click instead of 9, and no compile on the row click; the compile should appear on the TAB visit instead.
check:composite-freshness found stale on main: all-people-groups
cities-internet
country-posters
dashboard-plus-grid
demo-map
doxa-research-map
doxa-simple-map
gospel-internet
gospel-resources
my-upg-100-list
plugin-studio
pplr-data-maps
prayer-walk-planner
regions-dashboard
research-dashboard
template-bundle
upg-neighbor. Rebuilt, tree + SRI
regenerated; the landing precondition (fresh, one folder, SRI) passed in main-capture scan.
Caught by reading the normalizer instead of waiting for the render: my adapter in f6a9d9d2 read
`g.religion` first, and DataSourceManager normalizes that field to the CODE — so every people-group card
would have printed MSN or BLA under the name, with the readable form sitting unused in `religionName`
one line away (DataSourceManager.js:357-358).
The country line had the same shape of error for the same reason. The renderer prints
flat(g.country_code) as display text next to the population, but the FIELD NAME invites handing it an
ISO code; normalized rows keep the code in countryIso and the label in countryName. Both now read the
readable form first and fall back toward the code, rather than the reverse.
This is the failure the commit before it flagged as the thing that could not be verified headlessly — a
wrongly mapped field prints blank or cryptic rather than throwing. It was findable without a browser
after all: the answer was in the normalizer, not in the rendering.
Verified: gates:pre 12/12. rega's check still stands and is still worth doing — this fixes the two fields
I could prove wrong, not the ones a rendered page would reveal.
Driver-lens audit card 2: the packet button downloaded packet.pdf while its poster sibling, two
functions above it, opened the print dialog. Same map, same menu, two different answers to "export".
The fix is not to wrap that PDF in a dialog. A PDF blob cannot go in the <img> the in-place print frame
uses, and every workaround — embed, object, the browser's own viewer — prints on some engines and
silently does nothing on others, which is the dead-button class this file has already paid for twice.
usePosterPrints.printPacket ALREADY does this: the same packet, authored as HTML with @page rules and
printed in place. This map was running a private second implementation of a shared product feature, with
jspdf behind it. Routing to the shared one deletes the duplicate, drops jspdf from this bundle, and the
packet gains the pages the shared renderer has that the local one never did — the life calendar, the
funnel and the prayer wheel, in the Driver's own page order.
★ AN EXPLICIT ADAPTER, because the mismatch would have been SILENT. The renderer reads slug / name /
population / religion / country_code, and usePosterPrints reads image_url; this map's normalized rows
carry country and imageUrl in camelCase. Handed over unmapped, the packet would have printed with blank
names and missing photos rather than failing — so the mapping sits in one visible place instead of
either side renaming its vocabulary.
Page 1 keeps the LIVE map rather than a fetched poster: this half is the one that knows what is on
screen. The shared renderer interpolates mapSvg as raw markup, so the PNG this map already knows how to
produce rides in as an <img>, as a data URL because the print frame is a separate document and an object
URL created here would be revoked out from under it. If that render fails the packet still prints — a
cover without the map beats no packet, which is the call the shared composable already documents.
Dead code removed with it: the download helper that served packet.pdf, and the lazy-import cache for the
composable this no longer uses. Neither export hands over a file now.
Verified: gates:pre 12/12. NOT verified here, and it needs to be before anyone calls it done: I cannot
see a rendered packet. The specific risk is the adapter — a field I mapped wrongly prints blank rather
than failing. rega should print one and look at the people-group cards before this is released.
The poster dialog had only a download path: its Render button saved a file, the
same trap PosterWall's old default was (gospel's Driver-lens audit, card 7).
A new action prop defaults to 'print': a PNG poster opens the print dialog in
place (printImageBlob, sized to the chosen page), a PDF opens in the browser's
viewer, and a blocked print frame or popup falls back to the file. 'download'
stays an explicit opt-in.
exp-shepherd found research-dashboard rendering ENDONYMS in every locale — 中国 and Türkiye on a Russian
page — and fetching a glyph range for every script on screen. The reading was that its basemap carries no
localized names and the choice was to swap the basemap or give up on its labels. Measured against the live
tiles, neither is necessary: the names are there, under keys this chain never tried.
The galaxy style is VersaTiles Colorful on the SHORTBREAD schema, whose label keys are name_ar / name_de /
name_en, not name:ar. This coalesce only tried the colon spelling, so all 43 of that style's label layers
fell through to `name` — the local name — while the localizer ran and reported success. A running fix that
cannot reach its data looks exactly like a missing feature.
MEASURED, six z3/z4 tiles across Europe, Asia and Africa, identical key sets on every one: name_ar,
name_de, name_el, name_en, name_es, name_fr, name_it, name_nl, name_pl, name_pt, name_uk. Of the eleven
locales we ship that is SEVEN in their own language — ar, de, en, es, fr, it, pt — with no change to the
basemap the Driver chose and no change to how the map looks.
hi, ro, ru and zh are genuinely absent from those tiles, so they now take name_en: one Latin script rather
than mixed CJK and Devanagari endonyms, which is also the single-script glyph saving. That is the same role
name:latin already plays for the OpenFreeMap styles, so it sits in the same position in the chain rather
than becoming a special case.
NOTHING CHANGES FOR THE OPENFREEMAP STYLES, by construction: the colon spelling stays first, so their
name:<lang> or name:latin still wins and the underscore entries are never reached. The order is also what
keeps the already-rewritten sentinel working — it tests for a chain beginning ["coalesce",["get","name:,
which this still is. Verified both, rather than assumed.
WHAT I DID NOT DO: swap the basemap. galaxy resolves to VersaTiles Colorful because the Driver asked for
blue oceans like the Mapbox map and that style's water measured closest (#bfd9f2; OpenFreeMap Liberty is
#9ebdff, noticeably stronger). Trading a look he chose for four locales' labels is his call, not mine, and
with this change those four get readable Latin labels anyway.
Verified: gates:pre 12/12. NOT measured here: glyph KB per locale, which needs a browser — rega has the
instrument and the before numbers.
The Driver's first laptop run with DX_SYNC_REMOTE: when main moved the page went
white. A landing rewrites dozens of built leaf index.html files under public/, and
Vite answers each changed .html with a full page reload (measured: one edited leaf
page, one full-reload broadcast), so a pull fired a burst of reloads while git was
still writing. The dev server no longer watches public/, which it never serves
through its module graph (measured after: no reload). Rudder's corner pill also
grew with every streamed word; while recording it is now 300px wide and shows the
newest words, cut from the front.
AUDIT34 card 9 (LOW): window.__UPG_MAP__, __UPG_SELECT__ and __UPG_SEARCH__ are
last-writer-wins. Embed two copies of this map on one page and the second mount
overwrites the first, so a gate driving 'the' map silently drives whichever
happened to mount last.
Adds __UPG_MAPS__, __UPG_SELECTS__ and __UPG_SEARCHES__, each keyed by the
embedder's instanceId (the same field poster-wall.vue:55 keys on), falling back
to the profile name.
THE BARE HANDLES ARE KEPT AND STILL POINT AT THE LAST MOUNT, deliberately. Every
existing consumer reads them directly -- internal/check-map-data-painted.mjs plus
~10 call sites across data/visual-gate.mjs and data/palette-preview.mjs -- and
those are browser gates. Removing the bare handles would break them to fix a case
they never hit, since a gate page mounts exactly one copy. This change is purely
additive: nothing reads the registries yet, so no consumer behaviour moves.
The honest limit, stated in the code rather than papered over: two copies with NO
instanceId and the SAME profile still collide, because nothing in the embed
contract distinguishes them. A mount counter would give them unique keys but make
the key unpredictable between mounts, so a gate could not name the instance it
wants -- worse than a key that repeats.
AUDIT34 card 3 (gospel's Driver-lens audit): the on-screen prayer-card fetch
hardcoded &lang=en, so the wall showed English people-group and country names
whatever language the reader picked -- while every word of the interface around
them was translated.
Fixed at three more sites than the card named, all reader-visible:
poster-wall.vue:161 the prayer-card deck (the card's site)
upg-100-list.vue:1983 countryNameMap, which feeds resolveTerritory() -- so
these names are the territory labels and the legend
upg-100-list.vue:2294 the MAIN records fetch: the pins, their popups, the list
The first uses getRuntimeLocale() || detectLocale(), the pattern the sibling
plugins already use (people-groups-grid.vue:276,
research-map-plus-grid.vue:147). The other two use this file's existing
packetLang(), which its two packet fetches already used -- so all four fetches in
that file now read one source.
The printed poster is deliberately NOT changed: generate.mjs fetches lang=en of
its own accord, and whether a print artifact is English-only is an open question
with the Driver (raised by regions-plugin). This is the on-screen surface only.
exp-shepherd measured on main: SERIES=country ONLY=country-yem made a POSTER,
and the matching packet.mjs run died with 'no zone matches ONLY=country-yem'.
The packet half of SERIES=country could not run for any country.
Cause: allZones() built zones from clusters.json -- the UPG-100 zoning -- where a
country id never appears. generate.mjs got its country zone source from
countryMode in the previous commit; packet.mjs did not. The same 'one concept,
two switches' defect I had just removed from one file survived in the other.
MY OWN TEST MISSED IT BECAUSE IT WAS NOT END TO END. I ran the country packet
against a UPG-100 zone id (ONLY=west-africa), which exercised the WORDING and
never the zone source, then reported the parameter verified end to end. Running
the actual pairing is what found it, and exp-shepherd did that, not me.
allZones() now returns the country zones under countryMode, built to match
generate.mjs's country branch exactly -- same ids, labels, adm0 and the PLURAL
country_codes selector shape -- so a poster and its packet cannot describe
different zones. Same two-path country-list lookup and the same error naming the
command that writes the file.
Verified: country-yem builds 8 groups / 6 pages, SERIES-ISOLATION PASS with 9
country markers and zero UPG-100 framing; the UPG-100 default is unchanged at 12
pages and 13 markers; all four packet gates PASS.
rega read the rendered page and found a fourth leak site my gate had passed: a
SERIES=country poster still printed 'UNENGAGED PEOPLE GROUPS', 'Unengaged of West
Africa', 'Series: List of 100 UPG's · map 11 of 28' and a side box calling it one
of a 30-map set. I had reported that PASS to three agents as proof the render was
clean. It was not proof: the gate's vocabulary was a HAND LIST of the phrases I
had already found, so it could not see framing I never enumerated.
ONE SWITCH (dx-shepherd's call). SERIES=country now implies the country zone
source, the series name and the header strings. COUNTRIES=1 is kept as an ALIAS
that resolves to SERIES=country, so runs that predate series.json keep working
and now get the matching packet too. Two parameters for one concept is exactly
what let rega's render take the new packet path with a UPG-100 header.
The five header strings move into series.json: kicker, title qualifier, series
name, side-box heading and body. One file owns the wording.
THE GATE NO LONGER DEPENDS ON WHAT I HAPPENED TO KNOW. Its vocabulary is DERIVED
from series.json -- every string a series declares becomes that series' own
fingerprint, so adding wording polices it from that moment. Strings shared
between series are dropped (they identify nothing) and a 12-character floor keeps
short phrases from matching innocent prose. The hand list is kept and merged for
phrases no series declares, like 'Project 7000'.
It also strips HTML and CSS comments before matching: generate.mjs ships design
notes into the SVG that QUOTE these strings, and a correct country poster failed
on five of them. A detector that does not know the comment syntax of what it
reads attacks the documentation of the thing it protects.
FOUND WHILE WIRING IT, a pre-existing break: COUNTRY MODE HAS NOT RUN AT ALL.
generate.mjs read plugin-designs/country-posters/data/countries.json, but
ddde518a8 moved that data to library/data/country-posters.json and the reader was
never updated -- no file at that path on main, so every COUNTRIES=1 run died on
ENOENT. Both paths are now tried, newest first, with an error naming the command
that writes it. Country mode builds 173 countries for the first time.
(node --check passed on the fix while existsSync was still unimported -- a parse
check is not a will-it-run check, and the free variable sat in the one branch my
tests had never reached.)
Verified: country header now reads 'Unreached People Groups / Pray for Republic
of The Gambia / What is a Country Prayer Map?'; the UPG-100 default header is
unchanged; the stale poster still fails as country with 11 real hits, so the
comment strip did not blind the check; COUNTRIES=1 alias renders correctly; all
four packet gates PASS.
The ordering rule 'regenerate the poster with SERIES=country BEFORE the packet'
was a fact about the build living in a human's head. Documentation does not hold
that kind of fact: the same bet already failed for 'regenerating the tree alone
ships a stale .js', which is written down in the brief and still happened.
The hazard is silent and I walked into it while building the series parameter.
The three below-map blocks are DRAWN INTO the poster's SVG by generate.mjs, and
the packet lifts that SVG whole, so the text reaches the packet through the
IMAGE -- no string file mentions it and no grep of packet-strings.json can see
it. A country packet built against a leftover UPG-100 poster prints 'Thirty
disciples' on page 1 with no error, the right file size, and a generator
reporting success. It looks exactly like the SERIES parameter does not work.
So the build asserts it instead of remembering it. The packet reads the poster's
series out of the artwork it is about to lift -- fingerprinted by the NETWORKS
headings, which are the one marker that survives every stage of the pipeline into
the artefact -- and exits 1 with the remedy command when it disagrees with its
own SERIES. Suggested by regions-plugin, whose two near-misses today were the
same class: assert against the ARTIFACT with a marker that survives the pipeline.
Verified in all four directions: country packet vs upg100 poster FAILS, upg100
packet vs country poster FAILS, and both matching combinations build clean. An
unrecognised poster is deliberately not an error -- that is the existing 'no
<svg>' branch's failure, not this one's.
check:composite-freshness found stale on main: country-posters
doxa-research-map
key-cities
my-upg-100-list
regions-dashboard. Rebuilt, tree + SRI
regenerated; the landing precondition (fresh, one folder, SRI) passed in main-capture scan.
The Driver's "the research map isn't connecting with the legend", on the WAGF Regions tab. It connected
and then wiped itself 0.4 s later, which is why it read as not connecting at all.
rega's stack trace ended three rounds of my guessing. The highlight paints at +51 ms; at +438 ms
clearAllHighlights runs, reached from applyDimFilter({kind:null}) through the mapStore.selectedRegion
watcher — so selectedRegion had gone FALSY while externalRegion still read "asia". Not the echo guard I
suspected first, not the JS filter scan I suspected second, not the shared legend's nodes reset I fixed
third (that was a real and separate bug). The round trip did it.
A row click emits map-filtered; the composite pushes the state back down as externalFilters; that payload
carries every key it knows — region:"asia" AND language:"". This handler set the region and, one line
later, "cleared" a language that had never been set. mapStore.selectLanguage() nulls selectedRegion by
design (exclusive selection, Bug 13) and does so even when its argument is null, so selectLanguage(null)
is a clear-everything in disguise. The 0.4 s is the debounce plus the trip through the composite and the
grid, which is why it looked like a mysterious timer.
An empty incoming value is now a NO-OP unless that dimension is actually lit. A genuine host clear still
clears — the dimension is lit and the new value is empty — and a genuine host selection still swaps
exclusively, which is what Bug 13 asked for. Guarded on BOTH axes, not just the one that bit: region and
language are symmetric here and fixing one would have left its mirror image.
★ THE SHARP EDGE IS IN THE SHARED STORE, and I am not moving it from here. selectLanguage(null),
selectRegion(null) and their siblings each null the other dimensions unconditionally, so every caller that
means "clear this one axis" is silently clearing all of them. That is load-bearing for other maps and
changing it belongs to whoever owns the store, so this fix is profile-local and the edge is recorded
rather than patched from a plugin lane.
Verified: gates:pre 12/12. Needs eyes: rega — the WAGF "Asia (Region)" click should now still be lit at
5 s, with the legend row keeping its state class.
Driver rule 2026-09-28: an export-poster button opens the print dialog with a
preview, never a direct download. PosterWall defaulted posterAction to
'download', so country-posters and key-cities, which set nothing, downloaded.
The default is 'print'. check:poster-action-print (gates:pre) keeps the default
and fails a consumer that opts into 'download' without a
"posterAction-download: <why>" comment. PDF packets and CSV exports are not
poster buttons and stay downloads.
The Driver: red dots on a mostly green map look wrong. A dark slate dot (#1e293b) with the existing
white ring reads on every fill the map uses, the green/amber/red/grey country bands and India's
pale-to-deep blue states, and it carries no good/bad meaning (red is this map's 'under 25%').
Candidates compared side by side on the world, India and West Africa views before choosing.
Rebuilt cities-internet and regenerated the tree (integrity hashes).
SUBDOMAIN_APPS=host=bundle/profile makes e.g. research-gallery.<domain> serve the /cdn build of
doxa-maps-spa at its root. A separate server on its own port answers only those hosts and only with
published /cdn files (no admin, PocketBase or dev server behind it), so no sign-in is skipped anywhere
else. The root page gets a <base href> at the plugin's real folder, so its relative links resolve as on /cdn.
The country posters shipped carrying UPG-100 framing: the UNENGAGED PEOPLE
GROUPS kicker, 'This packet is 1 of {n} UPG-100 lists', the Project
2000/7000/4700/700 narrowing funnel, and a below-map block telling the reader to
divide the set among THIRTY disciples -- coherent for ~30 UPG-100 zones and
incoherent for ~170 countries. Reported by the Driver.
Not a mislabel: there was NO series parameter anywhere, so 'the packet' and 'the
UPG-100 packet' were one thing and nothing could disagree.
series.json declares each series' funnel flag, string overrides and the three
below-map NETWORKS blocks. Both packet.mjs and generate.mjs read it, because the
leak was in TWO independent places -- the packet strings and the poster's own
furniture, which is not a packet string at all, so fixing either alone leaves the
other printing. SERIES absent or unknown resolves to upg100.
The funnel page drops for country: it narrows every people group on earth to THIS
list, which is not a description of a country. Reusing the existing
missing-vision-funnel path means one way for the page to be absent, not two.
Four packet keys are overridden, not seventeen: 12 of the 17 are funnel_* and go
with the page; the 12 prayer points are invariant, because they are about a
PEOPLE and a country poster still prays for the peoples within it. Overrides
apply to every locale on purpose -- an English country kicker is visibly awaiting
translation, whereas inheriting the translated UPG-100 kicker prints the wrong
series fluently.
The country English is a DRAFT from the Driver's dictated shape, flagged
_draft_awaiting_driver; no translations until he signs off.
check:series-isolation asserts the property, with a self-test over 6 known-bad
inputs. It caught a real leak this commit would otherwise have claimed fixed: the
packet's page 1 lifts the whole SVG from the pre-rendered poster, so a country
packet built against a stale UPG-100 poster still printed 'Thirty disciples'.
Verified at the rendered layer -- country 11 pages, upg100 12, the delta exactly
the funnel page; upg100 networks byte-identical to the text they replaced.
The ROP Tree and Language Families legend tabs (and the 'Go up to …' crumb built
from them) were typed in English; they now read the existing tabs.* catalogue
entries. ROP25 stays as the registry's name.
pplr carried private copies of SemanticTreeLegend and usePplrInstance that had drifted.
The instance copy lacked activeTabTouched, so the legend tab snapped back to its
default after a pick; the legend copy spoke English accessibility labels in every
locale. Both are deleted and the views import the library versions.
file-map.md documented `SOW_EXPORT_BUTTON=false` as a hidden global switch that "hides the
fully-wired Export menu across ALL bundles". SOW_ appears nowhere in this repo, so anyone setting
it would get silence and conclude the Export menu was broken. Docs describe the code as it is, so
the claim is gone.
The per-component prop it mentioned is real and untouched: LegendMobile.vue passes
:export-enabled. Only the global env override was fictional. MAX_LIVE went with the staging.js row
it lived in, and neither name appears in README.md or documentation/ now.
If a global export switch is wanted, that is a card, not a doc edit.
Audited README.md and all 35 documentation/*.md against main. Script names and markdown links
were already clean — 59 documented scripts all exist in package.json, 0 broken links. What was
wrong was file-map.md describing files that no longer exist, and three paths that had moved.
REMOVED, all archived in afb998f17 or earlier and still documented as current:
internal/staging.js — the repo-root staging page; the plugin studio is the only entry now
index.html — the same page's HTML
library/composables/useDoxaSearch.js — deleted in ea0c47b27, and dropped from a prose list too
REPOINTED: the three colour strategies documented under plugin-designs/doxa-research-map/src/colors/
live in library/colors/. THIRD-PARTY.md's two vendored files live under data/vendor/.
embeddability.md's reference to the staging script now says it was archived and where the entry is.
WHAT I DID NOT "FIX", because it was not broken: ~20 relative shorthand paths in the architecture
docs (`composables/useMapData.js` inside a document about library/) read correctly in context, and
three paths name things that are not repo files at all — a runtime endpoint (public/api/maps/token.json),
a generated artifact (out/_records.json) and a placeholder (./bundle/bundle.js). Reporting those as
defects would have buried the five real ones in 23 false positives.
TWO MEASUREMENT TRAPS I fell into on the way, both corrected before reporting: my first env-var scan
grepped only internal/, plugin-designs/ and library/ and declared 13 documented env vars unread —
11 of them are read in internal/build-lock.sh, .env.example, deployment/cloud/Dockerfile and the Go
backend. And an earlier sweep of my own data files reported six "rega" hits that were all the word
"aggregates".
main is the shared sandbox the dev server follows; /cdn is built only from PRODUCTION_BRANCH (default
production), which moves when someone promotes: git push origin main:production. The container watches
that branch and rebuilds /cdn once it has held a new commit for CDN_AUTOBUILD_MINUTES (default 2; 0 = the
admin button only); a failed commit is not retried automatically. The build fetches the branch into its
own ref (never touching the dev server's FETCH_HEAD) and uses that commit's own exclusion list, so
loosening it takes a promotion too. DEPLOY.md documents the promote step.
SKILL.md claimed new plugins land in a Research Gallery Experimental section.
That section was replaced by capability tiers (plugin-studio.vue:630); stage is
now a label prefix, not a place. Both stale lines corrected.
Adds what the skill never said:
- capability is DERIVED from composes and an explicit type is CHECKED against it
(check-capability-derived.mjs), with its two traps;
- rail labels come from meta, and check-studio-rail.mjs:68 fails a bare id;
- presets merge OVER the profile's meta config, key by key -- rule 3 of
documentation/PRESET-PLUGIN-STANDARD.md, asserted by check:preset-config-merge
(generate-tree.mjs:1094, the comment there records the pinHint bug);
- targets is read (profile-meta.mjs:28) and consequential: the Research Gallery
skips a bundle that does not list web (doxa-maps-spa.vue:605).
Depth moves to references/: glossary, targets, backend, workflow. backend.md
carries the PocketBase user tiers marked DESIGN INTENT, NOT BUILT -- quoted
rather than linked, because their source lives outside this repo.
Verified against this tree, rebased on main: 61 path citations resolve, 10 line
citations land on the code they claim.
README: one line pointing an agent at the skill.
check:composite-freshness found stale on main: cities-internet
country-posters
doxa-maps-spa
doxa-research-map
doxa-simple-map
my-upg-100-list
rudder-voice. Rebuilt, tree + SRI
regenerated; the landing precondition (fresh, one folder, SRI) passed in main-capture scan.
The gallery added every experimental bundle as a tab, so Rudder's engine probes
(/api/rudder/status, /devices, /events) logged three 404s on every embed of the
gallery. The manifest now carries each profile's meta targets, and the gallery
skips a bundle whose profiles are all corner widgets or declare targets without
"web". Rudder itself contacts its engine only where one can exist: the dev
server, the desktop app, or an explicit apiBase.
rega measured the WAGF Regions tab on Simple Research: the polygon highlight lands 51-85 ms after the
click, then 433-667 ms later it is gone — fill-opacity back to 0.7, line-width back to 0.6, no row
marked, every region at full colour. That reads exactly like the Driver's "the legend isn't connecting
with the map", and it is worse than not connecting: it connected and then undid itself.
Those two values are the fingerprint. 0.7 and 0.6 are written only by this profile's CLEAR paths, every
one of which is reached from a deselect — so something sent select(null) half a second after the click,
and nobody clicked twice.
SemanticTreeLegend watches the IDENTITY of its nodes array, and every tree feeding it is a computed that
returns a NEW array each evaluation. Any late async that touches a dependency — pins still arriving, a
lazily loaded layer, a refresh — rebuilds the array with IDENTICAL CONTENT, and the reset in that
watcher then cancelled the reader's selection: expandedIds cleared, selectedNode nulled, select(null)
emitted. The reset is right for genuinely new data; what was wrong is that it could not tell the two
cases apart.
It asks one question now: does the selected row still EXIST in the new nodes? If it does, re-point at
the new object and leave the tab and the expansion alone, because nothing about the reader's state has
become invalid. If it is gone, the reset runs exactly as before and the deselect is honest. Re-pointed
rather than kept, because the old object belongs to the discarded array and holding it would compare the
highlight against a node the tree no longer contains.
This is shared, so it is every tree legend on every map, not only mine: any profile whose data arrives or
refreshes after a click was silently losing that click.
NOT changed, and deliberately: the WAGF Regions TAB click also compiles shaders for regions-fill and
regions-border (rega: 2.9-4.1 s on SwiftShader), because those layers are added on that click. Warming
them at boot would mean paying for two more layers on every reader of every tab to save a one-time cost
on one tab, and the 6.4 MB regions GeoJSON is lazily loaded there on purpose. SwiftShader inflates a
compile by roughly an order of magnitude — the pins variant measured 3.2-3.5 s there and +0.5 s of real
boot delta — so the honest expectation on a GPU is a few hundred milliseconds, once, on a tab that is
already loading megabytes. That is a trade I am declining to make rather than one I missed.
Verified: gates:pre 12/12. Needs eyes: rega, the same WAGF Regions row click — the highlight should now
still be there at 5 s, and the variant compile may appear on that click once the selection sticks.
regions' headless sweep reported interop row 10 as a KNOWN DEFECT owned by me, while rega had proven the
same path working on both engines. dx-shepherd asked which was true. It is the register that is wrong.
★ PROVEN BEFORE DELETING, because "someone said it is fixed" is exactly how a stale exemption becomes a
hidden live one. Four independent checks on main:
- THE ENTRY'S OWN CLAIM IS FALSE. It said onShowOnMap "answers with go('map') only; it never re-dispatches
doxa:select-zone". It does — and it also writes a `select-zone` ATTRIBUTE first, so a pane that mounts
later still reads the request. The attribute is the stronger half and the entry does not know it exists.
- THE CITED COORDINATE IS WRONG. upg-100-mfe.vue:143 is an unrelated comment about tab rendering; the
handler is at :274. A file:line in a register ages into a false claim the moment anyone edits above it —
the third time this cycle a line number has drifted into fiction (gospel's engine offender moved
1490→1505→1509 the same way).
- BOTH FIXES ARE ANCESTORS OF main: 12a64702 (release the retained request on success — the FIRST press
was losing the zone) and 67e9ede5 (the wall clears its memory when its pane goes inert).
- The receiving side still registers its listener, and rega proved it end to end at 810ffa26.
⚠ AND THE REGISTER'S OWN CONTRACT SAYS THIS SHOULD NOT HAVE HAPPENED. Six lines above it:
"DELETED in the same commit as the owner's fix so main never carries a stale exemption or a red gate on
someone else's bug." The entry outlived its defect by three landings because the deletion was not part of
the fix. A convention that depends on remembering is not a convention; the MAPBOX_KNOWN block right below
records the same rule being followed correctly, which is why that one is empty and honest.
The block stays with the reasoning in it rather than being emptied silently, so the next reader sees what
was checked and why — and the printing machinery is untouched for the next real entry.
A preset page baked only the instance's props, so every default in the profile's
.meta.json config that the instance did not repeat was lost (go-int: upg-neighbor's
five presets dropped pinHint:false). instanceHtml now takes the meta config first
and the props over it. check:preset-config-merge (gates:post) reads every built
preset page and fails on a lost default; the rule is written into
PRESET-PLUGIN-STANDARD.md. Affected before the fix: upg-neighbor only; no
production preset has profile defaults to lose.
40 console.log calls (fetch progress, grid sizes, sample rows, API stats) ran on every
load; public-hygiene flagged them. Warnings and errors are kept.
check:composite-freshness found stale on main: all-people-groups
cities-internet
country-posters
dashboard-plus-grid
demo-map
doxa-research-map
doxa-simple-map
gospel-internet
key-cities
my-upg-100-list
plugin-studio
pplr-data-maps
prayer-walk-planner
regions-dashboard
research-dashboard
template-bundle
upg-neighbor. Rebuilt, tree + SRI
regenerated; the landing precondition (fresh, one folder, SRI) passed in main-capture scan.
70a9ce16 gave the cleared state a data-driven shape and stopped there, which was half a fix. The pins
layer is ADDED with a constant "circle-opacity": 1, and applyPinOpacity is otherwise first reached from
a legend click — so the new expression was only ever the CLEARED state, and the first selection was
still the moment MapLibre needed a new shader variant. The 3.2-3.5 s freeze rega profiled would have
survived the fix meant to remove it.
One line: applyPinOpacity(map.value) in the map-ready path, right after the layer is added. matchExpr is
left UNDEFINED on purpose — "re-apply what is current" — so a selection already restored from a
share-link hash is preserved rather than cleared at boot.
★ HOW IT WAS FOUND, because the method is the point: reading the BUILT bundle, not the source. The
addLayer call with its constant opacity is right there in the shipped JS, next to the pre-warm
expression I had just added — two facts that cannot both be true of a working fix. I had been about to
hand rega a tree to measure and would have read a still-slow first click as "the engine folds the
expression away", which is the wrong lesson and a day of chasing it.
Verified: gates:pre 12/12. Unchanged: this is still zero visual difference — both branches of the
boot-time expression are the same alpha.
rega's contact sheet showed what no gate could: page 3 read as ONE SOLID GREEN BAR. 32 contiguous cells
fill most of row 1 of 52, so the page said "a big green block" where the Driver's intent is "32 boxes lost
in 5,200". Rarity is the whole message and a bar cannot carry it.
⚠ AND THE COMMENT I DELETED SAID THE OPPOSITE, WHICH IS THE PART WORTH KEEPING. It read: "Row-major, so the
shaded run reads as a contiguous block from the top — a scattered shading of the same COUNT would be honest
arithmetic and an unreadable picture." I wrote that reasoning about a picture I had never looked at, and it
was exactly backwards. Scattering is what makes the eye hunt for them, which IS the point.
DETERMINISTIC, NEVER Math.random(): a 32-bit FNV-1a hash of the ZONE ID feeds a mulberry32 PRNG, so the
same packet prints identically every time — two copies of one zone must not disagree — while different
zones get different constellations, so a print run is not 28 sheets wearing the same pattern. Verified: two
consecutive renders produce byte-identical files (md5 febafc1813c6 both times).
Partial Fisher-Yates over the index space rather than shuffling all 5,200, so a large share degrades into
an even spread rather than clumping.
★ AND THE NEW ASSERTION IS THE ONE THAT WOULD HAVE CAUGHT THIS. check-packet-render now measures the shaded
cells' VERTICAL SPREAD as a fraction of the grid height, at the RENDERED layer via getBoundingClientRect —
because a COUNT cannot tell a scatter from a bar. 32 shaded is 32 shaded either way, which is precisely why
every existing assertion stayed green on a broken page.
POISON-TESTED rather than trusted: rebuilding the same 32 cells as a contiguous run makes it FAIL with
"span only 0% of the grid height". The real packet measures 95% (32 cells across 26 distinct rows). The 50%
floor is deliberately loose — a genuine scatter spans nearly the whole page, a bar spans ~1%, and anything
between deserves a human look.
⚠ The poison test first ran on a copy in /tmp and the gate REFUSED it — "outside any checkout, so which
build produced it cannot be checked". A sensible guard, and it meant my first poison run proved nothing;
re-ran inside the tree.
en and ar both PASS at 9 sheets. Page 2 (the empty life grid reading as blank graph paper) is UNTOUCHED and
held for the Driver with the page-order question — shading a reader's life we do not know would be a guess
presented as a fact, which is a product decision, not mine.
English readers got the style's own bilingual labels, native script under every name,
so an English page fetched a glyph range for every script at world view: Prayer
Progress loaded 18 font/sprite requests (777 KB) in English against 5 (185 KB) in
Arabic. English now gets the same single-language labels as every other locale.
The Driver: on Simple Research the research map is not connecting with the legend. It was connecting —
two seconds later, which a reader correctly reads as "nothing happened".
I was wrong about the cause twice, and both wrong answers were cheap to discard because they were
measured. First guess: a late echo through applyDimFilter's 80ms dedupe, whose toggle branch would read
the echo as a second click and clear the highlight. rega instrumented the pin layer every frame for 2 s:
ZERO 'legend:highlight' events on every click and no apply-then-revert anywhere. Second guess: the JS
filter scan in zoomToLegendRow, which walks 2,106 features evaluating ['in',['get','uniqueId'],
['literal', ~700 ids]] linearly. Benchmarked at the real shapes: 13.9 ms per full scan, 4.4 ms with a
memoised Set. Two orders of magnitude short, so nothing there was "optimised".
rega's CPU profile named it. ONE task of 3.2-3.5 s, self time getShaderParameter 3,583 ms +
getProgramParameter 551 ms: the GPU compiling a NEW SHADER VARIANT. The click was the first moment
circle-opacity stopped being a constant and became a per-feature expression, and a data-driven paint
property needs a different program. The second click on the same map costs 205 ms — the program is
cached — which is the signature of a one-time compile and not of anything in our logic. The legend's DOM
was never involved: 34 rows before the click, 34 after.
So the fix is not a smaller expression. It is that the first SELECTION must not be the moment the
variant is first needed. With nothing selected the value is now ['case',['has','uniqueId'], a, a] —
both branches identical, so not one pixel changes — and circle-opacity is data-driven from the first
paint. The program compiles during boot beside every other one, and a selection afterwards only
re-uploads attribute values.
The trade, stated rather than hidden: boot compiles one more program. On a real GPU that is tens of
milliseconds; rega measured seconds only because that box rasterises on the CPU through SwiftShader.
Paying it at load beats freezing the map mid-gesture, and the variant was going to be compiled on the
first interaction anyway.
Second finding from the same profile, fixed here too: circle-opacity was being rewritten 4-8 times for a
single click (+1.52, +1.57, +1.94, +5.17 s) because the selection handler, the glow sync and the slider
watcher all end up in this one function, and every write re-uploads the per-feature attribute. It now
writes only when the value actually differs — asked of the MAP rather than of a remembered key, so a
layer that gets torn down and re-added cannot leave the skip holding a dim that is no longer painted.
Verified: gates:pre 12/12 (check:setup-order included, since this touches module-scope state next to a
function that reads it). Needs eyes: rega, the same affinity click — expect the dim at the first frame
and one long task at BOOT instead.
rega, ?lang=ar on landing 38: "People's Republic of China" and "Lao People's Democratic Republic" were
still English in the top-countries chart. The API labels some countries with their FORMAL name, and
library/data/countries.json does not carry those — it says "China" and "Laos" — so reversing the English
name missed precisely the countries whose names are longest. Measured: both miss the name table, and
both resolve through the row (CHN -> CN -> الصين, LAO -> LA -> لاوس).
Every row already carries both halves: the label the chart aggregates on, and the alpha-3 in
country_code.value — the same accessor the choropleth joins on. Reading the pair out of the data is
exact and needs no alias table. The name table stays as the fallback, so a label whose rows are
currently filtered out still resolves, and a synthetic key ("Other") resolves to neither and renders as
itself.
Deliberately NOT a looser text match. "Democratic Republic of the Congo" and "Republic of the Congo" are
two countries, so stripping formal prefixes is how one would silently become the other — the failure
countryName.js's own header warns about, and the reason that file refuses to guess.
Still English, and not fixable from CLDR: the regions "Middle East" and "North America & Non-Spanish
Caribbean". M49 145 is "Western Asia" and 021 is "Northern America"; neither means the region we draw,
so naming them from CLDR would redraw the region rather than translate it. They need a translation
source, which is the Driver's call — the same class as the affinity blocs and language families.
Verified: check:sfc-parse PASS. Needs eyes: rega, ?lang=ar top-countries.
Driver decision 2026-09-28: the prayer is kept word for word. The
operational-vocabulary rule now removes exactly the prayer text at
packet-strings.json <locale>.points[0].text (every locale, raw and as a string
literal) from a scanned file before it looks; the same word anywhere else is
still reported. Measured: 4 warnings to 0, and a planted file using the word
is still flagged.
The copyable script snippet 404'd on bun run dev for every production map, while
the iframe snippet worked through leafRewrite. /<b>/<b>.js and /<tree>/<b>/<b>.js
now serve the built IIFE from public/<tree>/; a bundle not built yet gets a plain
404 that says to build it, never the HTML fallback.
rega's load waterfall: the engine was requested only after the component tree
had rendered, and the dashboard's 384 KB country boundaries only after the whole
basemap style had loaded, so each waited on the other. The engine download now
starts when the map composable is set up, and the boundaries when the dashboard's
map panel mounts; the style handler awaits that same download.
The rail showed Gospel x Internet as "board": a row's name fell back to its
profile id unless a hand-kept DISPLAY_NAME table had an entry. The manifest now
carries each profile's .meta.json label (or title / displayName), and the rail
shows it without a trailing aside in parentheses; with no label it shows a name
made from the ids ("Gospel Resources · Charts"). DISPLAY_NAME is an override
only and is empty. check:studio-rail fails if any row would show a bare id.
