Three defects on country-map/country-map, which is production -- DX Life's
country pages. Found by auditing the element rather than from a report.
1. ANY profile-config change refetched all 2,106 records. The watch getter
returned a NEW ARRAY literal, and Vue compares a watch value with Object.is,
so a fresh array always reads as changed; both computeds read the injected
cfg, so the effect re-ran on every config change and the callback then fired
unconditionally. MEASURED with real Vue: flipping isDark refetched the whole
endpoint and redrew the map, and so did a host adding pinHint. A joined
STRING compares by value, so only a real language or country change refetches
-- which must still refetch, because the API localises the names.
This was latent waste until the composite re-forward made config writes
routine. It is now on every upstream locale switch, so the two changes are
complementary rather than independent.
2. A SLOW EARLIER RESPONSE OVERWROTE A NEWER ONE. loadPoints is async and had no
sequencing at all -- 0 abort or generation tokens in the file -- so two runs
could be in flight and the one that STARTED first could FINISH last. Both
assigned points, so the map showed the previous country's pins with no error
and nothing for the reader to notice. Every run now takes a ticket and a stale
ticket writes nothing: not the points, not loadFailed, not the spinner the
current run is still using. fetch is left to complete rather than aborted,
because the response is cached for the next switch back.
3. `loading` WAS WRITE-ONLY: assigned in three places and read in none, zero
template references. A reader on a country page waited on an empty map -- no
outline, no pins, no message -- which is indistinguishable from a country that
genuinely has no people groups, and this element fetches the whole list before
it can draw one pin. "Loading results..." was already in the catalogue in all
eleven locales, so saying so cost no new string and no parity debt.
check:country-map-refetch guards all three. It drives the REAL reactive chain
with real Vue and takes the watch key expression OUT OF THE SHIPPED SFC, because
bug 1 is entirely a question of what Vue considers changed and is invisible to
any test that fakes reactivity or re-types the watch.
Mutation-tested, 5 mutations, all caught by name. The first version of the gate
MISSED one: a bare /if \(!current\(\)\) return/ grep matched the guard in the
catch block, so removing only the guard before the points write still passed.
The assertions are positional now -- the guard must sit immediately before the
assignment it protects -- and the catch guard is asserted separately. That hole
existed because the behavioural half of the test reproduced the guard's shape
inside the gate instead of reading the component, which is the self-consistent
oracle this repo's other geometry gate explicitly warns against.
Registered in scripts and gates:pre, verified by re-parsing: one key, 107
checks. Bundle rebuilt (BUNDLE=country-map), then tree, then fix:sri.
The research page now gives this map three tabs (engagement, prayer,
adoption). With two or more tabs the bar was a row of plain buttons: a screen
reader heard three buttons, with nothing saying they are tabs or which one is
showing. The bar is now role=tablist, each button is role=tab with
aria-selected, and the buttons are type=button. A one-tab instance shows no
bar, so it is unchanged.
resources-map.meta.json carried NO stage, so the manifest defaulted it to
experimental — while doxa-maps-spa, the gallery that becomes Deep Research,
mounts it: doxa-maps-spa.meta.json composes "gospel-resources/resources-map"
and TAB_DEFS names it as the Gospel Resources tab. The record disagreed with
what ships, and the record was the part that was wrong.
WHAT THIS CHANGES: the manifest stage, and nothing a reader sees. Verified —
the diff moves one "experimental" to one "production" and ZERO sha384 hashes.
No bundle is rebuilt and no byte of any map changes.
The stricter production gates now apply to this profile and it passes them:
gates:pre 103 PASS / 0 FAIL / 1 NOT MEASURED of 104. That is the part worth
having, because marking something production is only honest if it then has to
meet the production bar.
NOT a decision about which profile the Deep Research page SHOULD show.
gospel-dashboard is the production composite and remains the better end state;
it is not switched to today because its .gd-map is sized 80svh — a viewport
unit inside a pane that is not the viewport — which against rega's measured
panes gives the map 91% of the pane at 1440x900 and 95% at 390x844, and
because it has never been shot inside the gallery. That fix is written and
held local for after the deadline.
- research-map.vue's openPacket read and wrote _packetBusy, a name 8ab3dc0f5 replaced with the
reactive packetBusy ref. Reading an undeclared name throws, so the packet export threw a
ReferenceError on every click, in doxa-research-map and in research-map-plus-grid, the production
map that compiles it. It uses packetBusy.value now (check:no-free-identifiers).
- check:list-prop-unit and check:optional-t-equivalence read every entry of library/i18n/locales as
a locale, and the folder has held a README.md since bba33e783. They read locale FOLDERS now.
- people-groups-grid's regions-look instance previewed two props (filters-toggle, backdrop) of a grid
merge that is not on development, so the page showed nothing different and
check:instance-config-keys failed. It returns with that merge.
useShadowStyles keys its <style> by the label and overwrites an existing
tag with the same key, so two components sharing a label keep only the
last one's CSS in a shared shadow root. The regions list lost its sheet
this way to the region card, which had kept the list's label when it was
extracted.
check:shadow-style-labels-unique (gates:pre) fails any label two calls
share, globally rather than per shadow root, so a copied or forked
component is caught when it is copied. It reads the real parsers,
classifies every reference to the function (a rebinding, .call, a
re-export or a namespace import fails instead of hiding a call),
cross-checks the AST against the token stream, and covers the other
ways two calls share a key: unlabelled calls under the dev server's
per-component key, and a fixed label over CSS that varies per caller.
countries-map.vue, a fork of research-map.vue, carried the same label;
it is now 'countries-map', with the country-posters bundle rebuilt and
the tree regenerated. The composable's JSDoc called the label a
debugging name; it now says it is the key.
Studio's 5B review found a dependency-tracking hole by testing the helper with
real Vue rather than reading it. entries was a plain array, so the watchEffect
tracked only what the FIRST registered getter read: with two panes reading
different refs, changing pane 2's source produced ZERO writes, and pane 2 caught
up only when pane 1's source happened to change. That is this helper's own bug
reappearing for any adopter whose panes read different sources. regions-mfe is
safe today because every getter reads the same cfg and spec -- which is exactly
why a reading could not find it.
entries is now shallowReactive, so register() re-runs the effect and every
getter is tracked. Also from the review:
- stale entries are PRUNED, not skipped: an adopter that rebuilds its pane
column would otherwise keep detached elements and re-run their getters on
every pass forever;
- a getter that throws is reported once per entry instead of being swallowed,
since silence recreates the symptom this helper exists to fix;
- the unused `tag` parameter is gone -- a documented check that does not exist
misleads the next adopter;
- the host argument now documents that it must be in the SAME tree as the
panes, because Element.contains() does not cross a shadow boundary and
passing the custom element would silently skip every write.
regions-dashboard rebuilt so public/ is servable at this sha for the browser
A/B. It is the only bundle that compiles the helper.
regions-mfe built each pane by creating the child element and writing its
profile-config once, inside onMounted, and nothing rewrote it. The pane is handed
a JSON STRING rather than a reactive object, so a later change to the OUTER
element's profile-config updated the composite's own cfg and reached nothing
below it. regions-dashboard is kept alive by v-show rather than remounted, so the
one code path that would have rewritten the attribute never runs again and the
panes stay in the old language for the whole session.
The leaf was never at fault: ProfileLoader's config is a computed over its prop,
lang is a computed over config, and a watchEffect drives appI18n on every change.
Every link is reactive -- it simply never receives a new prop.
library/composables/usePaneConfig.js owns the missing half: a composite registers
each pane element with a getter for its config, and a watchEffect re-forwards
whenever any reactive source that getter reads changes. It writes only when the
serialised config actually differs, because setting profile-config re-parses it
in the child and can remount what it drives -- an unconditional write per tick
would be a rerender loop. Stale elements (a rebuilt pane column) are skipped.
It is a library helper rather than three copies because regions-mfe, upg-100-mfe
and country-posters-mfe all assemble a config and set the attribute once; they
differ only in which keys they inherit, so each keeps its own
INHERITED_PAGE_KEYS and hands the result to the helper. This commit adopts it in
regions-mfe only; the other two are upg100's to adopt.
Two-sided, headlessly: a check that fails when a composite writes profile-config
on a child without re-forwarding names all THREE composites on development, and
on this branch regions-dashboard reports "re-forwarding=usePaneConfig" while the
other two still fail.
NOT VERIFIED IN A BROWSER YET. rega's LIVEDIAG is the falsifier: the nested
[profile-config] should show the OLD locale on development and the NEW one here.
country-map.vue's poster-wall wiring put the poster table, peopleGroupRecords
and the sheet code into the bundle of EVERY DX Life country page, for a wall
that is hidden by default. Measured: country-map.js 391,475 bytes on
development against 683,401 here -- +291,926, a 74% increase -- and with it
removed the bundle is 391,625, back within 150 bytes of development.
Making the three imports lazy does not help, and I measured that rather than
assuming it: with all three behind await import() and the table fetched at
mount, the bundle grew to 684,700. vite.config.js:282 sets
inlineDynamicImports, so a dynamic import is flattened straight back in and only
the wrapper is added. Removing the weight needs the pinned-served-asset pattern
that poster-render.js already uses -- card
PERF-LOAD-THE-PRINTING-SUBSYSTEM-ON-DEMAND.
library/packet/countrySheet.js stays: it is unreached at runtime (0 hits in the
built bundle) so it adds nothing, and it is the library half this element will
use once the printing code loads on demand.
ProfileLoader's comment now names both populations -- 20 of the 25 bundles are
single-profile, 10 of the 17 that actually mount this loader, which are the only
ones the default affects.
Resolving package.json conflicts by unioning unique lines is right for a file
that holds one entry per line. Applied to ordinary source it duplicates
declarations, because both sides of a hunk keep their copy and nothing
de-duplicates across the join.
Eleven plugin entry points ended with the same import twice, and the poster
renderer with two identical `const sheet`. Each is a SyntaxError: every one of
those bundles would have failed to build, and the poster export would not run.
Found by parse-checking every changed .js against the merge base, not by the
gates -- gates:pre passed 101 of 102 with all twelve files broken, because its
parse gates do not cover plugin entry points.
One measurement trap worth recording. `node --check` on a .js file parses it as
CommonJS, so a bare `import` is a syntax error and ALL 28 entry points appear
broken, on development too. That reading says the tree is on fire and is simply
wrong. Copying each file to a .mjs and checking that parses it as a module,
which separates the eleven that are genuinely broken here from the rest that
are fine.
The handoff merges brought in three references that name the Driver in shipped
text: two comments in CenteredSetTarget.vue and one _why in the people-groups-grid
instances. Reworded to state the engineering reason without the speaker -- an
open decision now reads as pending a product decision, which is what the rule
asks for. The instances file still parses as JSON.
Three lines brought in with the list-pattern doc cited the Driver; the
reference gate refuses them. The rule, the scope and the open exception
stay as stated, without the attribution.
Merging gospel/a5-cpu and gospel/panel-followup brought back two comments that
quote the Driver directly, and check:no-new-driver-references failed on them —
correctly. Development had already swept these; the merge reintroduced them from
an older base.
Both keep their reasoning and lose the attribution: a recording pill that
changes width on every word reads as instability, and on a phone the header can
take most of the panel, leaving the draft unscrollable and its last lines
unreachable. Those are the facts the rules encode; who said them is not.
The consolidation brought back two Driver citations from older branches that
the reference sweep had already cleared from development. Both keep their
engineering reason and lose the speaker: a gate's rationale for existing, and
the mechanism of the print collapse.
Caught by check:no-new-driver-references, which is the point of running the
full chain rather than the seven gates the merged package.json had left.
Each of these shadow roots is position:absolute; inset:0, while its :host had
display:block with no position. An absolutely positioned element resolves against
its nearest POSITIONED ancestor, and an unpositioned host is not one -- so the
root fell through to whatever the partner page happened to have, or to the
viewport. It only looked right standalone, where the page fills the viewport
anyway, and inside a composite's .pane, which is itself positioned. A partner
embedding the element directly gets it painted outside its box.
All three are the shadow root's own root element, with only comments between
<template> and them, so none of them has a positioned ancestor to fall back on:
regions-list.vue:408 .rl-root at :409
region-dashboards.vue:733 .rd-root at :734
regions-mfe.vue:194 .rmfe at :199
dx-shepherd's census named the first as confirmed and the other two as
candidates to check. Checked: both are roots, not children of something already
positioned, so all three are real.
The fix is the house pattern -- position:relative on :host -- matching the
already-fixed cities-internet/connectivity-map byte for byte in shape, and
prayer-walk-planner.vue:628.
One line changed per file. Verified in the ARTIFACT rather than the source: the
rebuilt regions-dashboard.js carries all three corrected host rules (two at
min-height 320px, one at 420px). check:sfc-parse, check:synced-css and
check:one-scroll-container pass; the manifest diff is integrity hashes only,
0 other lines.
country-map.vue fetched /api/people-groups/list and mapped the posts straight to
points, without applyPeopleGroupList(). Every other consumer of that endpoint
calls it -- DataSourceManager, useDashboardData, useCsvExport, peopleGroupRecords
and the people-groups grid. So a host that asked for one curated list got the
WHOLE endpoint drawn on its country map, and nothing said so: the pins looked
like data rather than like the wrong list.
The list is applied to the raw posts, before they become points, which is where
every other consumer applies it. That is what keeps this element and a dashboard
showing the same country from disagreeing about which groups are in it -- the
same reason the country filter beside it is client-side.
The default list returns the rows untouched and makes no fetch, so an
unconfigured host pays nothing. A list whose membership cannot be loaded yields
NO groups rather than the wrong ones: applyPeopleGroupList's own contract, and
the safer failure for a prayer surface.
Measured three ways, with a temporary copy of check:universal-props that adds
country-map to its BUNDLES list (the reason the gate never saw this: 19 bundles
listed, this one absent):
- RED on the base: BYPASS country-map LIST-FETCH country-map.vue:98, exit 1;
- GREEN here: 20 bundles read engine, token, locale and list through the
library, exit 0, the gate's 9 planted-bypass self-tests still caught;
- a real list value changes the drawn count, through the shipped
applyPeopleGroupList and the served membership files: Nepal draws 34 points
on the default list, 29 on "frontier", 0 on "frontier-500k". That last zero
is data, not a failed load: the same membership matches 2 India rows.
Rebuilt on this base rather than carrying the earlier bundle, because
development rebuilt the same files in between and generated output is
regenerated, never merged. The list-membership URL is present in the built
bundle; the manifest diff is integrity hashes only, 0 other lines.
Two defects, one cause: on a short panel the engine-offline notice took the room
the draft needed.
FIRST, A PHONE RULE THAT NEVER RAN. bac2431d0 (19:59) put notices below the
draft "so the draft stays in view". 4e94e9216 (21:58) added the corner widget
with `.compact .notice { order: 0 }`. A media query adds NO specificity, so
`.compact .notice` (0,2,0) beat `@media { .notice }` (0,1,0) at every width and
the phone rule has been dead since the day it landed. Ordering the compact rule
to 1 fixes the widget at every width, not only under 480px, because the corner
panel is short regardless of the viewport.
SECOND, ORDER ALONE WAS NOT ENOUGH, which I only found by doing the arithmetic
rather than trusting the first fix. .notice is flex:0 1 auto and .compact
.compose is flex:1 1 0, so the draft yields whatever the notice's content asks
for no matter where it sits. The offline notice is ~132px at 375 and ~179px at
390 — reference text a reader needs once, holding a third of a short panel
forever. Capping it at 96px with its own scroll gives the draft 58px at 375x450
and 124px at 390x560, against rega's measured 22px and 41px.
The panel is 60vh (plugin-studio.vue:466), NOT the viewport. Sizing off the
viewport is what made my earlier estimate wrong by 58px; the model that uses the
real panel reproduces rega's 22px exactly.
No floor anywhere: a floor is what clipped the draft to begin with.
(cherry picked from commit 79cbb41bc3)
Same defect as cities-internet's list and table: no :host rule at all, so the
element falls back to display:inline with no height while .gt is
position:absolute;inset:0 — which then resolves against whatever the partner
page has positioned.
TWO MOUNTS, ONE RULE, and they do not collide:
- STANDALONE (data-tables.meta.json) this is its own custom element, so :host
applies and .gt is finally contained.
- COMPOSED, gospel-dashboard renders <DataTables/> as an inline Vue component
with no shadow host of its own, so :host matches nothing there. Its
'.gd-tables .gt{position:static;…}' override (gospel-dashboard.vue:81) keeps
working, which is what lets the stacked section take its natural height and the
page scroll. That override records a MEASURED failure — docScrollHeight 900
against clientHeight 900, a wheel gesture moving 0px — and must not return.
min-height, not height alone: with no height, .gt's inset:0 would give the table
0px in an auto-height box, so it would stop escaping by becoming invisible. A
floor gives an auto host something to grow from; a definite-height slot still
decides through height:100%.
NOT the ProfileLoader 100dvh floor (ProfileLoader.vue:410), which is the separate
upstream cause of the connectivity-map overflow and is studio's card. .gt is not
that wrapper, so this fix stands on its own.
(cherry picked from commit 10f2a5c13f8d08c0086ada8eafa4ad819863bd0a)
Neither element had a :host rule at all, so each fell back to the custom-element
default — display:inline, no height of its own — while its root div carried
position:absolute;inset:0. Absolute positioning resolves against the nearest
POSITIONED ancestor, so on a partner page both resolved against the PARTNER's
layout rather than against this element.
rega measured it on a partner page with a 420px host box: connectivity-list
painted to y=16,799 and data-tables to y=1,698 — 16,355px and 1,254px below the
box, over whatever the partner had underneath.
The same fix as connectivity-map's (60cb4ff18), for the same reason:
:host{display:block;position:relative;width:100%;height:100%;min-height:320px}
display:block gives the element a box; position:relative makes that box the
containing block; height:100% lets a definite-height slot decide; min-height is
the floor so an auto-height host grows to fit instead of collapsing.
NO NEW SCROLL CONTAINER WAS NEEDED. .ci and .ct already carry overflow-y:auto —
the scroller existed, it simply had no host to be bounded by. Once contained,
the list and the table scroll inside the box, which is what an embed should do
in a definite-height slot.
min-height is 320px here, not the map's 420px: regions-list.vue:408 is the house
pattern for list profiles, and a list is legible shorter than a map is.
(cherry picked from commit 08ad95fa86e64dfe043befb156e358c91319e511)
country-map.vue fetched /api/people-groups/list and mapped the posts straight to
points, without applyPeopleGroupList(). Every other consumer of that endpoint
calls it -- DataSourceManager, useDashboardData, useCsvExport, peopleGroupRecords
and the people-groups grid. So a host that asked for one curated list got the
WHOLE endpoint drawn on its country map, and nothing said so: the pins looked
like data rather than like the wrong list.
The list is applied to the raw posts, before they become points, which is where
every other consumer applies it. That is what keeps this element and a dashboard
showing the same country from disagreeing about which groups are in it -- the
same reason the country filter beside it is client-side.
The default list returns the rows untouched and makes no fetch, so an
unconfigured host pays nothing. A list whose membership cannot be loaded yields
NO groups rather than the wrong ones: applyPeopleGroupList's own contract, and
the safer failure for a prayer surface.
Measured three ways, with a temporary copy of check:universal-props that adds
country-map to its BUNDLES list (the reason the gate never saw this: 19 bundles
listed, this one absent):
- RED on development: BYPASS country-map LIST-FETCH country-map.vue:98, exit 1;
- GREEN here: 20 bundles read engine, token, locale and list through the
library, exit 0, the gate's 9 planted-bypass self-tests still caught;
- a real list value changes the drawn count, through the shipped
applyPeopleGroupList and the served membership files: Nepal draws 34 points
on the default list, 29 on "frontier", 0 on "frontier-500k".
Rebuilt, tree regenerated, SRI refreshed. The list-membership URL is present in
the built bundle; the manifest diff is integrity hashes only, 0 other lines.
The row meta joined its first three languages with English commas and cut the rest silently, so a truncated list read as complete. It now goes through formatList (Intl.ListFormat) as a conjunction whose LAST item states the cut, from keyCities.list.moreLanguages ("{n} more", landed in R29): "Hindi, Urdu, Bengali, and 2 more" / "Hindi وUrdu وBengali و2 أخرى" / "Hindi、Urdu、Bengali和另外2种". Three or fewer languages show as they are. Not 'unit': in zh it joins names with no separator, and in ar and hi it still adds a conjunction (studio, measured).
join(', ') is English list grammar. The two user-facing sites now go through the library's formatList (Intl.ListFormat, R29), as conjunctions because each is the FULL list:
- PPLR's Rosetta popup, "Not yet cross-mapped in PPLR: {ids}";
- UPG Neighbor's bridge-language cards, "Believers in: {list}".
Key Cities' shortened language list follows once its "{n} more" key (keyCities.list.moreLanguages) is on development. The joins that stay are not reader lists: a thrown error's details, the bridge-languages CSV columns, and a gate's console line.
Two reader-visible lists were joined with a hardcoded ', ': the ADM1 names
in the selection error ('no people-group data for {names}') and the
selected territories in the batch HUD. Both use formatList now (R29), so
de reads 'Gujarat, Rajasthan und Odisha', zh 'Gujarat、Rajasthan和Odisha',
ar 'Gujarat وRajasthan وOdisha'. The locale is passed from the i18n ref
(packetLang), not left to getRuntimeLocale, so the HUD's list re-renders
on a runtime language switch. formatClipboard's join stays: its only
caller is a test handle.
Three places joined names with ", ", which is a list only in English. Arabic
separates with a word, Chinese uses a different mark entirely and no comma at
any position, and German and French put a word before the last item. The
separator and the final connector are the grammar, so no catalogue key can fix
them from outside.
Two of the three carried a second, worse error. A multi-value filter holds
ALTERNATIVES -- the bus matches a row if ANY of its values matches, in
useSelectionBus with some() and in the data table with includes() -- so the
chip means "Islam or Hinduism". Joined with a comma it reads as "and" in every
language with an explicit connector, which states the opposite of what the
filter does. Those two become a disjunction.
The third is the screen-reader live region, which names the facets currently
applied. Those hold together and narrow the set in turn, so it is a
conjunction. It is also the one string here that some readers rely on entirely,
having no chips to glance at.
The composable is the single owner: every surface that turns a filter into
words comes through it, so the chip bar, the research-filter chips and the CSV
provenance line now say the same thing in the same language.
Left alone deliberately: five joins in this plugin's gate scripts. Those are
failure messages read in a terminal by whoever runs the gate, not text shown to
a reader, and formatting them per locale would help nobody.
Source only. This needs the bundle rebuilt before it ships.
formatList (R29) replaces join(', ') at the two sites that a reader sees. The
separator was never a missing key — no key can fix punctuation between
interpolated values — which is why both survived every i18n gate.
connectivity-map.vue:365, CONJUNCTION. The Indian states with no boundary are
ALL missing one. The names land inside a translated sentence at :595, and the ar
message reads "لا توجد بعد حدود ولاية لـ {names}: …", so English commas were
being printed inside Arabic prose.
GiCombos.vue:49, DISJUNCTION, decided from the matcher and not from what reads
better: useSelectionBus.js:299 tests cur.includes(value) and toggleFilterValue
adds to the list, so two values on one facet WIDEN the result set. "Africa and
Asia" would state the opposite of what the filter does. This changes the English
too — "Africa, Asia" becomes "Africa or Asia" — which is a gain, because a comma
silently reads as ALL.
Measured, the same three names and two facet values through each type:
en "Ladakh, Puducherry, and Chandigarh" "Africa or Asia"
ar "Ladakh وPuducherry وChandigarh" "Africa أو Asia"
zh "Ladakh、Puducherry和Chandigarh" "Africa或Asia"
was "Ladakh, Puducherry, Chandigarh" "Africa, Asia"
ar joins with و and no comma at all; zh uses 、. Neither is reachable by keying.
Not converted, and deliberately: useBoard.js:126 is a console.warn for
developers, and rudder-voice.vue:971 is in a component with zero i18n imports and
no catalogue area — English by construction, so it converts with the rest of
Rudder if Rudder is ever localized.
Source only. The rebuilds are regenerated after this rebases onto landed R29.
