D-TH.6, ejecutada. El slot de tinta pasa a `fg` y el modificador interactivo
se pone delante, que es lo que theming §6.7 r7 documentaba sin guard desde
que se escribió. Value-preserving: renombra la clave en `recipes/base.ts` y
sus 793 referencias en 101 ficheros; no toca un solo valor.
claves renombradas 269 en 66 componentes
referencias 793 en 101 ficheros
censo antes/después 162 recetas · 5.203 knobs · 1.841 públicos (37 %) · 56
sin contrato — IDÉNTICO, como debe ser un rename
--names antes/después 269 desviadas → 0
Qué NO entra, y por qué:
47 `{rol}-{slot-de-rol}` canónicas: COLOR_ROLE_SLOTS pone el modificador
detrás POR CONSTRUCCIÓN (`primary-solid-hover`)
13 exentas firmadas `--focus-ring-*` es familia del sistema; el color
de `aura` es un sustantivo; `stop-color` ES una
parte de gradient-builder
47 hovers neutros por VALOR: migran a la capa de estado
(§38 + R-4.3), no se renombran — firma 3
6 ocurrencias en historia changelog, errores-toxico, PLAN-affix/background:
reescribir un registro fechado lo vuelve mentira
El clasificador vive en el censo (`--names`), no en un script suelto, para que
el guard R-5.3 consuma la MISMA gramática que el codemod. Su muta-prueba
(`__names-mutatest.ts`, 27 casos) es lo que hizo el trabajo: cazó que yo
promovía al frente CUALQUIER valor declarado por el morfo, y así
`--sidebar-width-icon` (la anchura del raíl colapsado) se convertía en
`--sidebar-icon-width` (la anchura de un icono), que es otra cosa. La firma
dice «delante lo interactivo, detrás lo dimensional y contextual»: ahora sólo
promociona el vocabulario interactivo cerrado, y lo contextual —`below`,
`loaded`, `vertical`, `icon`— se queda donde estaba. Los cinco casos de
regresión están en la muta-prueba.
También cazó que `dropdown-menu.item-bg-hover` lee `var(--color-primary-element)`:
es un hover CON VALENCIA (el palette swap de recipe-contract §2), no el hover
bespoke que §38 deprecó. Clasificar por el nombre lo habría metido en una
migración que no le toca; se clasifica por el VALOR.
Verificación (§7.4, artefacto por paso):
diff de generated/ 269 renombres 1:1 · 0 cambios de valor · 22 privados
reapuntados a los nombres nuevos (`__names-verify-diff`)
computed tag-group · field · tabs: 6.467 valores en 23 estados,
0 diffs — y comprobado en el navegador que sirve el CSS
nuevo, para que ese 0 no sea el de una copia cacheada
huérfanos 0 en código
suite eidos 434 pasan · 1 rojo, el conocido (`skin-media-player`).
`recipe-css-contract` verde: es el guard que caza el
`var()` sin fallback a un nombre que ya no se emite,
o sea el fallo exacto de un rename a medias
component:audit 163 PASS · 3 NEEDS-WORK, los tres SIN TOCAR por esto
check 0 errores en ficheros tocados (los 73 globales son de
otras sesiones de la rama; se atribuye por fichero)
rtl:check 0 · docs:check 0
formato el renombrado no alarga ninguna línea: ningún fichero
tiene más líneas de 100 chars que antes, así que no se
pasa prettier — hacerlo reformateaba 300 ficheros de
deriva ajena
Entra aquí la corrección del repaso de los 7 ya hechos:
`gradient-builder.checker-color` → `checker-fg` (el damero de transparencia;
ahí `color` era slot). Sus `stop-color-*` no se tocan.
Queda para el paso siguiente: R-5.3 no puede graduar a `error` directo
mientras los 47 hovers neutros sigan hablando el idioma viejo — o migran
antes (firma 3, ya firmada), o el guard necesita una exención greppable.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
El morfo declaraba emerge-present / emerge-dismiss / emerge-dismiss-escape y
nadie los emitia: el SILENT del pack nunca casaba, la puerta de temas que su
propio docblock documenta no podia funcionar, y el preset present-rise que
declara `event: ['emerge-present',...]` jamas pintaba. La excepcion del censo
lo decia desde agosto: «Today's silence is accidental, not the declared
design».
El re-analisis pedido por el autor cazo lo que la primera propuesta no vio:
la firma present-rise es GLOBAL (base.css:6520), la entrada por data-state es
el workaround de la emision ausente (el comentario del provider lo dice
literal), y la semantica autorizada distingue hover (entrada) de focus
(instant-open, «appears with no entrance — its semantics») — cosa que el
evento no distingue. De ahi la opcion B firmada: el estado conserva la
entrada; la emision añade la superficie que faltaba.
Cableado con la receta F4: `present` pre→post — y SIN su `commits` (fijaba
'delayed-open' y pisaria el 'instant-open' del camino focus; los overlays
binarios conservan el suyo porque su valor es total, desviacion declarada);
dismiss ×2 ganan `targetFallback: [trigger]` (Presence sostiene el content por
la salida; el fallback cubre la carrera). Provider: emision SOLO en
transiciones reales — un open-timer cancelado no presenta nada que retirar, un
re-hover abierto no re-emite — y `handleClose` lleva la causa para distinguir
el Escape.
MEDIDO en navegador, los tres caminos y la fuga exacta:
hover present CONTENT · entrada autorizada corriendo (scale-in+fade-in),
present-rise NO — el preset gana la cascada; miedo al doble
movimiento REFUTADO en el camino delayed
focus present CONTENT con instant-open y present-rise CORRIENDO — la fuga
predicha, confirmada → silencio visual MINIMO en la receta, scoped a
[data-state='instant-open'][data-event-phase='active'];
re-medido: anims [] y la entrada delayed intacta
leave dismiss CONTENT via Presence · salida autorizada (scale-out+
fade-out), dismiss-fade no
Escape dismiss-escape CONTENT · desestampado del hold a ~240ms
Deudas limpiadas en el mismo commit (el guard lo exige): 3 ids de
INERT_EVENT_DEBT + 3 excepciones de EMISSION_EXCEPTIONS. Quedan 4 de la parte
(ii): virtual-list/grid ×3 y gradient-picker commit-reset (S-14).
Verificado: soma navegador 1278✓/1 (el timeout ajeno) · tooltip 4/4 ·
sema+morfo+contracts 552✓/6 ajenos · check 69 = base aislada, diff VACIO ·
prettier: morfo/tooltip.ts limpio en HEAD queda limpio; los 3 avisos ya
fallaban en HEAD.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Fable audit follow-through (fable_audit.md + fable-eidos-audit.md):
- RECIPE_CONTRACT.md (E2 canon): the transversal systems every recipe must
consume, enforced by component-audit R-4.1-4.6 (all at error; escape
valves /* literal */ + /* functional */; WIP tracks excluded). Stale
audit rules fixed against the current architecture (E-2.2 wrapper
imports, D-1.2 v2 9-tab union, D-3.1 single snippet, TabsVariant
mirror) - verdicts went 0/117/15 -> 75/50/5.
- Motion migration 15/15: recipes off local @keyframes onto the channel -
preset stamps (dropdown/context/select/combobox/tooltip/link-preview/
clipboard), new expand/collapse + value-flash signatures, shared-axis
reverse pair, delayed-open open-alias in the preset trigger
(PRESET_STATE_ALIASES), materials pattern for irreducible triggers
(card/timeline/tabs/nav-menu/metrics). Duration/scale hooks keep every
recipe's tuned values.
- buildScheme (fase C): full a1..a12 alpha ramp per role (was a2/a3 -
stale alphas after applyColorScheme), intentSeeds so temper starts from
the ACTIVE theme's intent mapping (risk stays orange), alpha background
self-derived from the scheme's own neutral step 1 (was hardcoded
#fff/#111); mode now forces the donor variant.
- Inventory decisions (fase D): semanticTracking axis removed (all-zero),
border-hover slot dropped (0 consumers), separator slot adopted across
line dividers (step 6, Radix divider tone), size-bundle consumption
pilot on toggle (canonical coordinates consumed, deliberate deviations
kept visible).
- Docs: building-a-component.md (the one door, 9-phase route + known
traps), PLAN-docs-reconciliation.md (fase 6 kickoff for a fresh
session), THEMING wiring updates.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A perceptible line showed across the caret base on every side but bottom: the
panel's border draws across the caret's base edge, "cutting" it off from the panel.
Add the popover's per-side fix — pull the caret base into the panel by the border
width (margin toward the panel, keyed on data-side) so the fill covers that border
and the caret's side strokes continue the panel outline seamlessly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A single-direction drop-shadow on the caret is wrong: the caret sits at the panel
edge pointing AT the trigger, so on a bottom-side tooltip (caret at the top) the
downward shadow falls onto the panel below it — a dark smudge over the content.
Remove it. The caret keeps the panel-bg fill + border edges (no black blob, no
smudge). A shadow that reads correctly would have to wrap the whole panel+caret
silhouette (content-level drop-shadow), not the caret alone.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Follow-up to the un-clip fix (c3def5d6). With the caret no longer clipped it was
visible, but it rendered in `currentColor` (the dark text colour) — a black triangle
on a light panel. Colour it to the panel bg (a seamless caret) + the panel border on
the two exposed edges, and give it its own `drop-shadow` so the matched (light) caret
still reads against the page — the panel's box-shadow is rectangular and never reaches
the caret. This is the "matched + visible" caret (the drop-shadow-on-the-silhouette
idea) now that overflow:visible lets it show.
VERIFICATION CAVEAT: confirmed the colour at the computed-style level (polygon fill ==
panel background, was the dark text colour). Could NOT capture a stable screenshot —
the demo tooltip closes on every programmatic/cursor interaction, so the visual pop of
the drop-shadow is unverified here. If the matched caret reads too faint, bump the
drop-shadow alpha or fall back to the darker caret.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The tooltip content is `position: relative; overflow: auto`, so the caret
(`position: absolute`, containing block = the content) that pokes past the panel
edge was CLIPPED away — invisible regardless of its colour. Confirmed by measuring:
the caret sits fully outside the content box, and the content clips its overflow.
Change the content to `overflow: visible` — the same reason the popover content is
`overflow: visible` (its comment literally says "fixes the arrow that overflow:auto
used to clip"). Tooltips are short, so no inner scroll viewport is needed.
Verified in a real browser: the dark caret is now clearly visible pointing at the
trigger. Combined with a40d1573 (caret colour back to currentColor) + the earlier
`display: block` (correct arrow height → correct trigger gap), the tooltip arrow is
visible and correctly placed again.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
My previous commit (4afadfd7) over-corrected: it changed the caret fill from
`currentColor` (the panel's dark text colour — long-standing, visible) to the
panel background, aiming for a "seamless extension". On a light-on-light tooltip
that makes the caret INVISIBLE — the panel reads only because of its box-shadow,
which the separate caret element doesn't share. The user had a visible caret
before; the "correct" colour regressed it.
Revert the polygon/path colour rules (caret returns to `currentColor` = visible)
and keep ONLY the `display: block` on the SVG — that was the fix that mattered: it
removes the ~21px of inline-baseline phantom height that both misaligned the caret
and, because the positioner adds the measured arrow height to the gap, pushed the
panel too far from the trigger. So: caret visible again + panel sits at the correct
~13px gap. A properly matched-and-visible caret (bg fill + its own drop-shadow) is
deferred until it can be verified.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The tooltip arrow SVG carries `fill="currentColor"` / `stroke="currentColor"`
presentation attributes, so setting `fill`/`stroke` on the wrapping `<span>`
(`[data-tooltip-arrow] { fill: … }`) never reached the shapes — an element's own
presentation attribute beats an inherited value. The caret therefore painted with
`color` (the dark text colour) instead of the panel background, and the `<svg>` was
left `display: inline`, adding phantom baseline height (26px box for a 5px caret)
that misaligned it. Net effect: the caret didn't read as a caret.
Fix mirrors the popover recipe: colour the `polygon` (fill = panel bg) + `path`
(stroke = panel border) directly, and set the `<svg>` to `display: block`. The
tooltip caret is now byte-identical in treatment to the popover's (fill = content
bg, 1px border-coloured edge). Also fixes the outline variant, which had the same
span-vs-shape mistake.
Pre-existing bug, unrelated to the avoidCollisions:'flip' rollout. Verified in a
real browser: polygon fill now equals the content background, svg is block, caret
positioned correctly.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
24h picker driven by a horizontal daylight gradient — drag the sun/moon
thumb across the night→day→night sky, jump with moment chips, or nudge
with hour/minute/second steppers. Inline panel + trigger/popover variants,
built on the generic Picker (deferValue) composing Slider, Button,
PickerShell, Popover and Tooltip.
- granularity (hour|minute|second) + optional band props; steppers keep a
fixed 3-column grid; shape/size propagate to panel, trigger and chips.
- fixed-tone sky palette as named recipe tokens (theme-independent).
- fix soma Tooltip.Arrow (empty children slot hid the default SVG) and
make PickerShell footer labels locale-reactive.
- floating-gap canon: --floating-gap-menu 0 -> var(--space-1); FloatingContent
reads --floating-gap via a reactive $derived (was a one-shot rAF that never
reached portaled menus -> they fell back to sideOffset, flush); nav-menu wired
to the token via CSS.
- focus-ring: exclude item/option/content from the universal archetype ring; menu
rows use the canonical highlight on :focus-visible too (no thick keyboard border,
no ring framing the whole float).
- overscroll-behavior: contain on the ScrollArea viewport + 7 overlay scroll regions
(no scroll chaining to the page).
- archetypes.css: default rules wrapped in :where() so component recipes always win.
- radius decoupled from size (Radix model) in toggle/textarea/menubar/nav-menu/
virtual-list/grid; toggle `rounded` prop.
Canon-read + token reach portaled menus pending a fresh-load check (the dev tab
cached the floating module) -- see CONTINUE-floating-gap-canon.md.
The tooltip's `delayed-open` entrance (the per-side slide, and now the bounce) NEVER
fired: the content only mounts when `open` becomes true, and the hover-delay timer
set `internalTooltipState = 'instant-open'` AT THAT MOMENT — so the content always
mounted as `instant-open` (no entrance). `delayed-open` was only ever the pending
state while the content was still unmounted. A long-standing dead animation.
Fix: the hover-delay timer now keeps `'delayed-open'` as the content mounts, so the
eidos `[data-state='delayed-open']` entrance plays. `'instant-open'` stays reserved
for focus / skip-delay (intentionally no entrance). Test updated (it asserted the
buggy `instant-open`); tooltip provider suite 4/4.
Also makes the bounce clearly perceptible (it was too subtle to notice even when
forced): scale 0.5 → 1.18 (overshoot) → 0.9 (undershoot) → 1.06 (second bounce) → 1,
over `--duration-slow`. Verified end-to-end in-browser: hover → content mounts
`delayed-open` → `animation-name: eidos-tooltip-bounce-in`.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replaces the tooltip's per-side 2px slide-in with a scale BOUNCE: grow from 0.7,
overshoot to 1.06, dip to 0.97 (the undershoot — the "rebote"), settle to 1, over
`--duration-moderate`. It scales from `transform-origin` (the floating origin,
toward the anchor), so it keeps a directional feel without per-side keyframes —
8 keyframes collapse to 2.
Bespoke recipe keyframe, NOT a registered `motion` preset: the tooltip's state is
3-valued (`delayed-open` / `instant-open` / `closed`), not the generic `open` /
`closed` the shared presets key on — so a registered preset wouldn't fire here.
This also confirms the architecture boundary: components whose state shape doesn't
match open/closed use a bespoke recipe animation, not the `motion` selector.
`instant-open` (keyboard focus) keeps no entrance; the reduced-motion block (kills
the animation) already covers the new rules — verified the bounce curve in-browser
(scale 0.7 → 1.06 → 0.97 → 1). Recipe-only change, no generator/base.css touched.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Closes the second BROKEN component from the audit (after date-range-picker).
Five errors collapse to zero:
- Morfo declares the three `emerge` events the recipe animations were
already keying off (`open`, `close`, `close-dismiss`) plus a
`texts.label` slot. `data-disabled` is now declared on `Trigger` (with
optional severity) so the recipe rule has a real contract to land on.
`aria-disabled` and `texts.label` round out the surface; the new
`tooltip.label` catalog entry ships under
`src/uix/langs/components/tooltip.ts`.
- Recipe CSS: explicit `[data-tooltip-trigger][data-disabled]` and
`[data-tooltip-trigger]:focus-visible` rules. The disabled rule
suppresses pointer events; the focus rule provides a defensive
outline so a focusable trigger shell is at least keyboard-visible.
- README: convert the old `## Pendiente deliberado` section into a
proper `## Gaps` table with explicit dispositions
(implementar/diferir/descartar), and update `## Eventos Sema` to
reflect that tooltip now declares its three emerge events instead of
claiming zero. Comparativa was already complete; the regex fix from
the previous commit picked it up.
Audit: tooltip BROKEN → PASS. Project-wide goes from 1/64/2 to 3/64/0
(PASS / NEEDS-WORK / BROKEN). No more BROKEN components.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
dev