ink-divergences.md used to record only divergences from Ink. Broaden it into the
single record point for the *whole* Ink relationship: it now also records deliberate
alignments (intentional sameness) as first-class entries, not just differences.
- Reframe the title + intro: three relationship kinds (deliberate alignments,
intentional divergences, non-behavioral notes), each a conscious decision.
- Elevate + strengthen the governing principle: aligning to Ink is only a means to
reduce bugs, never the goal — correctness and Vue philosophy outrank parity, and
"because Ink does it" is never on its own a justification.
- Add a first-class "Deliberate Alignments" section; promote the two deliberate-
sameness records (commit-timing Ink-alignment, literal-tab measure-vs-paint, the
latter keeping its existing [VOUCHED @hyf0]) out of Non-Behavioral Notes into it.
- Restructure the classification flow to split deliberate match vs deliberate
divergence first ("How to Classify an Entry").
- Update AGENTS.md guidance and the two stale cross-references to the renamed
headings (accessibility-api.md + the in-file ARIA entry).
No existing divergence entry's substance changed; no [VOUCHED] stamp was added or
removed (AI cannot self-bless). Reviewed by Codex + an independent reviewer.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The note was overlong. Keep the essentials: string-width counts a tab as 0 but
paint expands it to the tab stop (ab\tcd measures 4, paints ~10); Ink-shared so
KEEP aligned; fix-if-needed pointer (expand at squash, upstream of string-width).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A literal tab in <Text> is measured as 0 columns by string-width but painted at
its tab-stop width (wrap-ansi / terminal), so the reserved yoga width disagrees
with what's drawn (ab\tcd measures 4, paints ~10). This is a shared upstream
quirk — Ink v7.0.4 does the same and likewise doesn't normalize tabs — so it's
aligned with Ink, not a divergence; recorded under Non-Behavioral Notes so it's
not rediscovered as a parity gap. KEEP (literal tabs in TUI text are vanishingly
rare). The note also captures the fix direction if ever needed (expand tabs to
spaces at the shared squash chokepoint, upstream of string-width) and the one
behavior change that would then become the actual divergence. [VOUCHED @hyf0]
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Encode the declarative invariant for the runtime `wrap` re-measure fix
(PR #193) as a full matrix: for all 6 wrap modes and all 30 ordered
transitions, toggling `wrap` at runtime produces the exact same frame as
a fresh mount with that wrap (measure == paint). Ground-truth fresh-mount
frames are derived at runtime, not hardcoded. Reverting the one-line fix
in node-ops.ts turns 16 of the 30 transitions red, so the matrix
genuinely guards the fix.
Vouch the divergence: add [VOUCHED @hyf0] to the ink-divergences.md entry
and reword it to lead with correctness (Ink v7.0.4 has the latent stale
measure bug; vue-tui keeps the correct invariant). Drop "pending a human
vouch" from the node-ops comment.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
assertBoxValid (Box) and text.vue's validate() run eager render-time validation
of paint-time VISUAL props (backgroundColor, border fg/bg colors, borderStyle
shape) and throw into the error boundary on an invalid value (e.g. a chalk
modifier name like "bold" used as a color). They were gated only by the per-node
ariaHidden skip (srHidden), not by GLOBAL screen-reader mode.
Under global SR mode (isScreenReaderEnabled; INK_SCREEN_READER=true) vue-tui,
like Ink, linearizes the whole tree to PLAIN TEXT and never colorizes / draws
borders for any node — Ink's colorize path is bypassed entirely, so it never
throws on an invalid color. vue-tui still ran the eager validation for non-
ariaHidden boxes under SR and threw, crashing a screen-reader user out of
accessible content over a paint-only prop value.
Skip the eager visual validation when global SR is on, in addition to the
existing per-node srHidden skip: box.vue gates `!srHidden && (srEnabled ||
assertBoxValid(props))`, text.vue gates `!srHidden && (srEnabled || validate())
&& hasContent`. The validation is all paint-time visual input (no structural
checks), so skipping it under SR is safe and matches Ink.
Verified against real Ink v7.0.4: with INK_SCREEN_READER=true a
<Box backgroundColor="bold"> renders plain text and does NOT throw; without it
Ink throws in colorize.js. This is an alignment fix (removes a vue-tui
over-throw), not a new divergence — the existing ink-divergences entry gets a
factual, unstamped note about the SR carve-out.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The `wrap` prop changes a <Text> node's MEASURED height (the yoga measure
func reads el.props.wrap to pick wrap/truncate/hard layout) but is NOT a
yoga prop, so a runtime wrap-only change took the generic STYLE_PROPS
branch in patchProp: it stored the new value into el.props and called
onCommit() WITHOUT markTextDirty(el). Yoga kept the OLD wrap mode's cached
height while paint rendered with the NEW wrap, so layout and paint
disagreed -- stale blank rows on wrap->truncate, overflow / overwritten
siblings on truncate->wrap.
Mark the text node dirty when the changed STYLE_PROP is `wrap` on a
tui-text node so yoga re-measures. `wrap` is the only STYLE_PROP that
affects measured dimensions (the rest are paint-only), so it is the sole
case.
Verified Ink v7.0.4 has the identical latent bug -- its applyStyles
ignores textWrap and never markDirty()s, so a wrap-only change goes stale
there too. Recorded as a blessed divergence in ink-divergences.md; the fix
matches the layout Ink produces whenever its measure func is invalidated.
Tests (text-wrap-remeasure.test.tsx) reproduce both directions:
RED produced Ink's stale frame ("aaaa …\n\n\nZZZZ"), GREEN the correct
re-measured layout.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
yoga's getComputedWidth()/getComputedHeight() return NaN for a node not yet
through a layout pass, and `?? 0` does NOT catch NaN (NaN ?? 0 === NaN), so a
pre-layout / mis-timed measureElement() read returned { width: NaN, height: NaN }
— poisoning user layout math (terminalWidth - measured.width → NaN → a NaN width
prop). Coerce non-finite computed dims to 0 (Number.isFinite(v) ? v : 0).
0 is a safe sentinel ("not yet computed"), not the box's true size — the correct
usage is to read AFTER layout (the JSDoc already steers callers to defer via
nextTick). It's chosen because it is Ink's clear intent (`?? 0`) and matches the
DOM precedent (getBoundingClientRect on display:none / img.naturalWidth pre-load
return 0, not NaN). Deliberate, low-risk robustness divergence from Ink v7.0.4's
NaN-leaking `?? 0`; recorded in ink-divergences.md.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(runtime): margin/padding edge removal falls back to the surviving shorthand
Withdrawing a per-edge/axis margin or padding override from a box that still
has a broader shorthand collapsed the edge to 0 instead of falling back. E.g.
`margin={5} marginTop={8}` with marginTop later removed: the setter ran
setMargin(EDGE_TOP, 0), and per yoga edge precedence EDGE_TOP=0 overrides the
surviving EDGE_ALL=5, so the top margin became 0 (the box jumps 5 cells) when
the declarative model (render = f(current props), current = {margin:5}) says 5.
A single per-prop yoga setter can't reconcile an edge that depends on the
specific edge + axis + all-edges shorthand together.
Fix mirrors the existing reconcileBorderEdges pattern: the 14 margin/padding
setters become no-ops, and reconcileMarginEdges/reconcilePaddingEdges recompute
all four physical edges from the box's full el.props with most-specific-wins
precedence (top = marginTop ?? marginY ?? margin ?? 0, ...), zeroing the
composite edges so nothing layers on top. A present-but-non-finite value
(NaN/Infinity) or a withdrawn prop falls THROUGH to the next precedence level,
preserving yoga's prior setMargin(NaN)->fallback behavior; an explicit 0 is
finite and still overrides. margin keeps EDGE_START/END and padding keeps
EDGE_LEFT/RIGHT for left/right, matching the prior setters.
Verified against real yoga-layout@3.2.1 that the SET path produces identical
computed edges as the old per-setter code (no layout regression) across all
combinations and patch orders, and the correct fallback on removal.
This is NOT an Ink-parity item: run against Ink v7.0.4, Ink and pre-fix vue-tui
both collapse to 0 (the identical bug). The fix diverges from Ink by being
declaratively correct under the already-documented G19 reset principle;
recorded in ink-divergences.md alongside the display / flexDirection entries.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(runtime): pin margin/padding spacing to a finite-number contract (Codex review)
The family recompute resolves an edge from a prop only when it is a finite
number (matching the `number` prop type + Ink's number-only spacing); numeric
strings (`margin="5"`) are coerced for Vue static-template ergonomics, but other
non-numeric values (`"50%"`, junk, `""`) are treated as not-set and fall through
to the surviving shorthand instead of being forwarded to yoga.
This makes intentional the behavior change the final review flagged: the OLD
per-setter code incidentally forwarded off-contract strings to yoga (so
`marginTop="50%"` became a percent and `marginTop="foo"` threw). That was
undocumented and non-Ink. Also excludes "" from the present() check so all
non-numeric strings fall through uniformly (Number("")===0 would otherwise
resolve to 0). The numeric/numeric-string SET path is unchanged (re-verified
across all 5040 patch orders).
Tests pin the contract (numeric, numeric-string, "50%"/"foo"/"" fall-through,
NaN/withdrawn fall-through, explicit 0), and ink-divergences.md records it.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Replace the homegrown "Context Engineering" convention with the canonical
Project Context Records (PCR) block in AGENTS.md, and migrate the
.agents/docs/ records to match.
- cross-links: [[wiki-link]] -> relative markdown [name](./name.md)
- provenance: the old "Maintainer decision (DATE): KEEP" markers -> canonical
[VOUCHED @hyf0] stamps (dates dropped, KEEP/OVERRIDE verdicts kept), covering
every variant ((DATE, user-blessed), (maintainer decision DATE), and
"(Decision recorded after review surfaced it.)")
- methodology prose describing the mechanism reworded to the vouch vocabulary
(generic [VOUCHED @handle])
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Rewrites Box/Text/Spacer/Static/Newline from h()/render functions to Vue
<script setup> template SFCs (Transform stays a render fn — it inspects its own
child vnodes), with vue-tsc-verified consumer types (template + JSX fixtures),
provide/inject text context, the always-validate Text divergence (color +
backgroundColor), and three renderer fixes the SFCs surfaced (static anchor skip,
transform line-index Ink-parity, useBoxMetrics subtree drill). Integrates the five
main commits landed after the branch point: #163 public-API audit, generic Static
scoped-slot typing, foreground color validation, useWindowSize/divergence docs.
Squashed from the SFC sub-commits + the two main-integration merges to keep a
linear, rebaseable history. See PR #165 for the full breakdown.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow-up cleanup for the 7 confirmed findings from a review of #163. The
dominant theme: #163 hard-renamed the public composable useTerminalSize ->
useWindowSize (no alias) but left stale references to the dead name in
user-facing docs.
- README.md + packages/runtime/README.md: the composable tables named the
removed `useTerminalSize()` (root README even framed the sole real export
`useWindowSize` as an "Ink-compat alias" — now inverted). Point both at
`useWindowSize()`.
- .agents/docs/ink-divergences.md: two vue-tui-side references to
`useTerminalSize` (the shallowRef "object of refs" example and the
"composables throw outside a render tree" list) -> `useWindowSize`. The
Ink-side `useWindowSize -> WindowSize` naming example is left unchanged.
- .agents/docs/accessibility-api.md: the intro cited three "blessed entries"
but only aria-camelCase is one; `renderToString` layout-only and the
`useWindowSize` name are now Ink parity, not divergences. Reword.
- .agents/docs/api-contract.md: tighten the `/internal` wording — the test
does assert one tripwire on `/internal`, so "not covered by
public-api.test.ts" was imprecise.
- public-api.test.ts / render-to-string.test.tsx: the public renderToString
dropped the `isScreenReaderEnabled` option but (unlike the sibling
`ScreenReaderOptions` type) had no compile-time guard. Replace an obscure,
fmt-fragile type-indexing guard with a readable call-site `@ts-expect-error`
in render-to-string.test.tsx; re-adding the option to the public
RenderToStringOptions makes the directive unused and fails `tsc --noEmit`.
- Rename terminal-size.test.tsx / .sequential.test.tsx ->
window-size.test.tsx / .sequential.test.tsx to match the migrated symbol.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* refactor(runtime)!: rename AnimationOptions to UseAnimationOptions
Align the useAnimation options type with VueUse's UseXOptions convention, matching its sibling composable options bags (UseInputOptions / UsePasteOptions / UseFocusOptions) and the already-correct UseAnimationReturn. Hard rename, no deprecated alias — done while the package is pre-1.0 (0.0.x), so no stability break.
Recorded under "Public composable naming follows Vue conventions" in .agents/docs/ink-divergences.md. Surfaced by the public-API audit.
BREAKING CHANGE: the exported type AnimationOptions is renamed to UseAnimationOptions; update `import { type AnimationOptions }` to `import { type UseAnimationOptions }`.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor(runtime)!: tighten public API to Ink + record aria decision & alignment principle
Public-API audit follow-ups. Where vue-tui had drifted from Ink with no real Vue reason, align to Ink; reduce speculative surface; and record decisions in .agents/docs/ink-divergences.md.
- renderToString: drop the public `isScreenReaderEnabled` option (Ink's public renderToString is layout-only). The SR-capable variant moves to `@vue-tui/runtime/internal` as `renderToStringWithScreenReader` for the accessibility test suite; SR output is unchanged.
- useTerminalSize -> useWindowSize: drop the invented name + alias, align to Ink's `useWindowSize`. The reactive ref return shape is unchanged (shallowRef divergence still applies).
- DevState/DevErrorInfo: move from the public barrel to `@vue-tui/runtime/internal` (internal HMR types, no public consumer; Ink exposes no HMR types).
- docs(divergences): add a standing "Why align to Ink — and when not to" principle (alignment is a means to reduce bugs, not an end; Vue idiom + reasonableness outrank parity); record the aria-props camelCase decision with its run-verified type-safety boundary; stamp the rawMode-default and measureElement-$el entries with their KEEP decisions.
BREAKING CHANGE: removed public exports `useTerminalSize`, `DevState`, `DevErrorInfo`, and `renderToString`'s `isScreenReaderEnabled` option. Use `useWindowSize`; import HMR types from `@vue-tui/runtime/internal`.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* refactor(runtime)!: move renderScreenReaderOutput to /internal-only
The screen-reader linearizer (ported from Ink's internal
`renderNodeToScreenReaderOutput`) was exported from the public barrel, but it
was never usefully public: its only parameter type `TuiNode` and the
node-construction primitives needed to build one are not public, so a public
consumer could not name or construct the argument. Ink keeps its counterpart
module-internal; we match that.
`renderScreenReaderOutput` + `ScreenReaderOptions` now live only in
`@vue-tui/runtime/internal` (already re-exported there). The live SR machinery
(render, the internal renderToStringWithScreenReader, the <Static> channel)
imports from the source module and is unaffected; public SR output is reached
via the mount `isScreenReaderEnabled` option.
public-api.test.ts: drop it from the public-members list; add a runtime guard
(absent from public, present on /internal) plus a compile-time @ts-expect-error
guard that the `ScreenReaderOptions` type cannot be re-added to the public
barrel.
Docs: new .agents/docs/accessibility-api.md (aria + SR design) and
api-contract.md (public surface = exports + their user-consumable types;
/internal is not the contract); resolve the open item and cross-link from
ink-divergences.md.
BREAKING CHANGE: renderScreenReaderOutput and ScreenReaderOptions are no longer
exported from @vue-tui/runtime; import from @vue-tui/runtime/internal if needed.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* test(runtime-tests): snapshot the exact public value-export set
Upgrade public-api.test.ts from "documented members present + targeted
negatives" to an exhaustive snapshot of the exact runtime value-export surface
of `@vue-tui/runtime`: adding, removing, or renaming any value export now fails
the test, so every public-surface change must be a deliberate edit to the list.
Type-only exports are erased at runtime and cannot be enumerated, so the type
surface stays guarded individually (the `@ts-expect-error` ScreenReaderOptions
guard); api-contract.md is updated to state this boundary precisely.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
A thrown non-Error whose .message is a string (throw {message:'x'})
displayed 'x' in the ErrorOverview but rejected waitUntilExit() with
new Error(String(value)) = '[object Object]' — display and reject
disagreed. Introduce one messageForNonError(value) helper (string
.message else String(value)) and feed it to BOTH the overview header
and the two non-Error reject-wrap sites, so the shown and rejected
messages can never drift. Overview output is byte-identical (the helper
is the prior inline logic extracted); real-Error, cross-realm, and
no-synthetic-stack paths are unchanged.
Blesses vue-tui's uniform show-the-error-and-reject behavior for any
thrown value (audit e17): Ink instead resolves waitUntilExit() with a
truthy thrown value and silently hangs on a falsy throw — abnormal, so
vue-tui deliberately diverges. Ledger entry rewritten to the full
run-verified scope with Maintainer decision (2026-06-12): KEEP.
Red-first: a consistency test asserting throw {message:'objmsg'} shows
AND rejects 'objmsg'.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
A focused input's caret zombied to the bottom-left corner whenever an
unrelated repaint (spinner tick, log line, progress bar) committed
without re-declaring the cursor: the active cursor was gated on a
per-commit dirty/reference change, so an unrelated commit dropped it.
Real terminal programs that own an edit point re-place the caret there
every frame (vim emits an absolute CUP after each repaint, readline
re-lands the buffer offset on SIGWINCH, nano homes to its edit cell).
Match that: the runtime now re-emits the last-declared caret at the end
of every commit until the declaration changes or is cleared, so the
caret survives unrelated repaints in all component topologies. The
position is clamped to the visible region (D5) and a cleared
declaration emits no caret, so teardown still hands the cursor back.
This is a deliberate divergence FROM Ink, which re-asserts only when
the cursor's React component re-renders and so zombies the caret in
sibling/leaf topology too (run-verified). Aligning to Ink reduces bugs
only when Ink is correct; here matching Ink would preserve abnormal
behavior. Overrides the prior 2026-06-01 KEEP, whose rationale (avoid
diverging from Ink in the sibling direction) was overturned by running
real terminal apps. The {x,y} setCursorPosition API is unchanged (it
remains the IME primitive); the fix is an internal per-commit re-emit.
Red-first: a real-TTY PTY test with sibling-topology spinner state
asserts the spinner-only frame ends with the caret-restore suffix.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Run-verified vs Ink v7.0.4 (audit e11 decisive experiment, 3/3 runs
byte-deterministic): the concurrent render option's root-tag half is
inert under react-reconciler 0.33.0 (every root becomes ConcurrentRoot;
hooks and preemption probes behave identically), but its dispatch half
is live — the default commits the first frame synchronously inside
render()/rerender(), concurrent:true schedules it on a later tick.
The old entry described the flag as having no observable surface at
all; vue-tui's mount() matches Ink's default dispatch.
Also append the audit provenance note to the paint-validation entry:
the scheduler-wedge rationale rests on the earlier paint-throw
investigation — the audit probe could not reach a paint throw from
public or raw-host input (paint's border fallback intercepts it).
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
The commit scheduler armed one fixed trailing timer at the start of a
throttle window, firing the trailing commit at windowStart+wait. Ink's
es-toolkit throttle re-arms on every throttled call: trailing fires at
lastCall+wait. Deterministic probe at maxFps=10 (updates t0/t0+43/
t0+86): Ink trailing median 192.5ms, vue-tui 103.6ms (audit e29).
Mirror the observable timing of es-toolkit's throttle: leading commit
when no window is active, per-call trailing re-arm (lastCall+wait), and
the maxWait edge (a call a full window after the first deferral commits
synchronously) so sustained updates keep the ~wait cadence instead of
debounce-starving. Resize cancellation (a separate blessed divergence)
is preserved: the post-fix cancel probe is byte-identical.
Red test is the discriminating multi-deferred-call shape: a single-
deferred-call test goes green under the wrong firstDeferredCall anchor.
Post-fix probes land at 188.3-189.2ms, inside Ink's 188.4-195.0 band;
CI=true vp run ci passes alongside vp run ready.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
The instance-reuse guard set a per-app skippedMount flag that was never
reset, so one guarded mount() call permanently disabled the app's own
teardown. Three run-confirmed wedges (audit e18), all absent in Ink:
- an owner double-firing mount() on its own live stdout kept painting
after unmount() and leaked its registry entry
- an app that once hit the guard could never unmount a later legitimate
mount on a free stdout
- an app live on stream A that merely targeted another app's busy
stream B became unkillable on A
Delete the flag; teardown()/resolveExit() now consult the actually
wired state (mountedAppContext / mountedAsOwner), so a guarded call is
inert for that call only. The blessed inert-no-op divergence from Ink's
reuse-and-rerender is unchanged; the ledger entry is reworded to the
call-scoped semantics.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Run-based audit against pinned Ink v7.0.4 (40b3a75) found two ledger
defects:
- The composable-naming entry claimed Ink names hook return types
XProps; real Ink 7.0.4 is mixed (XProps for stream/app hooks,
UseBoxMetricsResult/AnimationResult/WindowSize for newer ones, and
several hooks export nothing). Restate both sides accurately.
- The <Transform> all-comment-children entry sat under Intentional
Divergence Choices, but its forcing is the React-only false !== null
edge that Vue's comment-vnode materialization cannot see — a
model-implied difference per the doc's own classification flow. Move
it there, expand the run-verified boundary (empty slot array, ''/0
children, screen-reader label), and reword the Non-Behavioral Notes
cross-ref in the same change so it stays true.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
A [Vue warn] emitted during the initial mount (e.g. the missing-render-
function warn from a root setup() throw) escaped the stderr filter
because mount() installed the console patch only after originalMount.
Ink patches in its constructor before the first React render
(ink.tsx:435-436); move the install before originalMount to match.
The mount-throw catch already restores the console via teardown().
Verified red-first against real Ink v7.0.4 (audit e10): Ink's stderr
stays empty for a render-throwing component; vue-tui's initial-mount
warn reached a real PTY before this fix.
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
Addresses PR review: the containing block for an absolutely-positioned child
is the **padding box** (inside the borders), not the "border-box". Verified by
running yoga (an abs child at top:0/left:0 insets by the border only, never by
padding — confirmed across border/padding combos) and the real Ink/vue-tui
renderers (X lands at the inner-border edge, byte-identical in both).
- Tighten the regression test: exact-frame assertions instead of `toContain`,
including a border+padding case that distinguishes the padding box from the
content box — the assertion that would have caught the original wording slip
(presence-only assertions could not).
- Correct "containing block (border-box)" -> "padding box (inside the borders)"
in paint.ts, layout-guards.ts, and ink-divergences.md. (The unrelated
"border-box-like" *sizing* notes are correct and left as-is.)
No runtime behavior change; the code already used yoga's computed position.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Findings confirmed by running real Ink v7.0.4 + vue-tui harnesses (and
cross-checked against source), not source-reading:
- nonerror-message: reclassify Additive Supersets -> Intentional. Ink accepts
the same non-Error throw and renders a blank-message overview (same input,
different output), so it is not an additive superset.
- flexdir-reset: reframe as parity through the public <Box>. Ink's Box
re-injects flexDirection:'row'/flexWrap:'nowrap', so removing the prop
resets in BOTH engines (column -> row); the "Ink persists" framing held only
at the raw ink-box host layer. Kept as the explicit contrast to display.
- resize: retitle/reword — scheduler.cancel() is unconditional on every resize,
not narrowing-only; the dedup is driven by overflow + a pending commit.
- usecursor: correct the "Why" — a render-body-set cursor is still dropped on an
ancestor-only commit (Ink re-asserts); the gap is broader than "set-once".
- shallowref-state: useTerminalSize() returns { columns, rows } (object of
refs), not a single shallowRef read as .value.
- composable-naming: the XProps return-type convention holds only for the
stream/app hooks (StdinProps/AppProps); useInput/usePaste/useFocus don't fit.
- renderToString SR: note Ink's live render() does expose isScreenReaderEnabled,
so it's a string-API gap, not a missing SR capability.
- measureElement ref: $el is the primary path (Box is a defineComponent).
- trim two Non-Behavioral bullets that duplicated AGENTS.md house rules.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The zero-content-area guard (layout-guards + paint.ts) suppressed ALL
children of a Box whose inner content rect collapsed to zero, including
position:"absolute" children. An absolutely-positioned child is placed
against the containing block (border-box), not the content rect, so Ink
v7.0.4 paints it (verified by running real Ink: a w=2 h=2 single-border
box with an absolute child renders "┌┐#\n└X"); vue-tui dropped it,
rendering "┌┐#\n└┘".
- layout-guards: exempt POSITION_TYPE_ABSOLUTE children from the hide loop
so they keep their layout.
- paint: move overflow-clip setup above the zero-content early-return and,
in that branch, paint only absolute children (still clipped by
overflow:hidden, matching Ink) while keeping in-flow children suppressed
(the blessed degenerate-box divergence).
Flow-child suppression and overflow:hidden clipping both stay Ink-aligned
(verified byte-identical against real Ink). Known limitation: an absolute
descendant nested under a suppressed in-flow child is still dropped (the
flow ancestor is removed from layout) — scoped to direct absolute children.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
An explicit `display={undefined}` is applied as DISPLAY_NONE and hides on
mount (verified against real Ink v7.0.4) — only an omitted `display` stays
visible. The previous parenthetical conflated the two.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Verified against real Ink v7.0.4 (react 19.2) instead of source-reading:
- display-reset: Ink does not "persist" a withdrawn `display`. Its diff
emits a removed key as `display: undefined`, and applyDisplayStyles
sets DISPLAY_NONE for any non-`'flex'` value, so clearing a previously
set `display` hides the box (verified: flex -> removed goes visible ->
hidden). Describe the real behavior and the common-toggle consequence.
- react-concurrent: drop "Suspense" from the React-only list. Vue ships a
built-in `<Suspense>`; the genuine gap is interruptible concurrent
rendering / `useTransition`.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Align several user-observable runtime behaviors with the Ink v7.0.4 parity audit: live input/paste handler refs, duplicate focus id registration, string-only color props, noninteractive empty final newlines, cross-realm error headers, and contained zero-content box layout/paint.
Document Vue-specific KEEP decisions and require Conventional Commits for commit messages and PR titles.
Co-authored-by: Claude <noreply@anthropic.com>
Re-verified every ledger entry against Ink v7.0.4. Fixes: MI5 (remove false
"residual false-child divergence", verified by running Ink), MI3 (Model-Implied
-> Intentional), VI2 (-> Non-Behavioral), VI3 (-> Intentional), and an
error-overview test comment (Ink renders blank error.message, not String(value)).
ID4 verified accurate and left unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A second mount() on a live stdout stays a warn + no-op (no behavior change).
This reframes it as a deliberate, maintainer-blessed choice instead of the
vaguely-justified entry it was.
- warning (render.ts): rewrite the stderr message to state the situation plus
the two recovery paths (update reactive state, or unmount() the existing app
first), replacing the old "unsupported / call unmount() first" phrasing.
- divergence doc: rewrite the Why with the real rationale (it is a misuse path;
Ink warns it is unsupported too; vue-tui fails safe by keeping the live app
and warning; there is no clean public path to Ink's reuse-and-rerender under
the createApp model). Move the entry from "Vue-Idiomatic Choices" to
"Intentional Divergence Choices"; record "Maintainer decision: KEEP"; tighten
the inert-handle wording (unmount() only settles its own exit promise).
- tests: retarget the instance-reuse-guard assertions to the new warning text,
including a negative assertion that previously matched a stale substring.
Verified: vp run ready (integration 1104 + PTY 123 passing, lint/type/build).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The package.json `exports` shape — runtime/testing's bare-string target vs Ink's
explicit `types` condition — is build-toolchain plumbing, not a runtime or
user-facing-API behavior. TS resolves the declaration identically via the
`.d.mts`-next-to-`.mjs` adjacency tsdown emits, so nothing is observably
divergent. It fit none of the doc's categories (mis-filed under "Public API
surface") and was an AI-added defensive note guarding a non-event.
Alignment tracks runtime behavior + the user-facing API; packaging/internal
infra (exports shape, ./internal, dist vs build, .mjs vs .js) is out of scope
and isn't recorded as a divergence.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Audit of the "byte-identical reconciler/runtime mechanics" subsection, each entry
verified against vue-tui + Ink v7.0.4 source.
Doc:
- Rewrite the TuiComment/Transform entry cause-first (Vue materializes a comment
placeholder where React renders nothing -> vue makes it inert) for clarity.
- Fix the commit-throttle figure: it is `ceil(1000/maxFps)` = 34ms at the default
maxFps=30, not "~32ms". The 32 was vue's own dead fallback constant, never the
production value; Ink has no 32 either.
- Drop the keyed-lists (LIS) entry: it restated the section header and guarded no
vue-authored code (patchKeyedChildren is upstream Vue).
- Drop the wrapText-truncate and animation-scheduler entries: both are vue-tui
implementation choices, not Vue-vs-React framework differences, and both are
already explained by their in-code comments.
Code (no behavior change; verified by `vp run ready`):
- Remove the dead `DEFAULT_THROTTLE_MS = 32` fallback in scheduler.ts. Production
always passes throttleMs (render.ts derives it from maxFps) and the immediate
path never reads it, so the 32 fallback never gated a frame. Make throttleMs
required; render.ts always passes it (0 when unthrottled).
- Tighten the animation-scheduler ceil comment (drop the "busy-loop" overstatement;
the fractional-delay truncation it describes is real and keeps the Math.ceil).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The `useFocusManager().activeId` entry conflated two things and buried the
load-bearing one. Split and reframe:
- The real divergence is framework-semantic, not API-specific: a React hook
re-runs each render so it can return a plain snapshot, whereas a Vue
composable's setup() runs once and must wrap reactive state in a `shallowRef`.
Moved to the Vue != React section as a general rule; `activeId` is now just
one example of it.
- Folded the empty-value convention (`null` vs Ink's `undefined`) into that
entry as a Vue ecosystem idiom rather than a separate headline.
Doc-only.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
vue-tui validates invalid render input (a chalk-modifier backgroundColor like
"bold", an unknown borderStyle) at the component-render layer (Box.ts/Text.ts),
not sunk into the paint layer — so a bad value throws where the error boundary
catches it (ErrorOverview → reject waitUntilExit) instead of crashing.
Records the framework-semantic forcing function (vue-tui's paint runs in a Vue
post-flush callback, so a paint-layer throw escapes onErrorCaptured and wedges
the scheduler), notes the React/Vue symmetry (a paint-layer throw is uncatchable
by component boundaries in both engines, not a Vue weakness), and the honest
cost (eager render-time validation over-throws in a few degenerate cases Ink's
lazy paint check never reaches).
Verified against Ink v7.0.4 source; entry reviewed by Codex.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Ink's colorize throws on a chalk-modifier-name backgroundColor (e.g. "bold":
'bold' in chalk but chalk.bgBold is undefined -> "chalk.bgBold is not a
function"); vue-tui degraded to bare text. Align: validate backgroundColor at
component render (Text + Box own bg + drawn border edges) so the throw is caught
by the error boundary, not the post-flush paint pass (a throw there wedges Vue's
scheduler — cf. borderStyle #124). Detection mirrors Ink exactly: only the
in-chalk-but-no-bg-method case throws; valid colors / hex / ansi256 / rgb /
[r,g,b] / non-chalk strings and foreground modifiers (color="bold" still bolds)
are unaffected. Border bgs are gated to Ink's render-border conditions
(borderStyle + drawn edge + perEdge ?? general); empty/hidden elements don't throw.
Since Ink throws lazily at paint (with layout/squash info) while vue must validate
eagerly at render, a few degenerate cases (content-area<=0 box, degenerate
top/bottom border, nested-empty text) over-throw on the invalid modifier input —
documented in code as architecturally irreducible. Removes the A12 entry from
ink-divergences.md.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Ink's applyDisplayStyles hides any present `display` that isn't 'flex'
(DISPLAY_NONE); vue-tui hid only on exact 'none', leaving off-spec values
(reachable via TS-bypass — the prop type is 'flex'|'none') visible. Align: the
yoga display setter now hides any present (non-null) value except 'flex',
matching Ink even for non-string junk (display={5}). The blessed A19 divergence
is preserved — a removed/undefined display (null) still resets to the visible
default (Vue can't distinguish display={undefined} from an omitted prop).
Removes the now-obsolete A21 entry from .agents/docs/ink-divergences.md (the
A19 "removed display resets to visible" entry remains).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The rawMode 'always' divergence entry overstated the Ctrl+C benefit — it framed
"reaching the app's interrupt handler" as the headline, which only applies under
the non-default `exitOnCtrlC: false`. Both Ink and vue-tui default exitOnCtrlC to
true, so by default Ctrl+C exits either way; the lazy-vs-always difference there is
only the exit path/code (graceful 0 vs re-raised SIGINT 130). Reword to lead with
the real default consequence (echo into the frame on no-input screens) and state
the Ctrl+C difference accurately, noting it only bites an app that sets
exitOnCtrlC:false. No behavior change.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add a `rawMode?: 'always' | 'auto'` mount option (replacing the dead, unwired
`rawMode?: boolean`), defaulting to 'always'.
- 'always' (default): the App takes a lifetime raw-mode hold at mount (gated on
interactive + a TTY stdin), so raw mode is held for the whole run regardless of
which input composables are mounted. Keystrokes never echo into the rendered
frame on a no-input/streaming screen, and Ctrl+C is handled consistently on
every screen (e.g. it reaches an agent's "interrupt generation" handler instead
of becoming a kernel SIGINT). Because owning raw mode ref()s stdin, the app
stays alive until an explicit unmount()/exit() — it does NOT auto-exit when idle.
- 'auto': Ink's original lazy model — raw mode is acquired only while a useInput /
useFocus / usePaste is mounted, so a no-input screen returns to cooked mode and a
no-input app auto-exits. The opt-out for inline / render-and-exit tools.
This is a deliberate divergence from Ink (the cross-framework norm — Bubble Tea,
Textual, Ratatui, prompt_toolkit all own the terminal for the program lifetime;
Ink's hook-driven model is the outlier). Documented in
.agents/docs/ink-divergences.md.
Implementation: the App holds a `lifetimeFloor` ref via holdRawModeForLifetime();
input composables stack above it. The per-consumer clearInputState is re-based to
the floor so a buffered partial escape (e.g. a lone ESC at a screen transition)
can't bleed into the next consumer — cleared both when the last consumer releases
and when the first consumer re-acquires above the floor (covers same-tick swaps
AND a delayed idle→input transition). The data listener and raw toggle stay on
until teardown, where dispose() releases the floor ref (raw disabled + stdin
unref'd exactly once).
Tests: rawMode-lifecycle ('always' holds raw with no input; 'auto' stays cooked;
no mid-session oscillation; no partial-escape bleed across a swap or an idle gap);
PTY exit-rawmode-always (a no-input 'always' app stays alive and exits on Ctrl+C).
The 6 auto-exit PTY fixtures are pinned to 'auto' (they model render-and-exit).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add an Additive-features entry documenting that two apps sharing one stdin
(different stdout) both receive input in vue-tui, where Ink's first-registered
'readable' listener drains the buffer (second app deaf until the first
unmounts) and its per-App raw-mode count drops raw mode on the first unmount.
vue refcounts the raw-mode toggle per-stdin (shared) and attaches the 'data'
listener per-controller, so the push event broadcasts to both. Implemented in
#118; single-app behavior is byte-identical.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The round-2 parity audit surfaced ~15 places where vue-tui deliberately differs from
Ink v7.0.4 (verified against Ink source, not invented). Document them so they aren't
re-"fixed" back into Ink's behavior:
Public API surface: useFocusManager().activeId is null (reactive ShallowRef) not
undefined (+ lock test); second mount() on a live stdout is an inert no-op vs Ink
reusing the instance; bare-string package exports vs an explicit types condition.
Additive (strict superset): RGB [r,g,b] tuples on every color prop (Ink string-only,
throws on an array); backgroundColor=chalk-modifier-name degrades to bare text (Ink
throws); useAnimation outside a tree drives a real scheduler; measureElement/
useBoxMetrics also accept a Vue component ref via $el; renderToString accepts
isScreenReaderEnabled; narrowing resize cancels the redundant trailing clearTerminal.
Framework-semantic (Vue ≠ React): an off-spec display value stays visible (Ink hides
any non-'flex'); out-of-type flex/align values are forwarded not defensively coerced
(only flexShrink — flexGrow matches Ink; reachable only via TS-bypass); duplicate
explicit-id useFocus dedups to one entry; the terminal-bound composables fail fast
outside a tree (useBoxMetrics/useAnimation degrade); a setup()-throw emits a dev-only
[Vue warn]. Plus wrapText truncate's per-line short-circuit and the scheduler's
ceil'd delay under reconciler mechanics.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Investigating P15 (a stable-reference cursor dropped on an unrelated commit) found
the premise was false: Ink does NOT re-assert the cursor on every commit. Ink's
useCursor uses a no-deps useInsertionEffect that re-runs only when the cursor
COMPONENT re-renders. React re-renders a whole subtree on an ancestor's commit, so
Ink re-asserts when the cursor is in that subtree — but when an unrelated SIBLING
owns the changing state, the cursor component does not re-render and Ink drops the
cursor too. vue (watch on positionRef) already matches Ink in that sibling case and
for the recommended reactive usage; the two differ only in the narrow edge of a
set-once cursor plus an ancestor-driven commit (Vue's fine-grained reactivity vs
React's render cascade). A global per-commit re-assert would diverge from Ink in the
opposite (sibling) direction. So this is an unavoidable Vue ≠ React consequence:
document it and keep the reactivity-tied behavior rather than "fix" it.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Ink's <Transform> returns null (no node) when children are undefined/null, and
that guard runs BEFORE the accessibilityLabel substitution (Transform.tsx:28-30).
vue always created a "transform" host node, so:
- an empty <Transform> in a flex `gap` row consumed a gap slot Ink never adds (P13); and
- a childless <Transform accessibilityLabel> emitted the label even though Ink's
null guard wins over it (P19).
Add the null-children guard at the top of the render fn. Vue materializes a bare
null/false/undefined/v-if=false child as a single Comment vnode and cannot tell them
apart, so the predicate treats the whole group as "no children" (slot undefined OR
every vnode is a Comment) — matching Ink for the common `{null}`/`{cond ? x : null}`
idioms and keeping <Transform> consistent with vue-tui's documented comment-anchor
model (every other component already omits a false/v-if child). An empty-string ({''},
a Text vnode) or JSX empty array ({[]}, a Fragment) still renders, matching Ink.
This deliberately diverges from Ink only for a literal {false} / {cond && x}-false
child (React's false !== null → Ink renders an empty gap-slot node); documented in
ink-divergences.md and locked by a test, since Vue physically cannot distinguish it
from null.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The parseKeypress kitty query-response filter (ESC[?Nu -> {ignore:true}, dropped by useInput)
was assumed to be a redundant second net duplicating the upstream kitty-keyboard detection
strip. Verified it is NOT redundant: the upstream strip runs only in the one-shot
confirmKittySupport detection listener (a private buffer), not on the steady-state input path
(stdin 'data' -> inputParser -> emitInput -> useInput -> parseKeypress). Removing the filter
leaks a stray query-response to handlers as spurious "[?1u" input in enabled mode, auto mode,
and split delivery (inputParser reassembles, so the upstream partial-handling doesn't apply).
Ink has the same gap (its parse-keypress has no query filter; its steady-state input uses
'readable', not 'data').
Keeps the filter (adds a why-comment) and records it as an additive divergence in
ink-divergences.md. Adds 4 end-to-end regression tests proving the filter is load-bearing.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
vue-tui left `display` out of RESETTABLE_PROPS, so a removed/undefined `display` persisted
its prior value (a removed display="none" stayed hidden). Adds `display` to RESETTABLE_PROPS
-- the setter already maps undefined -> DISPLAY_FLEX -- so a withdrawn `display` returns to
the Box default (visible), per render = f(current props), like flexDirection/flexWrap (G19).
Deliberate, documented divergence from Ink (which hides on a present-undefined `display` via
DISPLAY_NONE, and persists on omitted) -- recorded in ink-divergences.md. The reset is
consistent across the visual and screen-reader paths (both read yoga's display state).
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
ErrorOverview was a one-line `Name: message` stub; Ink v7.0.4 renders a full
ERROR overview. Ports it faithfully:
- white-on-red ` ERROR ` label + message, dim cwd-relative file:line:column origin
- a code excerpt around the throwing line (padded gutter, error line highlighted,
`Line N` / `Line N, error` aria-labels)
- the parsed stack (`- fn (file:line:col)`, cwd-relative, StackUtils nodeInternals
filtering, unparsable-line fallback, fs.existsSync guard)
Adds code-excerpt@4.0.0 + stack-utils@2.0.6 (the versions Ink uses). The error
boundary now keeps the raw thrown value for display (Ink stores the raw value;
ErrorOverview renders a stack only when one exists) -- so a non-Error throw no
longer shows a misleading synthetic framework stack. The exit/reject path still
receives a wrapped Error (semantics unchanged). The unparsable-stack fallback
emits a literal backslash-t to match Ink's JSX.
Showing String(value) for a non-Error message (vs Ink's blank) is a documented
additive divergence (ink-divergences.md).
Adds 4 error-overview tests.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
useStdin() returned the full internal StdinContext (8 members incl. the raw-mode
ref-counting primitives acquireRawMode/releaseRawMode, setBracketedPasteMode, and
internal_eventEmitter/internal_exitOnCtrlC). Ink's useStdin() returns only its PublicProps
— { stdin, setRawMode, isRawModeSupported } — keeping the rest on the internal context,
reached via the internal useStdinContext()/inject. Verified against Ink 7.0.4
(src/hooks/use-stdin.ts:10, src/components/StdinContext.ts).
- Narrow useStdin(): UseStdinReturn (the 3 public fields). The full StdinContext stays
internal, reached by useInput/useFocus/usePaste via inject(StdinContextKey) — the runtime
object is unchanged, only the public type narrows (mirrors Ink's type-level narrowing).
- Name every stdio/app composable return type per VueUse's UseXReturn convention and export
them: UseStdinReturn, UseStdoutReturn, UseStderrReturn, UseAppReturn (shapes byte-identical
to Ink's StdinProps/StdoutProps/StderrProps/AppProps). vue-tui reserves XProps for component
props (BoxProps, via ExtractPublicPropTypes), so composable returns use UseXReturn — the
Vue-community-idiomatic name.
- Unify the two pre-existing return types onto the same convention:
AnimationResult → UseAnimationReturn, UseBoxMetricsResult → UseBoxMetricsReturn.
- Type-level test (public-types.test-d.ts) locks the shapes and asserts useStdin()'s public
return excludes the internal members.
- Docs: trim the type-reexport divergence area to genuine divergences (fold
RenderOptions/Instance → MountOptions/TuiApp into createApp; keep DOMElement → TuiNode);
drop the AppProps/StdinProps/StdoutProps/StderrProps "N/A" entry — those Ink names are the
hook return types, now mirrored as UseXReturn (recorded under Framework idioms). Supersedes #77.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The placeholder listed three candidates to evaluate as by-design divergences:
v-if/null comment host nodes, reactivity-driven re-render timing, and keyed
reconciliation order. Grounding each in the runtime code shows all three produce
byte-identical terminal output vs Ink — a commit always paints f(current host
tree), so *how* the tree was built never reaches the terminal. None is a
divergence, so the placeholder is removed and the three are recorded as concise
one-liners under "Framework idioms (noted, not behavioral divergences)":
- v-if=false / null|false|undefined children become an inert Comment vnode
(TuiComment) — no yoga node, paints nothing, doesn't shift a sibling's yoga
index, skipped for the positional <Transform> index (G52); output equals
omitting the element.
- Commit timing is deliberately Ink-aligned (~32ms ceil(1000/maxFps) throttle,
sync resize), so Vue's fine-grained re-render granularity stays unobservable.
- Keyed lists use Vue core's patchKeyedChildren, not React's fiber diff; output
depends on the final tree, not the move order.
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Move the exitOnCtrlC guard into the always-on stdin controller (emitInput),
encoding-agnostic via parseKeypress, so Ctrl+C exits under both the legacy
\x03 byte and the kitty CSI-u form regardless of which composable holds raw
mode (useInput / useFocus / usePaste, or none). Single source of truth —
dropped from useInput. Excludes Ctrl+Shift+C; fast-paths \x03 and only parses
escape-prefixed sequences. Adds TDD PTY coverage and updates the divergence doc.
Also: stop tracking docs/superpowers/plans/2026-05-27-ink-test-parity.md —
docs/ must stay out of git (AGENTS.md); it was committed before .gitignore
covered it.
Yunfei He