feat(tui): port claude-code's wheel accel state machine

Replaces the static WHEEL_SCROLL_STEP=1 multiplier on wheel events
with an adaptive accel state machine that infers user intent from
inter-event timing.

Algorithm ported straight from claude-code's
src/components/ScrollKeybindingHandler.tsx.  All tuning constants,
the native/xterm.js path split, the encoder-bounce detection, the
trackpad-burst signature → all theirs.  This file is a mechanical
port into our module structure.

What it does:

  precision click (>500ms gap)   1 row/event   (deliberate scan)
  sustained mouse (40-200ms)     2-6 rows      (decay curve)
  detected wheel bounce          ramps to 15   (sticky wheel-mode)
  trackpad flick (5+ <5ms)       1 row/event   (burst detect)
  direction reversal             reset to base

Two implementation paths:

  * native terminals (ghostty, iTerm2, Kitty, WezTerm) — linear
    window-ramp + optional wheel-mode curve triggered by detected
    encoder bounce.  SGR proportional reporting handled via the
    burst-count guard.

  * xterm.js (VS Code / Cursor / browser terminals) — pure
    exponential-decay curve with fractional carry.  Events arrive
    1-per-notch with no pre-amplification, so the curve is more
    aggressive.

Selected at construction via isXtermJs() from @hermes/ink (now
exported).  Per-user tune via HERMES_TUI_SCROLL_SPEED (alias
CLAUDE_CODE_SCROLL_SPEED for portability).

13 unit tests covering direction flip/bounce/reversal, idle
disengage, trackpad-burst disengage, frac invariants, and the
native vs xterm.js branches.

Profiled under --rate 30 (stress test) and --rate 10 (realistic
sustained scroll): accel ramps to cap=6 at 30Hz burst, decays to
1-3 rows at sparse 10Hz clicks.  Perf is comparable to baseline
because accel IS multiplying step — the win is perceptual (fast
flicks cover distance, slow clicks keep precision), not raw fps.

Companion to the earlier WHEEL_SCROLL_STEP=1 change: that set the
base; this modulates around it.
This commit is contained in:
Brooklyn Nicholson 2026-04-26 17:16:11 -05:00
parent 0cd98499bb
commit 4395c2b007
5 changed files with 433 additions and 5 deletions

View file

@ -0,0 +1,169 @@
import { describe, expect, it } from 'vitest'
import { computeWheelStep, initWheelAccel } from '../lib/wheelAccel.js'
describe('wheelAccel — native path', () => {
it('first click after init returns base', () => {
const s = initWheelAccel(false, 1)
expect(computeWheelStep(s, 1, 1000)).toBe(1)
})
it('same-direction fast events ramp mult (window-mode)', () => {
const s = initWheelAccel(false, 1)
// First click establishes dir. Subsequent clicks inside the 40ms
// window ramp by +0.3 each (capped at 6).
computeWheelStep(s, 1, 1000)
computeWheelStep(s, 1, 1020)
computeWheelStep(s, 1, 1040)
const fourth = computeWheelStep(s, 1, 1060)
// After 3 window events: mult starts at 1 → stays 1 on first ramp
// (first event just sets baseline), then +0.3 × 3 = 1.9 → floor=1.
// The key property: doesn't shrink below base.
expect(fourth).toBeGreaterThanOrEqual(1)
})
it('gap beyond window resets mult to base', () => {
const s = initWheelAccel(false, 1)
// Ramp up
for (let t = 1000; t < 1100; t += 20) {
computeWheelStep(s, 1, t)
}
// Long pause, then click
const afterPause = computeWheelStep(s, 1, 2000)
expect(afterPause).toBe(1)
})
it('direction flip defers one event for bounce detection', () => {
const s = initWheelAccel(false, 1)
computeWheelStep(s, 1, 1000)
// Flip — should defer
expect(computeWheelStep(s, -1, 1050)).toBe(0)
})
it('flip-back within bounce window engages wheelMode', () => {
const s = initWheelAccel(false, 1)
computeWheelStep(s, 1, 1000)
// Flip (deferred)
computeWheelStep(s, -1, 1050)
// Flip BACK within 200ms → bounce confirmed → wheelMode engaged
computeWheelStep(s, 1, 1100)
expect(s.wheelMode).toBe(true)
})
it('flip-back outside bounce window is a real reversal (no wheelMode)', () => {
const s = initWheelAccel(false, 1)
computeWheelStep(s, 1, 1000)
computeWheelStep(s, -1, 1050) // defer
// Flip-back arrives 300ms later → too late → real reversal
computeWheelStep(s, 1, 1400)
expect(s.wheelMode).toBe(false)
})
it('5 consecutive sub-5ms events disengage wheelMode (trackpad signature)', () => {
const s = initWheelAccel(false, 1)
s.wheelMode = true
s.dir = 1
s.time = 1000
// 5 bursts <5ms apart (trackpad flick)
computeWheelStep(s, 1, 1002)
computeWheelStep(s, 1, 1004)
computeWheelStep(s, 1, 1006)
computeWheelStep(s, 1, 1008)
computeWheelStep(s, 1, 1010)
expect(s.wheelMode).toBe(false)
})
it('1.5s idle disengages wheelMode', () => {
const s = initWheelAccel(false, 1)
s.wheelMode = true
s.dir = 1
s.time = 1000
computeWheelStep(s, 1, 3000) // 2 second gap
expect(s.wheelMode).toBe(false)
})
})
describe('wheelAccel — xterm.js path', () => {
it('first click returns 2 after long idle', () => {
const s = initWheelAccel(true, 1)
// First event — "sameDir && gap > WHEEL_DECAY_IDLE_MS" triggers
// reset-to-2 branch since dir starts at 0 and 0 !== 1.
const n = computeWheelStep(s, 1, 1000)
expect(n).toBeGreaterThanOrEqual(1)
})
it('sub-5ms burst returns 1 (same-direction, same-batch)', () => {
const s = initWheelAccel(true, 1)
computeWheelStep(s, 1, 1000)
const burst = computeWheelStep(s, 1, 1002)
expect(burst).toBe(1)
})
it('slow steady scroll stays in precision range', () => {
const s = initWheelAccel(true, 1)
// Simulated 30Hz sustained scroll: 33ms gap
const results: number[] = []
for (let t = 1000; t < 2000; t += 33) {
results.push(computeWheelStep(s, 1, t))
}
// Every event should produce 1-6 rows. No runaway.
for (const r of results) {
expect(r).toBeGreaterThanOrEqual(1)
expect(r).toBeLessThanOrEqual(6)
}
})
it('direction reversal resets mult', () => {
const s = initWheelAccel(true, 1)
// Ramp up
for (let t = 1000; t < 1100; t += 20) {
computeWheelStep(s, 1, t)
}
const beforeFlip = s.mult
// Flip
computeWheelStep(s, -1, 1200)
expect(s.mult).toBeLessThanOrEqual(beforeFlip)
// Reset branch sets mult=2
expect(s.mult).toBe(2)
})
it('frac stays in [0,1) across events', () => {
const s = initWheelAccel(true, 1)
// frac must never go negative or reach 1.0 — that's the correctness
// invariant of the fractional carry. Whether a specific series of
// inputs produces a nonzero frac depends on tuning constants; just
// check the bound is maintained across a realistic scroll pattern.
for (let t = 1000; t < 1200; t += 30) {
computeWheelStep(s, 1, t)
expect(s.frac).toBeGreaterThanOrEqual(0)
expect(s.frac).toBeLessThan(1)
}
})
})