Files
BadNote/lib/editor/engine/stroke_geometry.dart

174 lines
7.1 KiB
Dart
Raw Normal View History

// lib/editor/engine/stroke_geometry.dart
//
// Single source of stroke outline geometry for both screen render and export.
// The recipe is lifted verbatim from the proven live
// `ink_painters.buildStrokePath` (lib/editor/canvas/ink_painters.dart): points
// are scaled from normalized page coords to pixels, perfect_freehand produces
// the outline, and a closed fill Path is built. Keeping ONE implementation here
// kills the hairline-export divergence (R7).
import 'dart:ui';
import 'package:perfect_freehand/perfect_freehand.dart' as pf;
import 'brush.dart';
import 'stroke_model.dart';
feat(pen): pressure-responsive width, configurable thinning, native Windows pen (tilt/buttons) W1 — Custom pen width + pressure sensitivity (Saber-style): - Root cause of "压感没用": perfect_freehand 1.0.4 IGNORES real stylus pressure (hardcodes radius=size/2 when simulatePressure=false) — width never tracked pen force. Upgraded perfect_freehand ^1.0.0 -> ^2.0.0 (honors real pressure); migrated all 5 getStroke call sites to the 2.x API (PointVector / StrokeOptions / Offset). - De-hardcoded `thinning` into `kDefaultPenThinning` (0.85), single source shared by the on-screen painter and the PDF export path; exposed as PenConfig.pressureSensitivity with a Pressure Sensitivity slider; live-applies via a config listener. W3 — Native Windows pen plugin (tilt + barrel/eraser buttons): - windows/runner/pen_channel.{h,cpp}: observe WM_POINTER at the TOP of MessageHandler (before HandleTopLevelWindowProc, which Flutter uses to consume pen events), read GetPointerPenInfo penFlags + tilt, stream over EventChannel('badnote/pen'); non-consuming. - PenInputService: single latched hardware state (no Win32-pointerId<->event.pointer correlation); graceful no-op off-Windows. - pen_canvas maps barrel/inverted/eraser through PenConfig.sideButton/eraserEnd (eraser/undo/toggleTool/pan) and captures tilt into PenPoint.tilt -> EditorPoint.tilt. W2 — Zoom flicker: page raster isolated in its own RepaintBoundary (safe interim); definitive crisp-on-zoom fix gated on the on-device root-cause probe (plan M3). Plans: ralplan-consensus plan at docs/plans/2026-06-22-badnote-pen-polish.md (Architect APPROVE-WITH-MUST-FIX M1-M4 + Critic ITERATE->APPROVE). Tests: 58/58 pass incl. shared-thinning invariant + thinning-affects-outline + tilt-adapter round-trip. flutter analyze clean; linux debug build OK. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 02:10:05 +08:00
/// Canonical default for perfect_freehand's `thinning` (how strongly pressure
/// modulates stroke width). The SINGLE source of truth shared by the on-screen
/// painter ([buildStrokeOutline] here and `ink_painters.buildStrokePath`) and
/// the PDF export path, so screen and export can never diverge. `0.85` =
/// pressure visibly sweeps width; preserves the existing feel + export golden.
/// Overridable per-stroke via [PenConfig.pressureSensitivity].
const double kDefaultPenThinning = 0.85;
fix(zoom): drive pinch absolutely from gesture-start snapshot; snappier pen Zoom flicker root cause (from the on-device badnote_input_log.txt): the pinch computed its per-frame scale change as desiredScale / getMaxScaleOnAxis(), i.e. it fed the LIVE matrix back into its own update. Consecutive frames in the log show `cur` (the live read) dropping to 0.75-0.89 for a single frame while the result track stayed smooth, so the code demanded a 1.3-1.4x correction that popped the zoom bigger/smaller and snapped back. The >1.4 glitch guard missed it because the spikes sat at 1.31-1.40. Fix: the scale branch of PenInteractiveViewer now drives the transform ABSOLUTELY from a gesture-start snapshot (_scaleStart, _referenceFocalPoint) plus the recognizer's clean, monotonic cumulative details.scale. Each frame is fully re-derived in closed form (pure scale+translate, no matrix inversion, no live read-back), so a transient mis-read or interleaved write cannot survive into the next frame. The per-frame glitch guard now keys on the recognizer's own scale-ratio (the true finger motion) instead of the corrupted live read. 2-finger pan still falls out of the same focal-anchor formula. Pen feel: lower perfect_freehand streamline 0.5 -> 0.32 (new shared constants kPenStreamline/kPenSmoothing, single-sourced across screen + export so the parity test still holds). At 0.5 a quick flick lagged so far behind the pen that short fast strokes collapsed toward their start and rendered as a dot ("写字识别成单击"); 0.32 tracks the real path for a crisper, lower-latency feel. flutter analyze lib/editor clean; 66/66 tests pass (incl. screen==export parity). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-23 01:45:54 +08:00
/// perfect_freehand input-smoothing parameters, shared (single source of truth)
/// by the on-screen painter and the export path so the two can never diverge
/// (guarded by the screen==export parity test). [kPenStreamline] lowers the
/// per-point lag from freehand's 0.5 default to 0.32: at 0.5 a quick flick lags
/// so far behind the pen that a short fast stroke collapsed toward its start and
/// rendered as a dot ("写字识别成单击") and the pen felt sluggish; 0.32 tracks the
/// real path closely (crisper, lower-latency feel) while still damping digitizer
/// jitter. [kPenSmoothing] keeps freehand's 0.5 corner rounding.
const double kPenStreamline = 0.32;
const double kPenSmoothing = 0.5;
/// THE single perfect_freehand outline recipe — the raw outline points for a
/// stroke. Both the on-screen painter ([buildStrokeOutline] / the live
/// `ink_painters.buildStrokePath`) and the PDF export
/// (`pdf_service._buildFreehandPdfPath`) call THIS, so the `StrokeOptions`
/// (thinning / smoothing / streamline / simulatePressure) live in exactly one
/// place and screen↔export can never drift again (R7 — the hairline-export bug
/// was pdf_service hardcoding its own `thinning: 0.7, streamline: 0.5`).
///
/// Callers supply already-pixel-scaled [pfPoints] (because the two stroke
/// models scale differently) plus the per-stroke flags. Returns the closed
/// outline as `List<Offset>` (perfect_freehand 2.x); empty when freehand
/// produces nothing.
List<Offset> freehandOutlinePoints({
required List<pf.PointVector> pfPoints,
required double size,
required bool isHighlighter,
required bool hasRealPressure,
required bool isComplete,
double thinning = kDefaultPenThinning,
BrushProfile? brush,
}) {
if (pfPoints.isEmpty) return const <Offset>[];
return pf.getStroke(
pfPoints,
options: brush != null
// Brush-driven path: every geometry knob (thinning / streamline /
// smoothing / caps / taper / simulatePressure) comes from the
// BrushProfile so each brush renders distinctly. Pressure was already
// pre-warped by the brush's gamma at CAPTURE (PressureCurve), so the
// pre-warp is baked into pfPoints — perfect_freehand stays linear here.
// simulatePressure is forced true only when the device gave us NO real
// pressure, so velocity-thinning still kicks in for mice/trackpads.
? _optionsFromBrush(brush,
size: size,
isComplete: isComplete,
hasRealPressure: hasRealPressure)
: pf.StrokeOptions(
size: size,
// Highlighter keeps a constant width (no thinning); pen uses the
// configurable [thinning] so Surface-Pen pressure changes width.
thinning: isHighlighter ? 0.0 : thinning,
smoothing: kPenSmoothing,
streamline: kPenStreamline,
// Real stylus pressure -> don't simulate; no pressure -> let
// freehand fake it based on velocity (highlighter never simulates).
// perfect_freehand 2.x honors real pressure when simulatePressure
// is false.
simulatePressure: !hasRealPressure && !isHighlighter,
isComplete: isComplete,
),
);
}
/// Build perfect_freehand [pf.StrokeOptions] from a [BrushProfile] (spec §4).
///
/// TODO(brush-opacity): [BrushProfile.opacity] / [BrushProfile.blendMultiply]
/// are NOT consumed here — geometry only this increment. The caller still paints
/// fill color (with its own alpha) and BlendMode.srcOver; per-stroke opacity /
/// BlendMode.multiply for ballpoint/pencil/highlighter lands later.
pf.StrokeOptions _optionsFromBrush(
BrushProfile brush, {
required double size,
required bool isComplete,
required bool hasRealPressure,
}) {
return pf.StrokeOptions(
size: size,
thinning: brush.pfThinning,
smoothing: brush.pfSmoothing,
streamline: brush.pfStreamline,
// Honor REAL pressure (already gamma-pre-warped at capture). Only fall back
// to velocity simulation when the device reported no usable pressure.
simulatePressure: brush.simulatePressure || !hasRealPressure,
start: pf.StrokeEndOptions.start(
cap: brush.capStart,
taperEnabled: brush.taper,
),
end: pf.StrokeEndOptions.end(
cap: brush.capEnd,
taperEnabled: brush.taper,
),
isComplete: isComplete,
);
}
/// Builds a closed, fillable outline [Path] for one [stroke], scaled into the
/// pixel space of [pageSize] (which maps normalized [0,1] coords to pixels).
///
/// [isComplete] should be false for the in-progress live stroke so freehand
/// tapers the trailing end correctly, and true for committed strokes.
///
feat(pen): pressure-responsive width, configurable thinning, native Windows pen (tilt/buttons) W1 — Custom pen width + pressure sensitivity (Saber-style): - Root cause of "压感没用": perfect_freehand 1.0.4 IGNORES real stylus pressure (hardcodes radius=size/2 when simulatePressure=false) — width never tracked pen force. Upgraded perfect_freehand ^1.0.0 -> ^2.0.0 (honors real pressure); migrated all 5 getStroke call sites to the 2.x API (PointVector / StrokeOptions / Offset). - De-hardcoded `thinning` into `kDefaultPenThinning` (0.85), single source shared by the on-screen painter and the PDF export path; exposed as PenConfig.pressureSensitivity with a Pressure Sensitivity slider; live-applies via a config listener. W3 — Native Windows pen plugin (tilt + barrel/eraser buttons): - windows/runner/pen_channel.{h,cpp}: observe WM_POINTER at the TOP of MessageHandler (before HandleTopLevelWindowProc, which Flutter uses to consume pen events), read GetPointerPenInfo penFlags + tilt, stream over EventChannel('badnote/pen'); non-consuming. - PenInputService: single latched hardware state (no Win32-pointerId<->event.pointer correlation); graceful no-op off-Windows. - pen_canvas maps barrel/inverted/eraser through PenConfig.sideButton/eraserEnd (eraser/undo/toggleTool/pan) and captures tilt into PenPoint.tilt -> EditorPoint.tilt. W2 — Zoom flicker: page raster isolated in its own RepaintBoundary (safe interim); definitive crisp-on-zoom fix gated on the on-device root-cause probe (plan M3). Plans: ralplan-consensus plan at docs/plans/2026-06-22-badnote-pen-polish.md (Architect APPROVE-WITH-MUST-FIX M1-M4 + Critic ITERATE->APPROVE). Tests: 58/58 pass incl. shared-thinning invariant + thinning-affects-outline + tilt-adapter round-trip. flutter analyze clean; linux debug build OK. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 02:10:05 +08:00
/// [thinning] is perfect_freehand's pressure→width response (see
/// [kDefaultPenThinning]); highlighter always forces `0.0` (constant width).
///
/// Returns an empty [Path] when the stroke has no points (or freehand produces
/// no outline).
Path buildStrokeOutline(
EditorStroke stroke,
Size pageSize, {
required bool isComplete,
feat(pen): pressure-responsive width, configurable thinning, native Windows pen (tilt/buttons) W1 — Custom pen width + pressure sensitivity (Saber-style): - Root cause of "压感没用": perfect_freehand 1.0.4 IGNORES real stylus pressure (hardcodes radius=size/2 when simulatePressure=false) — width never tracked pen force. Upgraded perfect_freehand ^1.0.0 -> ^2.0.0 (honors real pressure); migrated all 5 getStroke call sites to the 2.x API (PointVector / StrokeOptions / Offset). - De-hardcoded `thinning` into `kDefaultPenThinning` (0.85), single source shared by the on-screen painter and the PDF export path; exposed as PenConfig.pressureSensitivity with a Pressure Sensitivity slider; live-applies via a config listener. W3 — Native Windows pen plugin (tilt + barrel/eraser buttons): - windows/runner/pen_channel.{h,cpp}: observe WM_POINTER at the TOP of MessageHandler (before HandleTopLevelWindowProc, which Flutter uses to consume pen events), read GetPointerPenInfo penFlags + tilt, stream over EventChannel('badnote/pen'); non-consuming. - PenInputService: single latched hardware state (no Win32-pointerId<->event.pointer correlation); graceful no-op off-Windows. - pen_canvas maps barrel/inverted/eraser through PenConfig.sideButton/eraserEnd (eraser/undo/toggleTool/pan) and captures tilt into PenPoint.tilt -> EditorPoint.tilt. W2 — Zoom flicker: page raster isolated in its own RepaintBoundary (safe interim); definitive crisp-on-zoom fix gated on the on-device root-cause probe (plan M3). Plans: ralplan-consensus plan at docs/plans/2026-06-22-badnote-pen-polish.md (Architect APPROVE-WITH-MUST-FIX M1-M4 + Critic ITERATE->APPROVE). Tests: 58/58 pass incl. shared-thinning invariant + thinning-affects-outline + tilt-adapter round-trip. flutter analyze clean; linux debug build OK. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 02:10:05 +08:00
double thinning = kDefaultPenThinning,
}) {
final path = Path();
if (stroke.points.isEmpty) return path;
final pfPoints = stroke.points
.map(
feat(pen): pressure-responsive width, configurable thinning, native Windows pen (tilt/buttons) W1 — Custom pen width + pressure sensitivity (Saber-style): - Root cause of "压感没用": perfect_freehand 1.0.4 IGNORES real stylus pressure (hardcodes radius=size/2 when simulatePressure=false) — width never tracked pen force. Upgraded perfect_freehand ^1.0.0 -> ^2.0.0 (honors real pressure); migrated all 5 getStroke call sites to the 2.x API (PointVector / StrokeOptions / Offset). - De-hardcoded `thinning` into `kDefaultPenThinning` (0.85), single source shared by the on-screen painter and the PDF export path; exposed as PenConfig.pressureSensitivity with a Pressure Sensitivity slider; live-applies via a config listener. W3 — Native Windows pen plugin (tilt + barrel/eraser buttons): - windows/runner/pen_channel.{h,cpp}: observe WM_POINTER at the TOP of MessageHandler (before HandleTopLevelWindowProc, which Flutter uses to consume pen events), read GetPointerPenInfo penFlags + tilt, stream over EventChannel('badnote/pen'); non-consuming. - PenInputService: single latched hardware state (no Win32-pointerId<->event.pointer correlation); graceful no-op off-Windows. - pen_canvas maps barrel/inverted/eraser through PenConfig.sideButton/eraserEnd (eraser/undo/toggleTool/pan) and captures tilt into PenPoint.tilt -> EditorPoint.tilt. W2 — Zoom flicker: page raster isolated in its own RepaintBoundary (safe interim); definitive crisp-on-zoom fix gated on the on-device root-cause probe (plan M3). Plans: ralplan-consensus plan at docs/plans/2026-06-22-badnote-pen-polish.md (Architect APPROVE-WITH-MUST-FIX M1-M4 + Critic ITERATE->APPROVE). Tests: 58/58 pass incl. shared-thinning invariant + thinning-affects-outline + tilt-adapter round-trip. flutter analyze clean; linux debug build OK. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 02:10:05 +08:00
(p) => pf.PointVector(
p.x * pageSize.width,
p.y * pageSize.height,
p.pressure ?? 0.5,
),
)
.toList();
// Resolve the brush so each stroke renders with its own geometry. The
// pressure pre-warp ([BrushProfile.pressureGamma]) was already applied at
// capture, so it is baked into the points here.
final brush = brushProfileFor(stroke.brush);
final outline = freehandOutlinePoints(
pfPoints: pfPoints,
size: stroke.width * pageSize.width,
isHighlighter: stroke.tool == EditorTool.highlighter,
hasRealPressure: stroke.points.any((p) => p.pressure != null),
isComplete: isComplete,
thinning: thinning,
brush: brush,
);
if (outline.isEmpty) return path;
feat(pen): pressure-responsive width, configurable thinning, native Windows pen (tilt/buttons) W1 — Custom pen width + pressure sensitivity (Saber-style): - Root cause of "压感没用": perfect_freehand 1.0.4 IGNORES real stylus pressure (hardcodes radius=size/2 when simulatePressure=false) — width never tracked pen force. Upgraded perfect_freehand ^1.0.0 -> ^2.0.0 (honors real pressure); migrated all 5 getStroke call sites to the 2.x API (PointVector / StrokeOptions / Offset). - De-hardcoded `thinning` into `kDefaultPenThinning` (0.85), single source shared by the on-screen painter and the PDF export path; exposed as PenConfig.pressureSensitivity with a Pressure Sensitivity slider; live-applies via a config listener. W3 — Native Windows pen plugin (tilt + barrel/eraser buttons): - windows/runner/pen_channel.{h,cpp}: observe WM_POINTER at the TOP of MessageHandler (before HandleTopLevelWindowProc, which Flutter uses to consume pen events), read GetPointerPenInfo penFlags + tilt, stream over EventChannel('badnote/pen'); non-consuming. - PenInputService: single latched hardware state (no Win32-pointerId<->event.pointer correlation); graceful no-op off-Windows. - pen_canvas maps barrel/inverted/eraser through PenConfig.sideButton/eraserEnd (eraser/undo/toggleTool/pan) and captures tilt into PenPoint.tilt -> EditorPoint.tilt. W2 — Zoom flicker: page raster isolated in its own RepaintBoundary (safe interim); definitive crisp-on-zoom fix gated on the on-device root-cause probe (plan M3). Plans: ralplan-consensus plan at docs/plans/2026-06-22-badnote-pen-polish.md (Architect APPROVE-WITH-MUST-FIX M1-M4 + Critic ITERATE->APPROVE). Tests: 58/58 pass incl. shared-thinning invariant + thinning-affects-outline + tilt-adapter round-trip. flutter analyze clean; linux debug build OK. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 02:10:05 +08:00
path.moveTo(outline.first.dx, outline.first.dy);
for (var i = 1; i < outline.length; i++) {
feat(pen): pressure-responsive width, configurable thinning, native Windows pen (tilt/buttons) W1 — Custom pen width + pressure sensitivity (Saber-style): - Root cause of "压感没用": perfect_freehand 1.0.4 IGNORES real stylus pressure (hardcodes radius=size/2 when simulatePressure=false) — width never tracked pen force. Upgraded perfect_freehand ^1.0.0 -> ^2.0.0 (honors real pressure); migrated all 5 getStroke call sites to the 2.x API (PointVector / StrokeOptions / Offset). - De-hardcoded `thinning` into `kDefaultPenThinning` (0.85), single source shared by the on-screen painter and the PDF export path; exposed as PenConfig.pressureSensitivity with a Pressure Sensitivity slider; live-applies via a config listener. W3 — Native Windows pen plugin (tilt + barrel/eraser buttons): - windows/runner/pen_channel.{h,cpp}: observe WM_POINTER at the TOP of MessageHandler (before HandleTopLevelWindowProc, which Flutter uses to consume pen events), read GetPointerPenInfo penFlags + tilt, stream over EventChannel('badnote/pen'); non-consuming. - PenInputService: single latched hardware state (no Win32-pointerId<->event.pointer correlation); graceful no-op off-Windows. - pen_canvas maps barrel/inverted/eraser through PenConfig.sideButton/eraserEnd (eraser/undo/toggleTool/pan) and captures tilt into PenPoint.tilt -> EditorPoint.tilt. W2 — Zoom flicker: page raster isolated in its own RepaintBoundary (safe interim); definitive crisp-on-zoom fix gated on the on-device root-cause probe (plan M3). Plans: ralplan-consensus plan at docs/plans/2026-06-22-badnote-pen-polish.md (Architect APPROVE-WITH-MUST-FIX M1-M4 + Critic ITERATE->APPROVE). Tests: 58/58 pass incl. shared-thinning invariant + thinning-affects-outline + tilt-adapter round-trip. flutter analyze clean; linux debug build OK. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-22 02:10:05 +08:00
path.lineTo(outline[i].dx, outline[i].dy);
}
path.close();
return path;
}