fix(studio): a heal stroke is one mask, and the fill comes from its rim
The dust brush had two faults and the second was hidden behind the first. A stroke was a row of independent repairs. Every circle the brush laid down found a patch of its own somewhere else in the frame and pasted it, so a stroke of twenty spots was twenty pastes with twenty rims to match, and what showed on the photo was the run of circles itself. The mask was their union on paper but not on the picture. And the mask was not a stroke either. The brush only ever dropped a spot where the pointer's own sample fell, and a fast drag reports a sample every few centimetres of screen — several brush radii apart. Circles that far apart do not touch, so their union was the row of dots the tool was complained about. The segment the pointer travelled is now walked at the brush's own spacing and a spot is laid at every step, whatever speed the stroke was drawn at. With the mask one shape, the patch had nothing left to justify it. The fill is now a push-pull pyramid over the whole mask: the painted pixels are cut out of the frame as a hole, the remaining pixels are halved and halved again until a level is small enough that no hole is left in it, and each level is drawn back over the one below it. What comes up out of that is the photo where the photo was, and the light of the mask's own rim carried inward where it was not — nothing is copied from anywhere, and a repair matches the grade, the blur and the grain around it because those are the pixels it was made from. Non-destructive as before: the recipe still holds the circles, and the preview and the full-size export run the two passes in engine/exportEngine.ts, the shaders and the mask stops in shared/utils/heal.ts. A recipe written by the old brush still loads; the patch coordinates it stored are simply not read any more.
This commit is contained in:
@@ -1,324 +1,111 @@
|
||||
import type { HealSpot } from '../types';
|
||||
|
||||
// FX tab > HEAL — the dust brush, and the patch search behind it.
|
||||
// FX tab > HEAL — the dust brush, and the fill behind it.
|
||||
//
|
||||
// A spot is a circle on the rendered photo plus the patch it borrows: the
|
||||
// renderer copies the pixels at (sx, sy) onto (x, y), feathers the edge, and
|
||||
// shifts what it pasted onto the light of the place it landed in, so a repair
|
||||
// is a draw of the same picture rather than a blur over the dust. All five
|
||||
// numbers are fractions of the render — x/y/sx/sy of its width and height, r of
|
||||
// its width — which is what makes one set of spots survive the preview and the
|
||||
// export rendering the same photo at two sizes, and keeps the circle round
|
||||
// whatever the photo's shape.
|
||||
// What the brush paints is a mask: the circles it laid down, all of them at
|
||||
// once, as ONE shape. What fills that shape is not borrowed from anywhere — the
|
||||
// renderer takes the photo around the mask's rim and works inward, so the
|
||||
// repair is made of the light and the colour that already border it. That is
|
||||
// the whole of the change from a patch brush: a stroke of twenty overlapping
|
||||
// spots used to be twenty independent repairs, each pasting its own patch and
|
||||
// matching its own rim, which is exactly what a run of circles looks like on the
|
||||
// photo; one mask and one solve have no circles in them to see.
|
||||
//
|
||||
// There is no ceiling on the list. The shader is built to carry exactly the
|
||||
// spots the recipe holds (healSkSL), so a new repair can never push an old one
|
||||
// out: the dust you healed first is the dust that stays healed.
|
||||
// Non-destructive like everything else here: the recipe keeps the circles, and
|
||||
// the preview and the export run the same two passes over them.
|
||||
//
|
||||
// The source is SEARCHED for rather than asked for. Lightroom picks the patch
|
||||
// and lets you drag the second circle afterwards; the search below is the same
|
||||
// idea without the second circle, and it is a pure function of a sampler so a
|
||||
// synthetic picture can hold it to account.
|
||||
//
|
||||
// Feather, as a fraction of the radius: inside it the patch is copied, outward
|
||||
// it fades to nothing, so the circle leaves no rim of its own. It is the outer
|
||||
// 15% and no more, because that band is the only place the dust being repaired
|
||||
// is mixed back into the patch — a wide fade keeps the speck's own edge alive
|
||||
// as a faint ring inside the circle, which is a blur of the dust rather than a
|
||||
// repair. sub-pixel at the default brush, still a soft edge at a big one.
|
||||
export const HEAL_FEATHER = 0.85;
|
||||
// x/y are fractions of the render's width and height, r of its width. The
|
||||
// fractions are what make one set of spots land on the preview and on the
|
||||
// full-size export, and keep a spot round whatever the photo's shape.
|
||||
|
||||
// The brush's radius, as a fraction of the photo's width — where this tool
|
||||
// starts, at the sensor-dust end, where a speck is a few thousandths of the
|
||||
// frame. The bounds and the wheel that moves it between them belong to the
|
||||
// brush both FX tools share (brush.ts).
|
||||
// frame. The bounds and the wheel that move it between them belong to the brush
|
||||
// both FX tools share (brush.ts).
|
||||
export const HEAL_DEFAULT_R = 0.012;
|
||||
// How far the search looks, in radii, and how many directions it looks in.
|
||||
const SEARCH_DISTANCES = [2.6, 4.2, 6.5];
|
||||
const SEARCH_DIRS = 8;
|
||||
// The taps that stand for "the light around a spot": a ring just OUTSIDE the
|
||||
// brush, where the dust being repaired is not — the scale the eye reads a
|
||||
// spot's surroundings at. Reading the inside instead compares a candidate with
|
||||
// the very dark the repair is trying to erase, and the patch that matches a
|
||||
// speck best is then the one carrying a speck of its own — which is how a
|
||||
// repair ends up moving dust a few pixels instead of removing it.
|
||||
//
|
||||
// Twelve taps read as their MEDIAN, not their mean. The ring can only be a
|
||||
// little way out — far enough out and it would be reading a light the spot does
|
||||
// not sit in — so a few of its taps land on the speck's own softened edge, or
|
||||
// on a neighbour's: readings in the minority, which the median drops and the
|
||||
// mean would drag the whole light down by.
|
||||
const RING_R = 1.15;
|
||||
const RING_TAPS: [number, number][] = Array.from({ length: 12 }, (_, i) => {
|
||||
const a = (i / 12) * Math.PI * 2;
|
||||
return [Math.cos(a), Math.sin(a)] as [number, number];
|
||||
});
|
||||
// The taps that decide whether a candidate patch is clean: the inside of the
|
||||
// patch, held against the patch's own mean. Dust is an outlier in its own
|
||||
// neighbourhood; skin, however grainy, is not. The last ring is the patch's own
|
||||
// edge — the circle is copied at full strength out to HEAL_FEATHER of its
|
||||
// radius, so that is where a neighbour's dust leaking into the patch shows up,
|
||||
// and the middle of the patch would never see it.
|
||||
const INSIDE_TAPS: [number, number][] = [
|
||||
[0, 0],
|
||||
[-0.5, 0],
|
||||
[0.5, 0],
|
||||
[0, -0.5],
|
||||
[0, 0.5],
|
||||
[-0.35, -0.35],
|
||||
[0.35, -0.35],
|
||||
[-0.35, 0.35],
|
||||
[0.35, 0.35],
|
||||
...Array.from({ length: 8 }, (_, i) => {
|
||||
const a = (i / 8) * Math.PI * 2;
|
||||
return [Math.cos(a) * HEAL_FEATHER, Math.sin(a) * HEAL_FEATHER] as [number, number];
|
||||
}),
|
||||
];
|
||||
// How far a candidate's light may sit from the spot's own before it is refused
|
||||
// outright, in the 0-255 levels the sampler answers in. Sending a patch from a
|
||||
// different light is exactly what makes a repair show up as a mark of its own —
|
||||
// a light square where dark ground was, a dark one where light was — so past
|
||||
// this the search gives up and the caller leaves the speck alone. The user sees
|
||||
// an untouched speck, which is honest, and tries another size or another spot.
|
||||
const LIGHT_GATE = 20;
|
||||
// Light outweighs cleanliness in the score, so the winner is the same light
|
||||
// first and the cleaner patch second; within the gate the cleanliness term only
|
||||
// breaks ties between patches the eye would call the same.
|
||||
const LIGHT_WEIGHT = 3;
|
||||
|
||||
const num = (v: unknown, fallback: number) => {
|
||||
const n = Number(v);
|
||||
return Number.isFinite(n) ? n : fallback;
|
||||
};
|
||||
const clamp01 = (v: number) => (v < 0 ? 0 : v > 1 ? 1 : v);
|
||||
// How much of a spot is painted at full strength, as a fraction of its radius:
|
||||
// out to here the mask is solid, and the last 15% is the feather that carries it
|
||||
// to nothing. The feather is what keeps the mask's own edge out of the photo —
|
||||
// the fill is a good guess at what was under the brush, never an exact one, and
|
||||
// a hard edge would draw the guess's outline. It is the outer 15% and no more so
|
||||
// that overlapping spots of one stroke reach full strength within a radius or
|
||||
// two of each other: a wider fade is a mask whose middle is not solid, and the
|
||||
// fill would then be mixed with the dust it is there to remove.
|
||||
export const HEAL_FEATHER = 0.85;
|
||||
|
||||
// How many halvings of the frame the fill reads, at most. Each level is half of
|
||||
// the one below it, so twelve covers any photo a camera makes and stops a
|
||||
// pathological surface size from walking off the end of the pyramid.
|
||||
export const HEAL_MAX_LEVELS = 12;
|
||||
|
||||
// The stored spots, made readable: numbers, inside the frame. Everything below
|
||||
// reads a recipe through this, so a hand-written or older file cannot produce a
|
||||
// spot the brush and the renderer disagree about.
|
||||
// spot the brush and the renderer disagree about. The patch the older brush
|
||||
// stored (sx/sy) is not read: a repair is no longer a copy of another place.
|
||||
export function readHeal(heal: HealSpot[] | undefined): HealSpot[] {
|
||||
if (!Array.isArray(heal)) return [];
|
||||
return heal
|
||||
.map((s) => ({
|
||||
x: clamp01(num(s?.x, 0)),
|
||||
y: clamp01(num(s?.y, 0)),
|
||||
x: Math.min(1, Math.max(0, num(s?.x, 0))),
|
||||
y: Math.min(1, Math.max(0, num(s?.y, 0))),
|
||||
r: Math.max(0, num(s?.r, 0)),
|
||||
sx: clamp01(num(s?.sx, s?.x ?? 0)),
|
||||
sy: clamp01(num(s?.sy, s?.y ?? 0)),
|
||||
}))
|
||||
.filter((s) => s.r > 0);
|
||||
}
|
||||
|
||||
// The uniform block the shader for `n` spots reads: the circles, the patches,
|
||||
// then the frame the fractions are of. Declaration order, arrays expanded —
|
||||
// that is how the runtime effect wants its uniforms, and one buffer is one
|
||||
// upload per render. Its length is a function of the list, not a fixed
|
||||
// capacity, because the shader carries exactly the spots the recipe holds.
|
||||
export function healUniforms(spots: HealSpot[], width: number, height: number): Float32Array {
|
||||
const list = readHeal(spots);
|
||||
const n = list.length;
|
||||
const u = new Float32Array((n * 2 + 1) * 4);
|
||||
for (let i = 0; i < n; i++) {
|
||||
const s = list[i];
|
||||
u.set([s.x, s.y, s.r, 0], i * 4);
|
||||
u.set([s.sx, s.sy, 0, 0], (n + i) * 4);
|
||||
}
|
||||
u.set([width, height, 0, 0], n * 2 * 4);
|
||||
return u;
|
||||
function num(v: unknown, fallback: number): number {
|
||||
const n = Number(v);
|
||||
return Number.isFinite(n) ? n : fallback;
|
||||
}
|
||||
|
||||
// The ring the pasted circle is matched at, and how its taps are laid out.
|
||||
// Measured just OUTSIDE the brush — RING_R of its radius, where the search
|
||||
// reads a spot's light too — because that is the only ring with nothing of the
|
||||
// repair in it: inside the circle is the borrowed patch at one radius and the
|
||||
// speck being covered at another, and a light read off either of those is the
|
||||
// dust talking rather than the photo. Outside, both sides are photographs — the
|
||||
// ground the repair has to sit in, and the ground the patch was borrowed from.
|
||||
// Sixteen taps are spread around it, at the SAME sixteen places on both sides,
|
||||
// so a tap's pair of readings differ by the light between the two places and
|
||||
// not by where in the photo they were taken. Sixteen rather than eight because
|
||||
// a tap that lands on an edge bends its own direction only as far as the next
|
||||
// ring tap: the finer the ring, the less of the rim one hard pixel of the photo
|
||||
// can reach around.
|
||||
const BLEND_TAPS = 16;
|
||||
// ...and one pixel further out again than RING_R. The ring is read by a pixel
|
||||
// that lies under the brush, so the reading has to clear the dust's own edge,
|
||||
// and that edge is soft: a speck r pixels across spreads about a pixel past
|
||||
// where it is drawn, in the pixels the shader samples. RING_R alone leaves the
|
||||
// nearest tap or two inside that spread, and a tap that lands on the dust is
|
||||
// the dust's light — which is exactly the blotch this reading exists to avoid.
|
||||
// One pixel is the width of the spread, in the same pixels the sampler works
|
||||
// in, so it is a pixel here rather than a fraction of the radius: a fraction
|
||||
// would be nothing at all at the sensor-dust end of the brush.
|
||||
const RING_PAD = 1;
|
||||
|
||||
// ...but one light for the whole circle is a light no ring has when the ring is
|
||||
// not all one thing. A twig against the sky, a hairline, the edge of a table
|
||||
// under the brush: a few taps of the ring land on the thing and the rest on the
|
||||
// ground, and the MEAN of them all is a colour that is neither — a level the
|
||||
// place does not have, pasted over the whole circle. The donor can be of exactly
|
||||
// the right light and still come out as a blotch, which is the mark the repair
|
||||
// was supposed to stop leaving. So the light is read once PER DIRECTION, and a
|
||||
// pixel takes the correction of the two readings it lies between, interpolated
|
||||
// by its own angle around the spot: the rim then meets the place all the way
|
||||
// round instead of on average, and the taps that landed on the twig correct the
|
||||
// part of the rim that is near the twig rather than the whole of it. It is the
|
||||
// same taps and the same one draw — each pixel reads them at its own angle
|
||||
// instead of sharing everyone's single number.
|
||||
|
||||
// One unrolled block per spot. SkSL indexes a uniform array by constant only
|
||||
// (see TONE_SKSL's mixer), so the spots are written out rather than looped, and
|
||||
// the shader is built for the count it is handed rather than for a capacity —
|
||||
// that is what lets the list be uncapped. A count costs one RuntimeEffect to
|
||||
// compile, so the renderer caches them by count (exportEngine's healEffectFor).
|
||||
//
|
||||
// What is pasted is the patch's own pixels SHIFTED onto the light of the place
|
||||
// it lands in — the gradient of the borrowed texture, carried at the level of
|
||||
// the destination: the border is made to match, and everything inside keeps the
|
||||
// texture it was borrowed for. That is the whole of "seamless" the brush needs:
|
||||
// at the rim the patch sits within a level or two of the photo around it,
|
||||
// instead of up to LIGHT_GATE levels off it, so the repair stops reading as a
|
||||
// soft blotch of its own and the feather has almost nothing left to hide. The
|
||||
// correction is read per direction around the ring, so the border is matched
|
||||
// where it is rather than on average, and the pixels of a twig that run through
|
||||
// the ring bend the part of the rim they are near instead of the whole circle.
|
||||
//
|
||||
// ponytail: the correction lives on the rim — every pixel takes the two ring
|
||||
// readings it lies between and interpolates — so it is the boundary of a Poisson
|
||||
// edit and not its interior: a repair laid over something the patch cannot
|
||||
// reproduce still carries the patch's own texture inside, bent to fit the rim.
|
||||
// The other half of the solve (a correction that relaxes inside the circle, a
|
||||
// Jacobi ping-pong per spot) lands when a real photo shows an interior the eye
|
||||
// can find from the rim alone. A direction's level is also one tap pair on each
|
||||
// side, so the grain the mean used to average away now rides the rim as a wedge
|
||||
// of a level or two across the circle — a tangential average of three taps per
|
||||
// direction is the rung for that, when a photo shows the wedge.
|
||||
const spotBlock = (i: number) => `
|
||||
{
|
||||
float4 s = spots[${i}];
|
||||
if (s.z > 0.0) {
|
||||
float rad = s.z * size.x;
|
||||
float d = distance(pos, s.xy * size.xy);
|
||||
if (d < rad) {
|
||||
float4 t = srcs[${i}];
|
||||
float2 pd = s.xy * size.xy;
|
||||
float2 ps = t.xy * size.xy;
|
||||
float r = rad * ${HEAL_FEATHER};
|
||||
float rr = rad * ${RING_R} + ${RING_PAD.toFixed(1)};
|
||||
// Which way round the spot this pixel sits, one millionth of a pixel off
|
||||
// the exact centre so the angle is defined there too.
|
||||
float ang = atan(pos.y - pd.y + 1e-6, pos.x - pd.x + 1e-6);
|
||||
half3 corr = half3(0.0);
|
||||
float wsum = 0.0;
|
||||
for (int k = 0; k < ${BLEND_TAPS}; k++) {
|
||||
float a = float(k) * 6.283185307 / float(${BLEND_TAPS});
|
||||
float2 o = float2(cos(a), sin(a)) * rr;
|
||||
// How far this pixel's angle is from this tap's, the short way round.
|
||||
float da = ang - a;
|
||||
da = abs(da - 6.283185307 * floor(da * 0.1591549431 + 0.5));
|
||||
float w = max(0.0, 1.0 - da * float(${BLEND_TAPS}) * 0.1591549431);
|
||||
corr += half(w) * (img.eval(pd + o).rgb - img.eval(ps + o).rgb);
|
||||
wsum += w;
|
||||
}
|
||||
corr *= half(1.0 / wsum);
|
||||
half m = half(1.0 - smoothstep(r, rad, d));
|
||||
c = mix(c, img.eval(pos + ps - pd) + half4(corr, 0.0), m);
|
||||
}
|
||||
}
|
||||
}
|
||||
`;
|
||||
|
||||
// The pass. It reads the pixels the pipeline has already built (the child is a
|
||||
// snapshot of the surface) and writes the borrowed patches back over them, so a
|
||||
// repair is one draw: no blur, no smoothing, and the grain and the frame land
|
||||
// on top of it afterwards exactly as they land on the rest of the photo.
|
||||
export function healSkSL(count: number): string {
|
||||
return `
|
||||
// The first of the fill's two passes: the photo with the painted pixels taken
|
||||
// out of it. The colour of every pixel is kept and its alpha drops to what the
|
||||
// brush did not touch, which is what turns the mask into a hole the second pass
|
||||
// can see the rim of. Written as premultiplied — rgb times that alpha — because
|
||||
// that is what every surface here holds, and the pyramid below is built on it:
|
||||
// a linear halving of premultiplied pixels is a weighted average of the pixels
|
||||
// that are still there, with the hole weighing nothing.
|
||||
export const HEAL_CUT_SKSL = `
|
||||
uniform shader img;
|
||||
uniform float4 spots[${count}];
|
||||
uniform float4 srcs[${count}];
|
||||
uniform float4 size;
|
||||
uniform shader mask;
|
||||
|
||||
half4 main(float2 pos) {
|
||||
half4 c = img.eval(pos);${Array.from({ length: count }, (_, i) => spotBlock(i)).join('')}
|
||||
return c;
|
||||
half4 c = img.eval(pos);
|
||||
half a = 1.0 - mask.eval(pos).a;
|
||||
half3 straight = c.a > 0.0001 ? c.rgb / c.a : c.rgb;
|
||||
return half4(straight * a, a);
|
||||
}
|
||||
`;
|
||||
}
|
||||
|
||||
// The patch to borrow for a spot at (x, y) of radius r, from a sampler that
|
||||
// answers fractions of the same photo. The candidates are a ring of offsets in
|
||||
// eight directions at three distances — the patch has to be far enough that the
|
||||
// dust is not in it, near enough that the light is the same — plus each one
|
||||
// mirrored through the spot, which is the pair Lightroom's own auto-source
|
||||
// leans on.
|
||||
// The second pass: the fill laid back under the brush. `fill` is the pyramid's
|
||||
// answer for every pixel of the frame — at the mask it is what the rim implies,
|
||||
// away from it it is the photo itself — and the mask is the weight it is mixed
|
||||
// at, so the rim is the only place the two meet and they already agree there.
|
||||
//
|
||||
// Each candidate is asked two questions. "Is this the same light?" holds the
|
||||
// mean of what would be pasted against the mean of the spot's own ring, the
|
||||
// pixels just outside the dust — the light the repair has to sit in, read where
|
||||
// the dust is not, because comparing against the spot's own inside compares a
|
||||
// candidate with the very dark the repair is trying to erase, and the patch
|
||||
// that matches a speck best is then the one carrying a speck of its own. "Is
|
||||
// there dust on it?" holds the candidate's inside against its own inside mean:
|
||||
// dust is an outlier in its own neighbourhood, grain is not. A candidate whose
|
||||
// light is off by more than LIGHT_GATE is not scored at all — a patch from the
|
||||
// wrong light is a mark of its own, and moving it a few pixels for a wrong
|
||||
// clone is not a repair.
|
||||
//
|
||||
// Returns null when the frame is too small to hold any candidate, or when every
|
||||
// candidate is refused: the caller then leaves the speck where it is rather
|
||||
// than pasting a patch it cannot stand behind.
|
||||
export function findHealSource(
|
||||
sample: (fx: number, fy: number) => { r: number; g: number; b: number },
|
||||
x: number,
|
||||
y: number,
|
||||
r: number
|
||||
): { sx: number; sy: number } | null {
|
||||
if (!(r > 0)) return null;
|
||||
const inside = (cx: number, cy: number) => cx - r >= 0 && cx + r <= 1 && cy - r >= 0 && cy + r <= 1;
|
||||
const rgb = (v: { r: number; g: number; b: number }) => (v.r + v.g + v.b) / 3;
|
||||
// The light the repair has to sit in: the median of the spot's ring, where
|
||||
// the dust is not. Measured once, because it is the same number for every
|
||||
// candidate.
|
||||
const ring = RING_TAPS.map(([dx, dy]) => rgb(sample(clamp01(x + dx * RING_R * r), clamp01(y + dy * RING_R * r)))).sort(
|
||||
(a, b) => a - b
|
||||
);
|
||||
const target = (ring[RING_TAPS.length / 2 - 1] + ring[RING_TAPS.length / 2]) / 2;
|
||||
// The score of one candidate: null when it is refused for its light, else how
|
||||
// far off that light is times its weight, plus how much the patch varies
|
||||
// inside itself.
|
||||
const score = (cx: number, cy: number): number | null => {
|
||||
let content = 0;
|
||||
const inside: number[] = [];
|
||||
for (const [dx, dy] of INSIDE_TAPS) {
|
||||
const v = rgb(sample(clamp01(cx + dx * r), clamp01(cy + dy * r)));
|
||||
inside.push(v);
|
||||
content += v;
|
||||
}
|
||||
content /= INSIDE_TAPS.length;
|
||||
const light = Math.abs(target - content);
|
||||
if (light > LIGHT_GATE) return null;
|
||||
let dirty = 0;
|
||||
for (const v of inside) dirty += Math.abs(v - content);
|
||||
return light * LIGHT_WEIGHT + dirty / INSIDE_TAPS.length;
|
||||
};
|
||||
let best: { sx: number; sy: number; score: number } | null = null;
|
||||
for (let d = 0; d < SEARCH_DIRS; d++) {
|
||||
const a = (d / SEARCH_DIRS) * Math.PI * 2;
|
||||
for (const dist of SEARCH_DISTANCES) {
|
||||
const cx = x + Math.cos(a) * dist * r;
|
||||
const cy = y + Math.sin(a) * dist * r;
|
||||
for (const [px, py] of [
|
||||
[cx, cy],
|
||||
[2 * x - cx, 2 * y - cy],
|
||||
]) {
|
||||
if (!inside(px, py)) continue;
|
||||
const s = score(px, py);
|
||||
if (s === null) continue;
|
||||
// A tie keeps the earlier candidate: the ring is walked from the right,
|
||||
// so the patch nearest the spot wins — the one most likely to share its
|
||||
// light.
|
||||
if (!best || s < best.score) best = { sx: px, sy: py, score: s };
|
||||
}
|
||||
}
|
||||
}
|
||||
return best ? { sx: best.sx, sy: best.sy } : null;
|
||||
// An alpha of nothing means the fill found no photo anywhere to work from (a
|
||||
// mask over the whole frame): the pixel is left as it was rather than painted
|
||||
// black, which is the one thing the user can still recognise as a mistake.
|
||||
export const HEAL_FILL_SKSL = `
|
||||
uniform shader img;
|
||||
uniform shader fill;
|
||||
uniform shader mask;
|
||||
|
||||
half4 main(float2 pos) {
|
||||
half4 c = img.eval(pos);
|
||||
half m = mask.eval(pos).a;
|
||||
if (m <= 0.0) return c;
|
||||
half4 f = fill.eval(pos);
|
||||
half3 here = c.a > 0.0001 ? c.rgb / c.a : c.rgb;
|
||||
half3 made = f.a > 0.0001 ? f.rgb / f.a : here;
|
||||
return half4(mix(here, made, m) * c.a, c.a);
|
||||
}
|
||||
`;
|
||||
|
||||
// How the mask is painted onto a surface, as the stops of one radial gradient
|
||||
// per spot: solid white out to HEAL_FEATHER of the radius, nothing at its edge,
|
||||
// a straight ramp between. The renderer builds the gradient per spot (Skia has
|
||||
// no one paint for a list of circles) and draws each at its own radius, so a
|
||||
// brush turned down to three pixels and one at a quarter of the frame feather by
|
||||
// the same fraction of themselves.
|
||||
export const HEAL_MASK_STOPS: number[] = [0, HEAL_FEATHER, 1];
|
||||
|
||||
Reference in New Issue
Block a user