/*
* Copyright 2006 The Android Open Source Project
*
* Use of this source code is governed by a BSD-style license that can be
* found in the LICENSE file.
*/
#include "SkBlurMask.h"
#include "SkBlurPriv.h"
#include "SkGpuBlurUtils.h"
#include "SkMaskFilterBase.h"
#include "SkReadBuffer.h"
#include "SkRRectPriv.h"
#include "SkWriteBuffer.h"
#include "SkMaskFilter.h"
#include "SkRRect.h"
#include "SkStringUtils.h"
#include "SkStrokeRec.h"
#include "SkVertices.h"
#if SK_SUPPORT_GPU
#include "GrClip.h"
#include "GrFragmentProcessor.h"
#include "GrRecordingContext.h"
#include "GrRecordingContextPriv.h"
#include "GrRenderTargetContext.h"
#include "GrResourceProvider.h"
#include "GrShaderCaps.h"
#include "GrShape.h"
#include "GrStyle.h"
#include "GrTextureProxy.h"
#include "effects/GrCircleBlurFragmentProcessor.h"
#include "effects/GrRectBlurEffect.h"
#include "effects/GrRRectBlurEffect.h"
#include "effects/GrSimpleTextureEffect.h"
#include "effects/GrTextureDomain.h"
#include "glsl/GrGLSLFragmentProcessor.h"
#include "glsl/GrGLSLFragmentShaderBuilder.h"
#include "glsl/GrGLSLProgramDataManager.h"
#include "glsl/GrGLSLUniformHandler.h"
#endif
class SkBlurMaskFilterImpl : public SkMaskFilterBase {
public:
SkBlurMaskFilterImpl(SkScalar sigma, SkBlurStyle, bool respectCTM);
// overrides from SkMaskFilter
SkMask::Format getFormat() const override;
bool filterMask(SkMask* dst, const SkMask& src, const SkMatrix&,
SkIPoint* margin) const override;
#if SK_SUPPORT_GPU
bool canFilterMaskGPU(const GrShape& shape,
const SkIRect& devSpaceShapeBounds,
const SkIRect& clipBounds,
const SkMatrix& ctm,
SkIRect* maskRect) const override;
bool directFilterMaskGPU(GrRecordingContext*,
GrRenderTargetContext* renderTargetContext,
GrPaint&&,
const GrClip&,
const SkMatrix& viewMatrix,
const GrShape& shape) const override;
sk_sp<GrTextureProxy> filterMaskGPU(GrRecordingContext*,
sk_sp<GrTextureProxy> srcProxy,
const SkMatrix& ctm,
const SkIRect& maskRect) const override;
#endif
void computeFastBounds(const SkRect&, SkRect*) const override;
bool asABlur(BlurRec*) const override;
protected:
FilterReturn filterRectsToNine(const SkRect[], int count, const SkMatrix&,
const SkIRect& clipBounds,
NinePatch*) const override;
FilterReturn filterRRectToNine(const SkRRect&, const SkMatrix&,
const SkIRect& clipBounds,
NinePatch*) const override;
bool filterRectMask(SkMask* dstM, const SkRect& r, const SkMatrix& matrix,
SkIPoint* margin, SkMask::CreateMode createMode) const;
bool filterRRectMask(SkMask* dstM, const SkRRect& r, const SkMatrix& matrix,
SkIPoint* margin, SkMask::CreateMode createMode) const;
bool ignoreXform() const { return !fRespectCTM; }
private:
SK_FLATTENABLE_HOOKS(SkBlurMaskFilterImpl)
// To avoid unseemly allocation requests (esp. for finite platforms like
// handset) we limit the radius so something manageable. (as opposed to
// a request like 10,000)
static const SkScalar kMAX_BLUR_SIGMA;
SkScalar fSigma;
SkBlurStyle fBlurStyle;
bool fRespectCTM;
SkBlurMaskFilterImpl(SkReadBuffer&);
void flatten(SkWriteBuffer&) const override;
SkScalar computeXformedSigma(const SkMatrix& ctm) const {
SkScalar xformedSigma = this->ignoreXform() ? fSigma : ctm.mapRadius(fSigma);
return SkMinScalar(xformedSigma, kMAX_BLUR_SIGMA);
}
friend class SkBlurMaskFilter;
typedef SkMaskFilter INHERITED;
friend void sk_register_blur_maskfilter_createproc();
};
const SkScalar SkBlurMaskFilterImpl::kMAX_BLUR_SIGMA = SkIntToScalar(128);
// linearly interpolate between y1 & y3 to match x2's position between x1 & x3
static SkScalar interp(SkScalar x1, SkScalar x2, SkScalar x3, SkScalar y1, SkScalar y3) {
SkASSERT(x1 <= x2 && x2 <= x3);
SkASSERT(y1 <= y3);
SkScalar t = (x2 - x1) / (x3 - x1);
return y1 + t * (y3 - y1);
}
// Insert 'lower' and 'higher' into 'array1' and insert a new value at each matching insertion
// point in 'array2' that linearly interpolates between the existing values.
// Return a bit mask which contains a copy of 'inputMask' for all the cells between the two
// insertion points.
static uint32_t insert_into_arrays(SkScalar* array1, SkScalar* array2,
SkScalar lower, SkScalar higher,
int* num, uint32_t inputMask, int maskSize) {
SkASSERT(lower < higher);
SkASSERT(lower >= array1[0] && higher <= array1[*num-1]);
int32_t skipMask = 0x0;
int i;
for (i = 0; i < *num; ++i) {
if (lower >= array1[i] && lower < array1[i+1]) {
if (!SkScalarNearlyEqual(lower, array1[i])) {
memmove(&array1[i+2], &array1[i+1], (*num-i-1)*sizeof(SkScalar));
array1[i+1] = lower;
memmove(&array2[i+2], &array2[i+1], (*num-i-1)*sizeof(SkScalar));
array2[i+1] = interp(array1[i], lower, array1[i+2], array2[i], array2[i+2]);
i++;
(*num)++;
}
break;
}
}
for ( ; i < *num; ++i) {
skipMask |= inputMask << (i*maskSize);
if (higher > array1[i] && higher <= array1[i+1]) {
if (!SkScalarNearlyEqual(higher, array1[i+1])) {
memmove(&array1[i+2], &array1[i+1], (*num-i-1)*sizeof(SkScalar));
array1[i+1] = higher;
memmove(&array2[i+2], &array2[i+1], (*num-i-1)*sizeof(SkScalar));
array2[i+1] = interp(array1[i], higher, array1[i+2], array2[i], array2[i+2]);
(*num)++;
}
break;
}
}
return skipMask;
}
bool SkComputeBlurredRRectParams(const SkRRect& srcRRect, const SkRRect& devRRect,
const SkRect& occluder,
SkScalar sigma, SkScalar xformedSigma,
SkRRect* rrectToDraw,
SkISize* widthHeight,
SkScalar rectXs[kSkBlurRRectMaxDivisions],
SkScalar rectYs[kSkBlurRRectMaxDivisions],
SkScalar texXs[kSkBlurRRectMaxDivisions],
SkScalar texYs[kSkBlurRRectMaxDivisions],
int* numXs, int* numYs, uint32_t* skipMask) {
unsigned int devBlurRadius = 3*SkScalarCeilToInt(xformedSigma-1/6.0f);
SkScalar srcBlurRadius = 3.0f * sigma;
const SkRect& devOrig = devRRect.getBounds();
const SkVector& devRadiiUL = devRRect.radii(SkRRect::kUpperLeft_Corner);
const SkVector& devRadiiUR = devRRect.radii(SkRRect::kUpperRight_Corner);
const SkVector& devRadiiLR = devRRect.radii(SkRRect::kLowerRight_Corner);
const SkVector& devRadiiLL = devRRect.radii(SkRRect::kLowerLeft_Corner);
const int devLeft = SkScalarCeilToInt(SkTMax<SkScalar>(devRadiiUL.fX, devRadiiLL.fX));
const int devTop = SkScalarCeilToInt(SkTMax<SkScalar>(devRadiiUL.fY, devRadiiUR.fY));
const int devRight = SkScalarCeilToInt(SkTMax<SkScalar>(devRadiiUR.fX, devRadiiLR.fX));
const int devBot = SkScalarCeilToInt(SkTMax<SkScalar>(devRadiiLL.fY, devRadiiLR.fY));
// This is a conservative check for nine-patchability
if (devOrig.fLeft + devLeft + devBlurRadius >= devOrig.fRight - devRight - devBlurRadius ||
devOrig.fTop + devTop + devBlurRadius >= devOrig.fBottom - devBot - devBlurRadius) {
return false;
}
const SkVector& srcRadiiUL = srcRRect.radii(SkRRect::kUpperLeft_Corner);
const SkVector& srcRadiiUR = srcRRect.radii(SkRRect::kUpperRight_Corner);
const SkVector& srcRadiiLR = srcRRect.radii(SkRRect::kLowerRight_Corner);
const SkVector& srcRadiiLL = srcRRect.radii(SkRRect::kLowerLeft_Corner);
const SkScalar srcLeft = SkTMax<SkScalar>(srcRadiiUL.fX, srcRadiiLL.fX);
const SkScalar srcTop = SkTMax<SkScalar>(srcRadiiUL.fY, srcRadiiUR.fY);
const SkScalar srcRight = SkTMax<SkScalar>(srcRadiiUR.fX, srcRadiiLR.fX);
const SkScalar srcBot = SkTMax<SkScalar>(srcRadiiLL.fY, srcRadiiLR.fY);
int newRRWidth = 2*devBlurRadius + devLeft + devRight + 1;
int newRRHeight = 2*devBlurRadius + devTop + devBot + 1;
widthHeight->fWidth = newRRWidth + 2 * devBlurRadius;
widthHeight->fHeight = newRRHeight + 2 * devBlurRadius;
const SkRect srcProxyRect = srcRRect.getBounds().makeOutset(srcBlurRadius, srcBlurRadius);
rectXs[0] = srcProxyRect.fLeft;
rectXs[1] = srcProxyRect.fLeft + 2*srcBlurRadius + srcLeft;
rectXs[2] = srcProxyRect.fRight - 2*srcBlurRadius - srcRight;
rectXs[3] = srcProxyRect.fRight;
rectYs[0] = srcProxyRect.fTop;
rectYs[1] = srcProxyRect.fTop + 2*srcBlurRadius + srcTop;
rectYs[2] = srcProxyRect.fBottom - 2*srcBlurRadius - srcBot;
rectYs[3] = srcProxyRect.fBottom;
texXs[0] = 0.0f;
texXs[1] = 2.0f*devBlurRadius + devLeft;
texXs[2] = 2.0f*devBlurRadius + devLeft + 1;
texXs[3] = SkIntToScalar(widthHeight->fWidth);
texYs[0] = 0.0f;
texYs[1] = 2.0f*devBlurRadius + devTop;
texYs[2] = 2.0f*devBlurRadius + devTop + 1;
texYs[3] = SkIntToScalar(widthHeight->fHeight);
SkRect temp = occluder;
*numXs = 4;
*numYs = 4;
*skipMask = 0;
if (!temp.isEmpty() && (srcProxyRect.contains(temp) || temp.intersect(srcProxyRect))) {
*skipMask = insert_into_arrays(rectXs, texXs, temp.fLeft, temp.fRight, numXs, 0x1, 1);
*skipMask = insert_into_arrays(rectYs, texYs, temp.fTop, temp.fBottom,
numYs, *skipMask, *numXs-1);
}
const SkRect newRect = SkRect::MakeXYWH(SkIntToScalar(devBlurRadius),
SkIntToScalar(devBlurRadius),
SkIntToScalar(newRRWidth),
SkIntToScalar(newRRHeight));
SkVector newRadii[4];
newRadii[0] = { SkScalarCeilToScalar(devRadiiUL.fX), SkScalarCeilToScalar(devRadiiUL.fY) };
newRadii[1] = { SkScalarCeilToScalar(devRadiiUR.fX), SkScalarCeilToScalar(devRadiiUR.fY) };
newRadii[2] = { SkScalarCeilToScalar(devRadiiLR.fX), SkScalarCeilToScalar(devRadiiLR.fY) };
newRadii[3] = { SkScalarCeilToScalar(devRadiiLL.fX), SkScalarCeilToScalar(devRadiiLL.fY) };
rrectToDraw->setRectRadii(newRect, newRadii);
return true;
}
///////////////////////////////////////////////////////////////////////////////
SkBlurMaskFilterImpl::SkBlurMaskFilterImpl(SkScalar sigma, SkBlurStyle style, bool respectCTM)
: fSigma(sigma)
, fBlurStyle(style)
, fRespectCTM(respectCTM) {
SkASSERT(fSigma > 0);
SkASSERT((unsigned)style <= kLastEnum_SkBlurStyle);
}
SkMask::Format SkBlurMaskFilterImpl::getFormat() const {
return SkMask::kA8_Format;
}
bool SkBlurMaskFilterImpl::asABlur(BlurRec* rec) const {
if (this->ignoreXform()) {
return false;
}
if (rec) {
rec->fSigma = fSigma;
rec->fStyle = fBlurStyle;
}
return true;
}
bool SkBlurMaskFilterImpl::filterMask(SkMask* dst, const SkMask& src,
const SkMatrix& matrix,
SkIPoint* margin) const {
SkScalar sigma = this->computeXformedSigma(matrix);
return SkBlurMask::BoxBlur(dst, src, sigma, fBlurStyle, margin);
}
bool SkBlurMaskFilterImpl::filterRectMask(SkMask* dst, const SkRect& r,
const SkMatrix& matrix,
SkIPoint* margin, SkMask::CreateMode createMode) const {
SkScalar sigma = computeXformedSigma(matrix);
return SkBlurMask::BlurRect(sigma, dst, r, fBlurStyle, margin, createMode);
}
bool SkBlurMaskFilterImpl::filterRRectMask(SkMask* dst, const SkRRect& r,
const SkMatrix& matrix,
SkIPoint* margin, SkMask::CreateMode createMode) const {
SkScalar sigma = computeXformedSigma(matrix);
return SkBlurMask::BlurRRect(sigma, dst, r, fBlurStyle, margin, createMode);
}
#include "SkCanvas.h"
static bool prepare_to_draw_into_mask(const SkRect& bounds, SkMask* mask) {
SkASSERT(mask != nullptr);
mask->fBounds = bounds.roundOut();
mask->fRowBytes = SkAlign4(mask->fBounds.width());
mask->fFormat = SkMask::kA8_Format;
const size_t size = mask->computeImageSize();
mask->fImage = SkMask::AllocImage(size, SkMask::kZeroInit_Alloc);
if (nullptr == mask->fImage) {
return false;
}
return true;
}
static bool draw_rrect_into_mask(const SkRRect rrect, SkMask* mask) {
if (!prepare_to_draw_into_mask(rrect.rect(), mask)) {
return false;
}
// FIXME: This code duplicates code in draw_rects_into_mask, below. Is there a
// clean way to share more code?
SkBitmap bitmap;
bitmap.installMaskPixels(*mask);
SkCanvas canvas(bitmap);
canvas.translate(-SkIntToScalar(mask->fBounds.left()),
-SkIntToScalar(mask->fBounds.top()));
SkPaint paint;
paint.setAntiAlias(true);
canvas.drawRRect(rrect, paint);
return true;
}
static bool draw_rects_into_mask(const SkRect rects[], int count, SkMask* mask) {
if (!prepare_to_draw_into_mask(rects[0], mask)) {
return false;
}
SkBitmap bitmap;
bitmap.installPixels(SkImageInfo::Make(mask->fBounds.width(),
mask->fBounds.height(),
kAlpha_8_SkColorType,
kPremul_SkAlphaType),
mask->fImage, mask->fRowBytes);
SkCanvas canvas(bitmap);
canvas.translate(-SkIntToScalar(mask->fBounds.left()),
-SkIntToScalar(mask->fBounds.top()));
SkPaint paint;
paint.setAntiAlias(true);
if (1 == count) {
canvas.drawRect(rects[0], paint);
} else {
// todo: do I need a fast way to do this?
SkPath path;
path.addRect(rects[0]);
path.addRect(rects[1]);
path.setFillType(SkPath::kEvenOdd_FillType);
canvas.drawPath(path, paint);
}
return true;
}
static bool rect_exceeds(const SkRect& r, SkScalar v) {
return r.fLeft < -v || r.fTop < -v || r.fRight > v || r.fBottom > v ||
r.width() > v || r.height() > v;
}
#include "SkMaskCache.h"
static SkCachedData* copy_mask_to_cacheddata(SkMask* mask) {
const size_t size = mask->computeTotalImageSize();
SkCachedData* data = SkResourceCache::NewCachedData(size);
if (data) {
memcpy(data->writable_data(), mask->fImage, size);
SkMask::FreeImage(mask->fImage);
mask->fImage = (uint8_t*)data->data();
}
return data;
}
static SkCachedData* find_cached_rrect(SkMask* mask, SkScalar sigma, SkBlurStyle style,
const SkRRect& rrect) {
return SkMaskCache::FindAndRef(sigma, style, rrect, mask);
}
static SkCachedData* add_cached_rrect(SkMask* mask, SkScalar sigma, SkBlurStyle style,
const SkRRect& rrect) {
SkCachedData* cache = copy_mask_to_cacheddata(mask);
if (cache) {
SkMaskCache::Add(sigma, style, rrect, *mask, cache);
}
return cache;
}
static SkCachedData* find_cached_rects(SkMask* mask, SkScalar sigma, SkBlurStyle style,
const SkRect rects[], int count) {
return SkMaskCache::FindAndRef(sigma, style, rects, count, mask);
}
static SkCachedData* add_cached_rects(SkMask* mask, SkScalar sigma, SkBlurStyle style,
const SkRect rects[], int count) {
SkCachedData* cache = copy_mask_to_cacheddata(mask);
if (cache) {
SkMaskCache::Add(sigma, style, rects, count, *mask, cache);
}
return cache;
}
static const bool c_analyticBlurRRect{true};
SkMaskFilterBase::FilterReturn
SkBlurMaskFilterImpl::filterRRectToNine(const SkRRect& rrect, const SkMatrix& matrix,
const SkIRect& clipBounds,
NinePatch* patch) const {
SkASSERT(patch != nullptr);
switch (rrect.getType()) {
case SkRRect::kEmpty_Type:
// Nothing to draw.
return kFalse_FilterReturn;
case SkRRect::kRect_Type:
// We should have caught this earlier.
SkASSERT(false);
// Fall through.
case SkRRect::kOval_Type:
// The nine patch special case does not handle ovals, and we
// already have code for rectangles.
return kUnimplemented_FilterReturn;
// These three can take advantage of this fast path.
case SkRRect::kSimple_Type:
case SkRRect::kNinePatch_Type:
case SkRRect::kComplex_Type:
break;
}
// TODO: report correct metrics for innerstyle, where we do not grow the
// total bounds, but we do need an inset the size of our blur-radius
if (kInner_SkBlurStyle == fBlurStyle) {
return kUnimplemented_FilterReturn;
}
// TODO: take clipBounds into account to limit our coordinates up front
// for now, just skip too-large src rects (to take the old code path).
if (rect_exceeds(rrect.rect(), SkIntToScalar(32767))) {
return kUnimplemented_FilterReturn;
}
SkIPoint margin;
SkMask srcM, dstM;
srcM.fBounds = rrect.rect().roundOut();
srcM.fFormat = SkMask::kA8_Format;
srcM.fRowBytes = 0;
bool filterResult = false;
if (c_analyticBlurRRect) {
// special case for fast round rect blur
// don't actually do the blur the first time, just compute the correct size
filterResult = this->filterRRectMask(&dstM, rrect, matrix, &margin,
SkMask::kJustComputeBounds_CreateMode);
}
if (!filterResult) {
filterResult = this->filterMask(&dstM, srcM, matrix, &margin);
}
if (!filterResult) {
return kFalse_FilterReturn;
}
// Now figure out the appropriate width and height of the smaller round rectangle
// to stretch. It will take into account the larger radius per side as well as double
// the margin, to account for inner and outer blur.
const SkVector& UL = rrect.radii(SkRRect::kUpperLeft_Corner);
const SkVector& UR = rrect.radii(SkRRect::kUpperRight_Corner);
const SkVector& LR = rrect.radii(SkRRect::kLowerRight_Corner);
const SkVector& LL = rrect.radii(SkRRect::kLowerLeft_Corner);
const SkScalar leftUnstretched = SkTMax(UL.fX, LL.fX) + SkIntToScalar(2 * margin.fX);
const SkScalar rightUnstretched = SkTMax(UR.fX, LR.fX) + SkIntToScalar(2 * margin.fX);
// Extra space in the middle to ensure an unchanging piece for stretching. Use 3 to cover
// any fractional space on either side plus 1 for the part to stretch.
const SkScalar stretchSize = SkIntToScalar(3);
const SkScalar totalSmallWidth = leftUnstretched + rightUnstretched + stretchSize;
if (totalSmallWidth >= rrect.rect().width()) {
// There is no valid piece to stretch.
return kUnimplemented_FilterReturn;
}
const SkScalar topUnstretched = SkTMax(UL.fY, UR.fY) + SkIntToScalar(2 * margin.fY);
const SkScalar bottomUnstretched = SkTMax(LL.fY, LR.fY) + SkIntToScalar(2 * margin.fY);
const SkScalar totalSmallHeight = topUnstretched + bottomUnstretched + stretchSize;
if (totalSmallHeight >= rrect.rect().height()) {
// There is no valid piece to stretch.
return kUnimplemented_FilterReturn;
}
SkRect smallR = SkRect::MakeWH(totalSmallWidth, totalSmallHeight);
SkRRect smallRR;
SkVector radii[4];
radii[SkRRect::kUpperLeft_Corner] = UL;
radii[SkRRect::kUpperRight_Corner] = UR;
radii[SkRRect::kLowerRight_Corner] = LR;
radii[SkRRect::kLowerLeft_Corner] = LL;
smallRR.setRectRadii(smallR, radii);
const SkScalar sigma = this->computeXformedSigma(matrix);
SkCachedData* cache = find_cached_rrect(&patch->fMask, sigma, fBlurStyle, smallRR);
if (!cache) {
bool analyticBlurWorked = false;
if (c_analyticBlurRRect) {
analyticBlurWorked =
this->filterRRectMask(&patch->fMask, smallRR, matrix, &margin,
SkMask::kComputeBoundsAndRenderImage_CreateMode);
}
if (!analyticBlurWorked) {
if (!draw_rrect_into_mask(smallRR, &srcM)) {
return kFalse_FilterReturn;
}
SkAutoMaskFreeImage amf(srcM.fImage);
if (!this->filterMask(&patch->fMask, srcM, matrix, &margin)) {
return kFalse_FilterReturn;
}
}
cache = add_cached_rrect(&patch->fMask, sigma, fBlurStyle, smallRR);
}
patch->fMask.fBounds.offsetTo(0, 0);
patch->fOuterRect = dstM.fBounds;
patch->fCenter.fX = SkScalarCeilToInt(leftUnstretched) + 1;
patch->fCenter.fY = SkScalarCeilToInt(topUnstretched) + 1;
SkASSERT(nullptr == patch->fCache);
patch->fCache = cache; // transfer ownership to patch
return kTrue_FilterReturn;
}
// Use the faster analytic blur approach for ninepatch rects
static const bool c_analyticBlurNinepatch{true};
SkMaskFilterBase::FilterReturn
SkBlurMaskFilterImpl::filterRectsToNine(const SkRect rects[], int count,
const SkMatrix& matrix,
const SkIRect& clipBounds,
NinePatch* patch) const {
if (count < 1 || count > 2) {
return kUnimplemented_FilterReturn;
}
// TODO: report correct metrics for innerstyle, where we do not grow the
// total bounds, but we do need an inset the size of our blur-radius
if (kInner_SkBlurStyle == fBlurStyle || kOuter_SkBlurStyle == fBlurStyle) {
return kUnimplemented_FilterReturn;
}
// TODO: take clipBounds into account to limit our coordinates up front
// for now, just skip too-large src rects (to take the old code path).
if (rect_exceeds(rects[0], SkIntToScalar(32767))) {
return kUnimplemented_FilterReturn;
}
SkIPoint margin;
SkMask srcM, dstM;
srcM.fBounds = rects[0].roundOut();
srcM.fFormat = SkMask::kA8_Format;
srcM.fRowBytes = 0;
bool filterResult = false;
if (count == 1 && c_analyticBlurNinepatch) {
// special case for fast rect blur
// don't actually do the blur the first time, just compute the correct size
filterResult = this->filterRectMask(&dstM, rects[0], matrix, &margin,
SkMask::kJustComputeBounds_CreateMode);
} else {
filterResult = this->filterMask(&dstM, srcM, matrix, &margin);
}
if (!filterResult) {
return kFalse_FilterReturn;
}
/*
* smallR is the smallest version of 'rect' that will still guarantee that
* we get the same blur results on all edges, plus 1 center row/col that is
* representative of the extendible/stretchable edges of the ninepatch.
* Since our actual edge may be fractional we inset 1 more to be sure we
* don't miss any interior blur.
* x is an added pixel of blur, and { and } are the (fractional) edge
* pixels from the original rect.
*
* x x { x x .... x x } x x
*
* Thus, in this case, we inset by a total of 5 (on each side) beginning
* with our outer-rect (dstM.fBounds)
*/
SkRect smallR[2];
SkIPoint center;
// +2 is from +1 for each edge (to account for possible fractional edges
int smallW = dstM.fBounds.width() - srcM.fBounds.width() + 2;
int smallH = dstM.fBounds.height() - srcM.fBounds.height() + 2;
SkIRect innerIR;
if (1 == count) {
innerIR = srcM.fBounds;
center.set(smallW, smallH);
} else {
SkASSERT(2 == count);
rects[1].roundIn(&innerIR);
center.set(smallW + (innerIR.left() - srcM.fBounds.left()),
smallH + (innerIR.top() - srcM.fBounds.top()));
}
// +1 so we get a clean, stretchable, center row/col
smallW += 1;
smallH += 1;
// we want the inset amounts to be integral, so we don't change any
// fractional phase on the fRight or fBottom of our smallR.
const SkScalar dx = SkIntToScalar(innerIR.width() - smallW);
const SkScalar dy = SkIntToScalar(innerIR.height() - smallH);
if (dx < 0 || dy < 0) {
// we're too small, relative to our blur, to break into nine-patch,
// so we ask to have our normal filterMask() be called.
return kUnimplemented_FilterReturn;
}
smallR[0].set(rects[0].left(), rects[0].top(), rects[0].right() - dx, rects[0].bottom() - dy);
if (smallR[0].width() < 2 || smallR[0].height() < 2) {
return kUnimplemented_FilterReturn;
}
if (2 == count) {
smallR[1].set(rects[1].left(), rects[1].top(),
rects[1].right() - dx, rects[1].bottom() - dy);
SkASSERT(!smallR[1].isEmpty());
}
const SkScalar sigma = this->computeXformedSigma(matrix);
SkCachedData* cache = find_cached_rects(&patch->fMask, sigma, fBlurStyle, smallR, count);
if (!cache) {
if (count > 1 || !c_analyticBlurNinepatch) {
if (!draw_rects_into_mask(smallR, count, &srcM)) {
return kFalse_FilterReturn;
}
SkAutoMaskFreeImage amf(srcM.fImage);
if (!this->filterMask(&patch->fMask, srcM, matrix, &margin)) {
return kFalse_FilterReturn;
}
} else {
if (!this->filterRectMask(&patch->fMask, smallR[0], matrix, &margin,
SkMask::kComputeBoundsAndRenderImage_CreateMode)) {
return kFalse_FilterReturn;
}
}
cache = add_cached_rects(&patch->fMask, sigma, fBlurStyle, smallR, count);
}
patch->fMask.fBounds.offsetTo(0, 0);
patch->fOuterRect = dstM.fBounds;
patch->fCenter = center;
SkASSERT(nullptr == patch->fCache);
patch->fCache = cache; // transfer ownership to patch
return kTrue_FilterReturn;
}
void SkBlurMaskFilterImpl::computeFastBounds(const SkRect& src,
SkRect* dst) const {
SkScalar pad = 3.0f * fSigma;
dst->set(src.fLeft - pad, src.fTop - pad,
src.fRight + pad, src.fBottom + pad);
}
sk_sp<SkFlattenable> SkBlurMaskFilterImpl::CreateProc(SkReadBuffer& buffer) {
const SkScalar sigma = buffer.readScalar();
SkBlurStyle style = buffer.read32LE(kLastEnum_SkBlurStyle);
uint32_t flags = buffer.read32LE(0x3); // historically we only recorded 2 bits
bool respectCTM = !(flags & 1); // historically we stored ignoreCTM in low bit
if (buffer.isVersionLT(SkReadBuffer::kRemoveOccluderFromBlurMaskFilter)) {
SkRect unused;
buffer.readRect(&unused);
}
return SkMaskFilter::MakeBlur((SkBlurStyle)style, sigma, respectCTM);
}
void SkBlurMaskFilterImpl::flatten(SkWriteBuffer& buffer) const {
buffer.writeScalar(fSigma);
buffer.writeUInt(fBlurStyle);
buffer.writeUInt(!fRespectCTM); // historically we recorded ignoreCTM
}
#if SK_SUPPORT_GPU
bool SkBlurMaskFilterImpl::directFilterMaskGPU(GrRecordingContext* context,
GrRenderTargetContext* renderTargetContext,
GrPaint&& paint,
const GrClip& clip,
const SkMatrix& viewMatrix,
const GrShape& shape) const {
SkASSERT(renderTargetContext);
if (fBlurStyle != kNormal_SkBlurStyle) {
return false;
}
if (!viewMatrix.isScaleTranslate()) {
return false;
}
// TODO: we could handle blurred stroked circles
if (!shape.style().isSimpleFill()) {
return false;
}
SkScalar xformedSigma = this->computeXformedSigma(viewMatrix);
if (xformedSigma <= 0) {
return false;
}
SkRRect srcRRect;
bool inverted;
if (!shape.asRRect(&srcRRect, nullptr, nullptr, &inverted) || inverted) {
return false;
}
SkRRect devRRect;
if (!srcRRect.transform(viewMatrix, &devRRect)) {
return false;
}
if (!SkRRectPriv::AllCornersCircular(devRRect)) {
return false;
}
GrProxyProvider* proxyProvider = context->priv().proxyProvider();
std::unique_ptr<GrFragmentProcessor> fp;
if (devRRect.isRect() || SkRRectPriv::IsCircle(devRRect)) {
if (devRRect.isRect()) {
SkScalar pad = 3.0f * xformedSigma;
const SkRect dstCoverageRect = devRRect.rect().makeOutset(pad, pad);
fp = GrRectBlurEffect::Make(proxyProvider, *context->priv().caps()->shaderCaps(),
dstCoverageRect, xformedSigma);
} else {
fp = GrCircleBlurFragmentProcessor::Make(proxyProvider, devRRect.rect(), xformedSigma);
}
if (!fp) {
return false;
}
paint.addCoverageFragmentProcessor(std::move(fp));
SkRect srcProxyRect = srcRRect.rect();
SkScalar outsetX = 3.0f*fSigma;
SkScalar outsetY = 3.0f*fSigma;
if (this->ignoreXform()) {
// When we're ignoring the CTM the padding added to the source rect also needs to ignore
// the CTM. The matrix passed in here is guaranteed to be just scale and translate so we
// can just grab the X and Y scales off the matrix and pre-undo the scale.
outsetX /= SkScalarAbs(viewMatrix.getScaleX());
outsetY /= SkScalarAbs(viewMatrix.getScaleY());
}
srcProxyRect.outset(outsetX, outsetY);
renderTargetContext->drawRect(clip, std::move(paint), GrAA::kNo, viewMatrix, srcProxyRect);
return true;
}
fp = GrRRectBlurEffect::Make(context, fSigma, xformedSigma, srcRRect, devRRect);
if (!fp) {
return false;
}
if (!this->ignoreXform()) {
SkRect srcProxyRect = srcRRect.rect();
srcProxyRect.outset(3.0f*fSigma, 3.0f*fSigma);
SkVertices::Builder builder(SkVertices::kTriangles_VertexMode, 4, 6, 0);
srcProxyRect.toQuad(builder.positions());
static const uint16_t fullIndices[6] = { 0, 1, 2, 0, 2, 3 };
memcpy(builder.indices(), fullIndices, sizeof(fullIndices));
sk_sp<SkVertices> vertices = builder.detach();
paint.addCoverageFragmentProcessor(std::move(fp));
renderTargetContext->drawVertices(clip, std::move(paint), viewMatrix, std::move(vertices),
nullptr, 0);
} else {
SkMatrix inverse;
if (!viewMatrix.invert(&inverse)) {
return false;
}
float extra=3.f*SkScalarCeilToScalar(xformedSigma-1/6.0f);
SkRect proxyRect = devRRect.rect();
proxyRect.outset(extra, extra);
paint.addCoverageFragmentProcessor(std::move(fp));
renderTargetContext->fillRectWithLocalMatrix(clip, std::move(paint), GrAA::kNo,
SkMatrix::I(), proxyRect, inverse);
}
return true;
}
bool SkBlurMaskFilterImpl::canFilterMaskGPU(const GrShape& shape,
const SkIRect& devSpaceShapeBounds,
const SkIRect& clipBounds,
const SkMatrix& ctm,
SkIRect* maskRect) const {
SkScalar xformedSigma = this->computeXformedSigma(ctm);
if (xformedSigma <= 0) {
maskRect->setEmpty();
return false;
}
if (maskRect) {
float sigma3 = 3 * SkScalarToFloat(xformedSigma);
// Outset srcRect and clipRect by 3 * sigma, to compute affected blur area.
SkIRect clipRect = clipBounds.makeOutset(sigma3, sigma3);
SkIRect srcRect = devSpaceShapeBounds.makeOutset(sigma3, sigma3);
if (!srcRect.intersect(clipRect)) {
srcRect.setEmpty();
}
*maskRect = srcRect;
}
// We prefer to blur paths with small blur radii on the CPU.
if (ctm.rectStaysRect()) {
static const SkScalar kMIN_GPU_BLUR_SIZE = SkIntToScalar(64);
static const SkScalar kMIN_GPU_BLUR_SIGMA = SkIntToScalar(32);
if (devSpaceShapeBounds.width() <= kMIN_GPU_BLUR_SIZE &&
devSpaceShapeBounds.height() <= kMIN_GPU_BLUR_SIZE &&
xformedSigma <= kMIN_GPU_BLUR_SIGMA) {
return false;
}
}
return true;
}
sk_sp<GrTextureProxy> SkBlurMaskFilterImpl::filterMaskGPU(GrRecordingContext* context,
sk_sp<GrTextureProxy> srcProxy,
const SkMatrix& ctm,
const SkIRect& maskRect) const {
// 'maskRect' isn't snapped to the UL corner but the mask in 'src' is.
const SkIRect clipRect = SkIRect::MakeWH(maskRect.width(), maskRect.height());
SkScalar xformedSigma = this->computeXformedSigma(ctm);
SkASSERT(xformedSigma > 0);
// If we're doing a normal blur, we can clobber the pathTexture in the
// gaussianBlur. Otherwise, we need to save it for later compositing.
bool isNormalBlur = (kNormal_SkBlurStyle == fBlurStyle);
sk_sp<GrRenderTargetContext> renderTargetContext(
SkGpuBlurUtils::GaussianBlur(context,
srcProxy,
nullptr,
clipRect,
SkIRect::EmptyIRect(),
xformedSigma,
xformedSigma,
GrTextureDomain::kIgnore_Mode,
kPremul_SkAlphaType));
if (!renderTargetContext) {
return nullptr;
}
if (!isNormalBlur) {
GrPaint paint;
// Blend pathTexture over blurTexture.
paint.addCoverageFragmentProcessor(GrSimpleTextureEffect::Make(std::move(srcProxy),
SkMatrix::I()));
if (kInner_SkBlurStyle == fBlurStyle) {
// inner: dst = dst * src
paint.setCoverageSetOpXPFactory(SkRegion::kIntersect_Op);
} else if (kSolid_SkBlurStyle == fBlurStyle) {
// solid: dst = src + dst - src * dst
// = src + (1 - src) * dst
paint.setCoverageSetOpXPFactory(SkRegion::kUnion_Op);
} else if (kOuter_SkBlurStyle == fBlurStyle) {
// outer: dst = dst * (1 - src)
// = 0 * src + (1 - src) * dst
paint.setCoverageSetOpXPFactory(SkRegion::kDifference_Op);
} else {
paint.setCoverageSetOpXPFactory(SkRegion::kReplace_Op);
}
renderTargetContext->drawRect(GrNoClip(), std::move(paint), GrAA::kNo, SkMatrix::I(),
SkRect::Make(clipRect));
}
return renderTargetContext->asTextureProxyRef();
}
#endif // SK_SUPPORT_GPU
void sk_register_blur_maskfilter_createproc() { SK_REGISTER_FLATTENABLE(SkBlurMaskFilterImpl); }
sk_sp<SkMaskFilter> SkMaskFilter::MakeBlur(SkBlurStyle style, SkScalar sigma, bool respectCTM) {
if (SkScalarIsFinite(sigma) && sigma > 0) {
return sk_sp<SkMaskFilter>(new SkBlurMaskFilterImpl(sigma, style, respectCTM));
}
return nullptr;
}