blob: 2c68156cf64d05aa5bf7d03201e305125fe65c43 [file] [log] [blame]
/*
* Copyright 2015 Google Inc.
*
* Use of this source code is governed by a BSD-style license that can be
* found in the LICENSE file.
*/
#include "dm/DMSrcSink.h"
#include "gm/verifiers/gmverifier.h"
#include "include/codec/SkAndroidCodec.h"
#include "include/codec/SkCodec.h"
#include "include/core/SkColorSpace.h"
#include "include/core/SkData.h"
#include "include/core/SkDeferredDisplayListRecorder.h"
#include "include/core/SkDocument.h"
#include "include/core/SkExecutor.h"
#include "include/core/SkImageGenerator.h"
#include "include/core/SkMallocPixelRef.h"
#include "include/core/SkPictureRecorder.h"
#include "include/core/SkSerialProcs.h"
#include "include/core/SkStream.h"
#include "include/core/SkSurface.h"
#include "include/core/SkSurfaceCharacterization.h"
#include "include/docs/SkPDFDocument.h"
#include "include/gpu/GrBackendSurface.h"
#include "include/gpu/GrDirectContext.h"
#include "include/ports/SkImageGeneratorCG.h"
#include "include/ports/SkImageGeneratorNDK.h"
#include "include/ports/SkImageGeneratorWIC.h"
#include "include/private/base/SkTLogic.h"
#include "include/utils/SkNullCanvas.h"
#include "include/utils/SkPaintFilterCanvas.h"
#include "modules/skcms/skcms.h"
#include "modules/skottie/utils/SkottieUtils.h"
#include "src/base/SkAutoMalloc.h"
#include "src/base/SkRandom.h"
#include "src/base/SkTLazy.h"
#include "src/codec/SkCodecImageGenerator.h"
#include "src/codec/SkSwizzler.h"
#include "src/core/SkAutoPixmapStorage.h"
#include "src/core/SkImageInfoPriv.h"
#include "src/core/SkOSFile.h"
#include "src/core/SkOpts.h"
#include "src/core/SkPictureData.h"
#include "src/core/SkPicturePriv.h"
#include "src/core/SkRecordDraw.h"
#include "src/core/SkRecorder.h"
#include "src/core/SkTaskGroup.h"
#include "src/gpu/ganesh/GrDirectContextPriv.h"
#include "src/gpu/ganesh/GrGpu.h"
#include "src/utils/SkJSONWriter.h"
#include "src/utils/SkMultiPictureDocumentPriv.h"
#include "src/utils/SkOSPath.h"
#include "src/utils/SkTestCanvas.h"
#include "tools/DDLPromiseImageHelper.h"
#include "tools/DDLTileHelper.h"
#include "tools/Resources.h"
#include "tools/RuntimeBlendUtils.h"
#include "tools/ToolUtils.h"
#include "tools/UrlDataManager.h"
#include "tools/debugger/DebugCanvas.h"
#include "tools/gpu/BackendSurfaceFactory.h"
#include "tools/gpu/MemoryCache.h"
#if defined(SK_BUILD_FOR_WIN)
#include "include/docs/SkXPSDocument.h"
#include "src/utils/win/SkAutoCoInitialize.h"
#include "src/utils/win/SkHRESULT.h"
#include "src/utils/win/SkTScopedComPtr.h"
#include <XpsObjectModel.h>
#endif
#if defined(SK_ENABLE_SKOTTIE)
#include "modules/skottie/include/Skottie.h"
#include "modules/skresources/include/SkResources.h"
#endif
#if defined(SK_ENABLE_SVG)
#include "include/svg/SkSVGCanvas.h"
#include "modules/svg/include/SkSVGDOM.h"
#include "modules/svg/include/SkSVGNode.h"
#include "src/xml/SkXMLWriter.h"
#endif
#if defined(SK_GRAPHITE)
#include "include/gpu/graphite/Context.h"
#include "include/gpu/graphite/Recorder.h"
#include "include/gpu/graphite/Recording.h"
// TODO: Remove this src include once we figure out public readPixels call for Graphite.
#include "src/gpu/graphite/Surface_Graphite.h"
#include "tools/graphite/ContextFactory.h"
#include "tools/graphite/GraphiteTestContext.h"
#endif
#if defined(SK_ENABLE_ANDROID_UTILS)
#include "client_utils/android/BitmapRegionDecoder.h"
#endif
#include "tests/TestUtils.h"
#include <cmath>
#include <functional>
using namespace skia_private;
static DEFINE_bool(multiPage, false,
"For document-type backends, render the source into multiple pages");
static DEFINE_bool(RAW_threading, true, "Allow RAW decodes to run on multiple threads?");
DECLARE_int(gpuThreads);
using sk_gpu_test::GrContextFactory;
using sk_gpu_test::ContextInfo;
namespace DM {
GMSrc::GMSrc(skiagm::GMFactory factory) : fFactory(factory) {}
Result GMSrc::draw(SkCanvas* canvas) const {
std::unique_ptr<skiagm::GM> gm(fFactory());
SkString msg;
skiagm::DrawResult gpuSetupResult = gm->gpuSetup(canvas, &msg);
switch (gpuSetupResult) {
case skiagm::DrawResult::kOk : break;
case skiagm::DrawResult::kFail: return Result(Result::Status::Fatal, msg);
case skiagm::DrawResult::kSkip: return Result(Result::Status::Skip, msg);
default: SK_ABORT("");
}
skiagm::DrawResult drawResult = gm->draw(canvas, &msg);
switch (drawResult) {
case skiagm::DrawResult::kOk : return Result(Result::Status::Ok, msg);
case skiagm::DrawResult::kFail: return Result(Result::Status::Fatal, msg);
case skiagm::DrawResult::kSkip: return Result(Result::Status::Skip, msg);
default: SK_ABORT("");
}
// Note: we don't call "gpuTeardown" here because, when testing DDL recording, we want
// the gpu-backed images to live past the lifetime of the GM.
}
SkISize GMSrc::size() const {
std::unique_ptr<skiagm::GM> gm(fFactory());
return gm->getISize();
}
Name GMSrc::name() const {
std::unique_ptr<skiagm::GM> gm(fFactory());
return gm->getName();
}
void GMSrc::modifyGrContextOptions(GrContextOptions* options) const {
std::unique_ptr<skiagm::GM> gm(fFactory());
gm->modifyGrContextOptions(options);
}
std::unique_ptr<skiagm::verifiers::VerifierList> GMSrc::getVerifiers() const {
std::unique_ptr<skiagm::GM> gm(fFactory());
return gm->getVerifiers();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static SkString get_scaled_name(const Path& path, float scale) {
return SkStringPrintf("%s_%.3f", SkOSPath::Basename(path.c_str()).c_str(), scale);
}
#ifdef SK_ENABLE_ANDROID_UTILS
BRDSrc::BRDSrc(Path path, Mode mode, CodecSrc::DstColorType dstColorType, uint32_t sampleSize)
: fPath(path)
, fMode(mode)
, fDstColorType(dstColorType)
, fSampleSize(sampleSize)
{}
bool BRDSrc::veto(SinkFlags flags) const {
// No need to test to non-raster or indirect backends.
return flags.type != SinkFlags::kRaster
|| flags.approach != SinkFlags::kDirect;
}
static std::unique_ptr<android::skia::BitmapRegionDecoder> create_brd(Path path) {
sk_sp<SkData> encoded(SkData::MakeFromFileName(path.c_str()));
return android::skia::BitmapRegionDecoder::Make(encoded);
}
static inline void alpha8_to_gray8(SkBitmap* bitmap) {
// Android requires kGray8 bitmaps to be tagged as kAlpha8. Here we convert
// them back to kGray8 so our test framework can draw them correctly.
if (kAlpha_8_SkColorType == bitmap->info().colorType()) {
SkImageInfo newInfo = bitmap->info().makeColorType(kGray_8_SkColorType)
.makeAlphaType(kOpaque_SkAlphaType);
*const_cast<SkImageInfo*>(&bitmap->info()) = newInfo;
}
}
Result BRDSrc::draw(SkCanvas* canvas) const {
SkColorType colorType = canvas->imageInfo().colorType();
if (kRGB_565_SkColorType == colorType &&
CodecSrc::kGetFromCanvas_DstColorType != fDstColorType)
{
return Result::Skip("Testing non-565 to 565 is uninteresting.");
}
switch (fDstColorType) {
case CodecSrc::kGetFromCanvas_DstColorType:
break;
case CodecSrc::kGrayscale_Always_DstColorType:
colorType = kGray_8_SkColorType;
break;
default:
SkASSERT(false);
break;
}
auto brd = create_brd(fPath);
if (nullptr == brd) {
return Result::Skip("Could not create brd for %s.", fPath.c_str());
}
auto recommendedCT = brd->computeOutputColorType(colorType);
if (kRGB_565_SkColorType == colorType && recommendedCT != colorType) {
return Result::Skip("Skip decoding non-opaque to 565.");
}
colorType = recommendedCT;
auto colorSpace = brd->computeOutputColorSpace(colorType, nullptr);
const uint32_t width = brd->width();
const uint32_t height = brd->height();
// Visually inspecting very small output images is not necessary.
if ((width / fSampleSize <= 10 || height / fSampleSize <= 10) && 1 != fSampleSize) {
return Result::Skip("Scaling very small images is uninteresting.");
}
switch (fMode) {
case kFullImage_Mode: {
SkBitmap bitmap;
if (!brd->decodeRegion(&bitmap, nullptr, SkIRect::MakeXYWH(0, 0, width, height),
fSampleSize, colorType, false, colorSpace)) {
return Result::Fatal("Cannot decode (full) region.");
}
alpha8_to_gray8(&bitmap);
canvas->drawImage(bitmap.asImage(), 0, 0);
return Result::Ok();
}
case kDivisor_Mode: {
const uint32_t divisor = 2;
if (width < divisor || height < divisor) {
return Result::Skip("Divisor is larger than image dimension.");
}
// Use a border to test subsets that extend outside the image.
// We will not allow the border to be larger than the image dimensions. Allowing
// these large borders causes off by one errors that indicate a problem with the
// test suite, not a problem with the implementation.
const uint32_t maxBorder = std::min(width, height) / (fSampleSize * divisor);
const uint32_t scaledBorder = std::min(5u, maxBorder);
const uint32_t unscaledBorder = scaledBorder * fSampleSize;
// We may need to clear the canvas to avoid uninitialized memory.
// Assume we are scaling a 780x780 image with sampleSize = 8.
// The output image should be 97x97.
// Each subset will be 390x390.
// Each scaled subset be 48x48.
// Four scaled subsets will only fill a 96x96 image.
// The bottom row and last column will not be touched.
// This is an unfortunate result of our rounding rules when scaling.
// Maybe we need to consider testing scaled subsets without trying to
// combine them to match the full scaled image? Or maybe this is the
// best we can do?
canvas->clear(0);
for (uint32_t x = 0; x < divisor; x++) {
for (uint32_t y = 0; y < divisor; y++) {
// Calculate the subset dimensions
uint32_t subsetWidth = width / divisor;
uint32_t subsetHeight = height / divisor;
const int left = x * subsetWidth;
const int top = y * subsetHeight;
// Increase the size of the last subset in each row or column, when the
// divisor does not divide evenly into the image dimensions
subsetWidth += (x + 1 == divisor) ? (width % divisor) : 0;
subsetHeight += (y + 1 == divisor) ? (height % divisor) : 0;
// Increase the size of the subset in order to have a border on each side
const int decodeLeft = left - unscaledBorder;
const int decodeTop = top - unscaledBorder;
const uint32_t decodeWidth = subsetWidth + unscaledBorder * 2;
const uint32_t decodeHeight = subsetHeight + unscaledBorder * 2;
SkBitmap bitmap;
if (!brd->decodeRegion(&bitmap, nullptr, SkIRect::MakeXYWH(decodeLeft,
decodeTop, decodeWidth, decodeHeight), fSampleSize, colorType, false,
colorSpace)) {
return Result::Fatal("Cannot decode region.");
}
alpha8_to_gray8(&bitmap);
canvas->drawImageRect(bitmap.asImage().get(),
SkRect::MakeXYWH((SkScalar) scaledBorder, (SkScalar) scaledBorder,
(SkScalar) (subsetWidth / fSampleSize),
(SkScalar) (subsetHeight / fSampleSize)),
SkRect::MakeXYWH((SkScalar) (left / fSampleSize),
(SkScalar) (top / fSampleSize),
(SkScalar) (subsetWidth / fSampleSize),
(SkScalar) (subsetHeight / fSampleSize)),
SkSamplingOptions(), nullptr,
SkCanvas::kStrict_SrcRectConstraint);
}
}
return Result::Ok();
}
default:
SkASSERT(false);
return Result::Fatal("Error: Should not be reached.");
}
}
SkISize BRDSrc::size() const {
auto brd = create_brd(fPath);
if (brd) {
return {std::max(1, brd->width() / (int)fSampleSize),
std::max(1, brd->height() / (int)fSampleSize)};
}
return {0, 0};
}
Name BRDSrc::name() const {
// We will replicate the names used by CodecSrc so that images can
// be compared in Gold.
if (1 == fSampleSize) {
return SkOSPath::Basename(fPath.c_str());
}
return get_scaled_name(fPath, 1.0f / (float) fSampleSize);
}
#endif // SK_ENABLE_ANDROID_UTILS
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static bool serial_from_path_name(const SkString& path) {
if (!FLAGS_RAW_threading) {
static const char* const exts[] = {
"arw", "cr2", "dng", "nef", "nrw", "orf", "raf", "rw2", "pef", "srw",
"ARW", "CR2", "DNG", "NEF", "NRW", "ORF", "RAF", "RW2", "PEF", "SRW",
};
const char* actualExt = strrchr(path.c_str(), '.');
if (actualExt) {
actualExt++;
for (auto* ext : exts) {
if (0 == strcmp(ext, actualExt)) {
return true;
}
}
}
}
return false;
}
CodecSrc::CodecSrc(Path path, Mode mode, DstColorType dstColorType, SkAlphaType dstAlphaType,
float scale)
: fPath(path)
, fMode(mode)
, fDstColorType(dstColorType)
, fDstAlphaType(dstAlphaType)
, fScale(scale)
, fRunSerially(serial_from_path_name(path))
{}
bool CodecSrc::veto(SinkFlags flags) const {
// Test to direct raster backends (8888 and 565).
return flags.type != SinkFlags::kRaster || flags.approach != SinkFlags::kDirect;
}
// Allows us to test decodes to non-native 8888.
static void swap_rb_if_necessary(SkBitmap& bitmap, CodecSrc::DstColorType dstColorType) {
if (CodecSrc::kNonNative8888_Always_DstColorType != dstColorType) {
return;
}
for (int y = 0; y < bitmap.height(); y++) {
uint32_t* row = (uint32_t*) bitmap.getAddr(0, y);
SkOpts::RGBA_to_BGRA(row, row, bitmap.width());
}
}
static bool get_decode_info(SkImageInfo* decodeInfo, SkColorType canvasColorType,
CodecSrc::DstColorType dstColorType, SkAlphaType dstAlphaType) {
switch (dstColorType) {
case CodecSrc::kGrayscale_Always_DstColorType:
if (kRGB_565_SkColorType == canvasColorType) {
return false;
}
*decodeInfo = decodeInfo->makeColorType(kGray_8_SkColorType);
break;
case CodecSrc::kNonNative8888_Always_DstColorType:
if (kRGB_565_SkColorType == canvasColorType
|| kRGBA_F16_SkColorType == canvasColorType) {
return false;
}
#ifdef SK_PMCOLOR_IS_RGBA
*decodeInfo = decodeInfo->makeColorType(kBGRA_8888_SkColorType);
#else
*decodeInfo = decodeInfo->makeColorType(kRGBA_8888_SkColorType);
#endif
break;
default:
if (kRGB_565_SkColorType == canvasColorType &&
kOpaque_SkAlphaType != decodeInfo->alphaType()) {
return false;
}
*decodeInfo = decodeInfo->makeColorType(canvasColorType);
break;
}
*decodeInfo = decodeInfo->makeAlphaType(dstAlphaType);
return true;
}
static void draw_to_canvas(SkCanvas* canvas, const SkImageInfo& info, void* pixels, size_t rowBytes,
CodecSrc::DstColorType dstColorType,
SkScalar left = 0, SkScalar top = 0) {
SkBitmap bitmap;
bitmap.installPixels(info, pixels, rowBytes);
swap_rb_if_necessary(bitmap, dstColorType);
canvas->drawImage(bitmap.asImage(), left, top);
}
// For codec srcs, we want the "draw" step to be a memcpy. Any interesting color space or
// color format conversions should be performed by the codec. Sometimes the output of the
// decode will be in an interesting color space. On our srgb and f16 backends, we need to
// "pretend" that the color space is standard sRGB to avoid triggering color conversion
// at draw time.
static void set_bitmap_color_space(SkImageInfo* info) {
*info = info->makeColorSpace(SkColorSpace::MakeSRGB());
}
Result CodecSrc::draw(SkCanvas* canvas) const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
if (!encoded) {
return Result::Fatal("Couldn't read %s.", fPath.c_str());
}
std::unique_ptr<SkCodec> codec(SkCodec::MakeFromData(encoded));
if (nullptr == codec) {
return Result::Fatal("Couldn't create codec for %s.", fPath.c_str());
}
SkImageInfo decodeInfo = codec->getInfo();
if (!get_decode_info(&decodeInfo, canvas->imageInfo().colorType(), fDstColorType,
fDstAlphaType)) {
return Result::Skip("Skipping uninteresting test.");
}
// Try to scale the image if it is desired
SkISize size = codec->getScaledDimensions(fScale);
std::unique_ptr<SkAndroidCodec> androidCodec;
if (1.0f != fScale && fMode == kAnimated_Mode) {
androidCodec = SkAndroidCodec::MakeFromData(encoded);
size = androidCodec->getSampledDimensions(1 / fScale);
}
if (size == decodeInfo.dimensions() && 1.0f != fScale) {
return Result::Skip("Test without scaling is uninteresting.");
}
// Visually inspecting very small output images is not necessary. We will
// cover these cases in unit testing.
if ((size.width() <= 10 || size.height() <= 10) && 1.0f != fScale) {
return Result::Skip("Scaling very small images is uninteresting.");
}
decodeInfo = decodeInfo.makeDimensions(size);
const int bpp = decodeInfo.bytesPerPixel();
const size_t rowBytes = size.width() * bpp;
const size_t safeSize = decodeInfo.computeByteSize(rowBytes);
SkAutoMalloc pixels(safeSize);
SkCodec::Options options;
if (kCodecZeroInit_Mode == fMode) {
memset(pixels.get(), 0, size.height() * rowBytes);
options.fZeroInitialized = SkCodec::kYes_ZeroInitialized;
}
SkImageInfo bitmapInfo = decodeInfo;
set_bitmap_color_space(&bitmapInfo);
if (kRGBA_8888_SkColorType == decodeInfo.colorType() ||
kBGRA_8888_SkColorType == decodeInfo.colorType()) {
bitmapInfo = bitmapInfo.makeColorType(kN32_SkColorType);
}
switch (fMode) {
case kAnimated_Mode: {
SkAndroidCodec::AndroidOptions androidOptions;
if (fScale != 1.0f) {
SkASSERT(androidCodec);
androidOptions.fSampleSize = 1 / fScale;
auto dims = androidCodec->getSampledDimensions(androidOptions.fSampleSize);
decodeInfo = decodeInfo.makeDimensions(dims);
}
std::vector<SkCodec::FrameInfo> frameInfos = androidCodec
? androidCodec->codec()->getFrameInfo() : codec->getFrameInfo();
if (frameInfos.size() <= 1) {
return Result::Fatal("%s is not an animated image.", fPath.c_str());
}
// As in CodecSrc::size(), compute a roughly square grid to draw the frames
// into. "factor" is the number of frames to draw on one row. There will be
// up to "factor" rows as well.
const float root = sqrt((float) frameInfos.size());
const int factor = sk_float_ceil2int(root);
// Used to cache a frame that future frames will depend on.
SkAutoMalloc priorFramePixels;
int cachedFrame = SkCodec::kNoFrame;
for (int i = 0; static_cast<size_t>(i) < frameInfos.size(); i++) {
androidOptions.fFrameIndex = i;
// Check for a prior frame
const int reqFrame = frameInfos[i].fRequiredFrame;
if (reqFrame != SkCodec::kNoFrame && reqFrame == cachedFrame
&& priorFramePixels.get()) {
// Copy into pixels
memcpy(pixels.get(), priorFramePixels.get(), safeSize);
androidOptions.fPriorFrame = reqFrame;
} else {
androidOptions.fPriorFrame = SkCodec::kNoFrame;
}
SkCodec::Result result = androidCodec
? androidCodec->getAndroidPixels(decodeInfo, pixels.get(), rowBytes,
&androidOptions)
: codec->getPixels(decodeInfo, pixels.get(), rowBytes, &androidOptions);
if (SkCodec::kInvalidInput == result && i > 0) {
// Some of our test images have truncated later frames. Treat that
// the same as incomplete.
result = SkCodec::kIncompleteInput;
}
switch (result) {
case SkCodec::kSuccess:
case SkCodec::kErrorInInput:
case SkCodec::kIncompleteInput: {
// If the next frame depends on this one, store it in priorFrame.
// It is possible that we may discard a frame that future frames depend on,
// but the codec will simply redecode the discarded frame.
// Do this before calling draw_to_canvas, which premultiplies in place. If
// we're decoding to unpremul, we want to pass the unmodified frame to the
// codec for decoding the next frame.
if (static_cast<size_t>(i+1) < frameInfos.size()
&& frameInfos[i+1].fRequiredFrame == i) {
memcpy(priorFramePixels.reset(safeSize), pixels.get(), safeSize);
cachedFrame = i;
}
SkAutoCanvasRestore acr(canvas, true);
const int xTranslate = (i % factor) * decodeInfo.width();
const int yTranslate = (i / factor) * decodeInfo.height();
canvas->translate(SkIntToScalar(xTranslate), SkIntToScalar(yTranslate));
draw_to_canvas(canvas, bitmapInfo, pixels.get(), rowBytes, fDstColorType);
if (result != SkCodec::kSuccess) {
return Result::Ok();
}
break;
}
case SkCodec::kInvalidConversion:
if (i > 0 && (decodeInfo.colorType() == kRGB_565_SkColorType)) {
return Result::Skip(
"Cannot decode frame %i to 565 (%s).", i, fPath.c_str());
}
[[fallthrough]];
default:
return Result::Fatal(
"Couldn't getPixels for frame %i in %s.", i, fPath.c_str());
}
}
break;
}
case kCodecZeroInit_Mode:
case kCodec_Mode: {
switch (codec->getPixels(decodeInfo, pixels.get(), rowBytes, &options)) {
case SkCodec::kSuccess:
// We consider these to be valid, since we should still decode what is
// available.
case SkCodec::kErrorInInput:
case SkCodec::kIncompleteInput:
break;
default:
// Everything else is considered a failure.
return Result::Fatal("Couldn't getPixels %s.", fPath.c_str());
}
draw_to_canvas(canvas, bitmapInfo, pixels.get(), rowBytes, fDstColorType);
break;
}
case kScanline_Mode: {
void* dst = pixels.get();
uint32_t height = decodeInfo.height();
const bool useIncremental = [this]() {
auto exts = { "png", "PNG", "gif", "GIF" };
for (auto ext : exts) {
if (fPath.endsWith(ext)) {
return true;
}
}
return false;
}();
// ico may use the old scanline method or the new one, depending on whether it
// internally holds a bmp or a png.
const bool ico = fPath.endsWith("ico");
bool useOldScanlineMethod = !useIncremental && !ico;
if (useIncremental || ico) {
if (SkCodec::kSuccess == codec->startIncrementalDecode(decodeInfo, dst,
rowBytes, &options)) {
int rowsDecoded;
auto result = codec->incrementalDecode(&rowsDecoded);
if (SkCodec::kIncompleteInput == result || SkCodec::kErrorInInput == result) {
codec->fillIncompleteImage(decodeInfo, dst, rowBytes,
SkCodec::kNo_ZeroInitialized, height,
rowsDecoded);
}
} else {
if (useIncremental) {
// Error: These should support incremental decode.
return Result::Fatal("Could not start incremental decode");
}
// Otherwise, this is an ICO. Since incremental failed, it must contain a BMP,
// which should work via startScanlineDecode
useOldScanlineMethod = true;
}
}
if (useOldScanlineMethod) {
if (SkCodec::kSuccess != codec->startScanlineDecode(decodeInfo)) {
return Result::Fatal("Could not start scanline decoder");
}
// We do not need to check the return value. On an incomplete
// image, memory will be filled with a default value.
codec->getScanlines(dst, height, rowBytes);
}
draw_to_canvas(canvas, bitmapInfo, dst, rowBytes, fDstColorType);
break;
}
case kStripe_Mode: {
const int height = decodeInfo.height();
// This value is chosen arbitrarily. We exercise more cases by choosing a value that
// does not align with image blocks.
const int stripeHeight = 37;
const int numStripes = (height + stripeHeight - 1) / stripeHeight;
void* dst = pixels.get();
// Decode odd stripes
if (SkCodec::kSuccess != codec->startScanlineDecode(decodeInfo, &options)) {
return Result::Fatal("Could not start scanline decoder");
}
// This mode was designed to test the new skip scanlines API in libjpeg-turbo.
// Jpegs have kTopDown_SkScanlineOrder, and at this time, it is not interesting
// to run this test for image types that do not have this scanline ordering.
// We only run this on Jpeg, which is always kTopDown.
SkASSERT(SkCodec::kTopDown_SkScanlineOrder == codec->getScanlineOrder());
for (int i = 0; i < numStripes; i += 2) {
// Skip a stripe
const int linesToSkip = std::min(stripeHeight, height - i * stripeHeight);
codec->skipScanlines(linesToSkip);
// Read a stripe
const int startY = (i + 1) * stripeHeight;
const int linesToRead = std::min(stripeHeight, height - startY);
if (linesToRead > 0) {
codec->getScanlines(SkTAddOffset<void>(dst, rowBytes * startY), linesToRead,
rowBytes);
}
}
// Decode even stripes
const SkCodec::Result startResult = codec->startScanlineDecode(decodeInfo);
if (SkCodec::kSuccess != startResult) {
return Result::Fatal("Failed to restart scanline decoder with same parameters.");
}
for (int i = 0; i < numStripes; i += 2) {
// Read a stripe
const int startY = i * stripeHeight;
const int linesToRead = std::min(stripeHeight, height - startY);
codec->getScanlines(SkTAddOffset<void>(dst, rowBytes * startY), linesToRead,
rowBytes);
// Skip a stripe
const int linesToSkip = std::min(stripeHeight, height - (i + 1) * stripeHeight);
if (linesToSkip > 0) {
codec->skipScanlines(linesToSkip);
}
}
draw_to_canvas(canvas, bitmapInfo, dst, rowBytes, fDstColorType);
break;
}
case kCroppedScanline_Mode: {
const int width = decodeInfo.width();
const int height = decodeInfo.height();
// This value is chosen because, as we move across the image, it will sometimes
// align with the jpeg block sizes and it will sometimes not. This allows us
// to test interestingly different code paths in the implementation.
const int tileSize = 36;
SkIRect subset;
for (int x = 0; x < width; x += tileSize) {
subset = SkIRect::MakeXYWH(x, 0, std::min(tileSize, width - x), height);
options.fSubset = &subset;
if (SkCodec::kSuccess != codec->startScanlineDecode(decodeInfo, &options)) {
return Result::Fatal("Could not start scanline decoder.");
}
codec->getScanlines(SkTAddOffset<void>(pixels.get(), x * bpp), height, rowBytes);
}
draw_to_canvas(canvas, bitmapInfo, pixels.get(), rowBytes, fDstColorType);
break;
}
case kSubset_Mode: {
// Arbitrarily choose a divisor.
int divisor = 2;
// Total width/height of the image.
const int W = codec->getInfo().width();
const int H = codec->getInfo().height();
if (divisor > W || divisor > H) {
return Result::Skip("Cannot codec subset: divisor %d is too big "
"for %s with dimensions (%d x %d)", divisor,
fPath.c_str(), W, H);
}
// subset dimensions
// SkWebpCodec, the only one that supports subsets, requires even top/left boundaries.
const int w = SkAlign2(W / divisor);
const int h = SkAlign2(H / divisor);
SkIRect subset;
options.fSubset = &subset;
SkBitmap subsetBm;
// We will reuse pixel memory from bitmap.
void* dst = pixels.get();
// Keep track of left and top (for drawing subsetBm into canvas). We could use
// fScale * x and fScale * y, but we want integers such that the next subset will start
// where the last one ended. So we'll add decodeInfo.width() and height().
int left = 0;
for (int x = 0; x < W; x += w) {
int top = 0;
for (int y = 0; y < H; y+= h) {
// Do not make the subset go off the edge of the image.
const int preScaleW = std::min(w, W - x);
const int preScaleH = std::min(h, H - y);
subset.setXYWH(x, y, preScaleW, preScaleH);
// And scale
// FIXME: Should we have a version of getScaledDimensions that takes a subset
// into account?
const int scaledW = std::max(1, SkScalarRoundToInt(preScaleW * fScale));
const int scaledH = std::max(1, SkScalarRoundToInt(preScaleH * fScale));
decodeInfo = decodeInfo.makeWH(scaledW, scaledH);
SkImageInfo subsetBitmapInfo = bitmapInfo.makeWH(scaledW, scaledH);
size_t subsetRowBytes = subsetBitmapInfo.minRowBytes();
const SkCodec::Result result = codec->getPixels(decodeInfo, dst, subsetRowBytes,
&options);
switch (result) {
case SkCodec::kSuccess:
case SkCodec::kErrorInInput:
case SkCodec::kIncompleteInput:
break;
default:
return Result::Fatal("subset codec failed to decode (%d, %d, %d, %d) "
"from %s with dimensions (%d x %d)\t error %d",
x, y, decodeInfo.width(), decodeInfo.height(),
fPath.c_str(), W, H, result);
}
draw_to_canvas(canvas, subsetBitmapInfo, dst, subsetRowBytes, fDstColorType,
SkIntToScalar(left), SkIntToScalar(top));
// translate by the scaled height.
top += decodeInfo.height();
}
// translate by the scaled width.
left += decodeInfo.width();
}
return Result::Ok();
}
default:
SkASSERT(false);
return Result::Fatal("Invalid fMode");
}
return Result::Ok();
}
SkISize CodecSrc::size() const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
std::unique_ptr<SkCodec> codec(SkCodec::MakeFromData(encoded));
if (nullptr == codec) {
return {0, 0};
}
if (fMode != kAnimated_Mode) {
return codec->getScaledDimensions(fScale);
}
// We'll draw one of each frame, so make it big enough to hold them all
// in a grid. The grid will be roughly square, with "factor" frames per
// row and up to "factor" rows.
const size_t count = codec->getFrameInfo().size();
const float root = sqrt((float) count);
const int factor = sk_float_ceil2int(root);
auto androidCodec = SkAndroidCodec::MakeFromCodec(std::move(codec));
auto imageSize = androidCodec->getSampledDimensions(1 / fScale);
imageSize.fWidth = imageSize.fWidth * factor;
imageSize.fHeight = imageSize.fHeight * sk_float_ceil2int((float) count / (float) factor);
return imageSize;
}
Name CodecSrc::name() const {
Name name = SkOSPath::Basename(fPath.c_str());
if (fMode == kAnimated_Mode) {
name.append("_animated");
}
if (1.0f == fScale) {
return name;
}
return get_scaled_name(name.c_str(), fScale);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
AndroidCodecSrc::AndroidCodecSrc(Path path, CodecSrc::DstColorType dstColorType,
SkAlphaType dstAlphaType, int sampleSize)
: fPath(path)
, fDstColorType(dstColorType)
, fDstAlphaType(dstAlphaType)
, fSampleSize(sampleSize)
, fRunSerially(serial_from_path_name(path))
{}
bool AndroidCodecSrc::veto(SinkFlags flags) const {
// No need to test decoding to non-raster or indirect backend.
return flags.type != SinkFlags::kRaster
|| flags.approach != SinkFlags::kDirect;
}
Result AndroidCodecSrc::draw(SkCanvas* canvas) const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
if (!encoded) {
return Result::Fatal("Couldn't read %s.", fPath.c_str());
}
std::unique_ptr<SkAndroidCodec> codec(SkAndroidCodec::MakeFromData(encoded));
if (nullptr == codec) {
return Result::Fatal("Couldn't create android codec for %s.", fPath.c_str());
}
SkImageInfo decodeInfo = codec->getInfo();
if (!get_decode_info(&decodeInfo, canvas->imageInfo().colorType(), fDstColorType,
fDstAlphaType)) {
return Result::Skip("Skipping uninteresting test.");
}
// Scale the image if it is desired.
SkISize size = codec->getSampledDimensions(fSampleSize);
// Visually inspecting very small output images is not necessary. We will
// cover these cases in unit testing.
if ((size.width() <= 10 || size.height() <= 10) && 1 != fSampleSize) {
return Result::Skip("Scaling very small images is uninteresting.");
}
decodeInfo = decodeInfo.makeDimensions(size);
int bpp = decodeInfo.bytesPerPixel();
size_t rowBytes = size.width() * bpp;
SkAutoMalloc pixels(size.height() * rowBytes);
SkBitmap bitmap;
SkImageInfo bitmapInfo = decodeInfo;
set_bitmap_color_space(&bitmapInfo);
if (kRGBA_8888_SkColorType == decodeInfo.colorType() ||
kBGRA_8888_SkColorType == decodeInfo.colorType()) {
bitmapInfo = bitmapInfo.makeColorType(kN32_SkColorType);
}
// Create options for the codec.
SkAndroidCodec::AndroidOptions options;
options.fSampleSize = fSampleSize;
switch (codec->getAndroidPixels(decodeInfo, pixels.get(), rowBytes, &options)) {
case SkCodec::kSuccess:
case SkCodec::kErrorInInput:
case SkCodec::kIncompleteInput:
break;
default:
return Result::Fatal("Couldn't getPixels %s.", fPath.c_str());
}
draw_to_canvas(canvas, bitmapInfo, pixels.get(), rowBytes, fDstColorType);
return Result::Ok();
}
SkISize AndroidCodecSrc::size() const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
std::unique_ptr<SkAndroidCodec> codec(SkAndroidCodec::MakeFromData(encoded));
if (nullptr == codec) {
return {0, 0};
}
return codec->getSampledDimensions(fSampleSize);
}
Name AndroidCodecSrc::name() const {
// We will replicate the names used by CodecSrc so that images can
// be compared in Gold.
if (1 == fSampleSize) {
return SkOSPath::Basename(fPath.c_str());
}
return get_scaled_name(fPath, 1.0f / (float) fSampleSize);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
ImageGenSrc::ImageGenSrc(Path path, Mode mode, SkAlphaType alphaType, bool isGpu)
: fPath(path)
, fMode(mode)
, fDstAlphaType(alphaType)
, fIsGpu(isGpu)
, fRunSerially(serial_from_path_name(path))
{}
bool ImageGenSrc::veto(SinkFlags flags) const {
if (fIsGpu) {
// MSAA runs tend to run out of memory and tests the same code paths as regular gpu configs.
return flags.type != SinkFlags::kGPU || flags.approach != SinkFlags::kDirect ||
flags.multisampled == SinkFlags::kMultisampled;
}
return flags.type != SinkFlags::kRaster || flags.approach != SinkFlags::kDirect;
}
Result ImageGenSrc::draw(SkCanvas* canvas) const {
if (kRGB_565_SkColorType == canvas->imageInfo().colorType()) {
return Result::Skip("Uninteresting to test image generator to 565.");
}
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
if (!encoded) {
return Result::Fatal("Couldn't read %s.", fPath.c_str());
}
#if defined(SK_BUILD_FOR_WIN)
// Initialize COM in order to test with WIC.
SkAutoCoInitialize com;
if (!com.succeeded()) {
return Result::Fatal("Could not initialize COM.");
}
#endif
std::unique_ptr<SkImageGenerator> gen(nullptr);
switch (fMode) {
case kCodec_Mode:
gen = SkCodecImageGenerator::MakeFromEncodedCodec(encoded);
if (!gen) {
return Result::Fatal("Could not create codec image generator.");
}
break;
case kPlatform_Mode: {
#if defined(SK_BUILD_FOR_MAC) || defined(SK_BUILD_FOR_IOS)
gen = SkImageGeneratorCG::MakeFromEncodedCG(encoded);
#elif defined(SK_BUILD_FOR_WIN)
gen = SkImageGeneratorWIC::MakeFromEncodedWIC(encoded);
#elif defined(SK_ENABLE_NDK_IMAGES)
gen = SkImageGeneratorNDK::MakeFromEncodedNDK(encoded);
#endif
if (!gen) {
return Result::Fatal("Could not create platform image generator.");
}
break;
}
default:
SkASSERT(false);
return Result::Fatal("Invalid image generator mode");
}
// Test deferred decoding path on GPU
if (fIsGpu) {
sk_sp<SkImage> image(SkImage::MakeFromGenerator(std::move(gen)));
if (!image) {
return Result::Fatal("Could not create image from codec image generator.");
}
canvas->drawImage(image, 0, 0);
return Result::Ok();
}
// Test various color and alpha types on CPU
SkImageInfo decodeInfo = gen->getInfo().makeAlphaType(fDstAlphaType);
int bpp = decodeInfo.bytesPerPixel();
size_t rowBytes = decodeInfo.width() * bpp;
SkAutoMalloc pixels(decodeInfo.height() * rowBytes);
if (!gen->getPixels(decodeInfo, pixels.get(), rowBytes)) {
Result::Status status = Result::Status::Fatal;
#if defined(SK_BUILD_FOR_WIN)
if (kPlatform_Mode == fMode) {
// Do not issue a fatal error for WIC flakiness.
status = Result::Status::Skip;
}
#endif
return Result(
status,
SkStringPrintf("Image generator could not getPixels() for %s\n", fPath.c_str()));
}
set_bitmap_color_space(&decodeInfo);
draw_to_canvas(canvas, decodeInfo, pixels.get(), rowBytes,
CodecSrc::kGetFromCanvas_DstColorType);
return Result::Ok();
}
SkISize ImageGenSrc::size() const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
std::unique_ptr<SkCodec> codec(SkCodec::MakeFromData(encoded));
if (nullptr == codec) {
return {0, 0};
}
return codec->getInfo().dimensions();
}
Name ImageGenSrc::name() const {
return SkOSPath::Basename(fPath.c_str());
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
ColorCodecSrc::ColorCodecSrc(Path path, bool decode_to_dst) : fPath(path)
, fDecodeToDst(decode_to_dst) {}
bool ColorCodecSrc::veto(SinkFlags flags) const {
// Test to direct raster backends (8888 and 565).
return flags.type != SinkFlags::kRaster || flags.approach != SinkFlags::kDirect;
}
Result ColorCodecSrc::draw(SkCanvas* canvas) const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
if (!encoded) {
return Result::Fatal("Couldn't read %s.", fPath.c_str());
}
std::unique_ptr<SkCodec> codec(SkCodec::MakeFromData(encoded));
if (nullptr == codec) {
return Result::Fatal("Couldn't create codec for %s.", fPath.c_str());
}
SkImageInfo info = codec->getInfo();
if (fDecodeToDst) {
SkImageInfo canvasInfo = canvas->imageInfo();
if (!canvasInfo.colorSpace()) {
// This will skip color conversion, and the resulting images will
// look different from images they are compared against in Gold, but
// that doesn't mean they are wrong. We have a test verifying that
// passing a null SkColorSpace skips conversion, so skip this
// misleading test.
return Result::Skip("Skipping decoding without color transform.");
}
info = canvasInfo.makeDimensions(info.dimensions());
}
auto [image, result] = codec->getImage(info);
switch (result) {
case SkCodec::kSuccess:
case SkCodec::kErrorInInput:
case SkCodec::kIncompleteInput:
canvas->drawImage(image, 0,0);
return Result::Ok();
case SkCodec::kInvalidConversion:
// TODO(mtklein): why are there formats we can't decode to?
return Result::Skip("SkCodec can't decode to this format.");
default:
return Result::Fatal("Couldn't getPixels %s. Error code %d", fPath.c_str(), result);
}
}
SkISize ColorCodecSrc::size() const {
sk_sp<SkData> encoded(SkData::MakeFromFileName(fPath.c_str()));
std::unique_ptr<SkCodec> codec(SkCodec::MakeFromData(encoded));
if (nullptr == codec) {
return {0, 0};
}
return {codec->getInfo().width(), codec->getInfo().height()};
}
Name ColorCodecSrc::name() const {
return SkOSPath::Basename(fPath.c_str());
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static DEFINE_int(skpViewportSize, 1000,
"Width & height of the viewport used to crop skp rendering.");
SKPSrc::SKPSrc(Path path) : fPath(path) { }
Result SKPSrc::draw(SkCanvas* canvas) const {
struct DeserializationContext {
GrDirectContext* fDirectContext = nullptr;
#if defined(SK_GRAPHITE)
skgpu::graphite::Recorder* fRecorder = nullptr;
#endif
} ctx {
GrAsDirectContext(canvas->recordingContext()),
#if defined(SK_GRAPHITE)
canvas->recorder()
#endif
};
SkDeserialProcs procs;
procs.fImageProc = [](const void* data, size_t size, void* ctx) -> sk_sp<SkImage> {
sk_sp<SkData> tmpData = SkData::MakeWithoutCopy(data, size);
sk_sp<SkImage> image = SkImage::MakeFromEncoded(std::move(tmpData));
image = image->makeRasterImage(); // force decoding
if (image) {
DeserializationContext* context = reinterpret_cast<DeserializationContext*>(ctx);
if (context->fDirectContext) {
image = image->makeTextureImage(context->fDirectContext);
}
}
return image;
};
procs.fImageCtx = &ctx;
std::unique_ptr<SkStream> stream = SkStream::MakeFromFile(fPath.c_str());
if (!stream) {
return Result::Fatal("Couldn't read %s.", fPath.c_str());
}
sk_sp<SkPicture> pic(SkPicture::MakeFromStream(stream.get(), &procs));
if (!pic) {
return Result::Fatal("Couldn't parse file %s.", fPath.c_str());
}
stream = nullptr; // Might as well drop this when we're done with it.
canvas->clipRect(SkRect::MakeWH(FLAGS_skpViewportSize, FLAGS_skpViewportSize));
canvas->drawPicture(pic);
return Result::Ok();
}
static SkRect get_cull_rect_for_skp(const char* path) {
std::unique_ptr<SkStream> stream = SkStream::MakeFromFile(path);
if (!stream) {
return SkRect::MakeEmpty();
}
SkPictInfo info;
if (!SkPicture_StreamIsSKP(stream.get(), &info)) {
return SkRect::MakeEmpty();
}
return info.fCullRect;
}
SkISize SKPSrc::size() const {
SkRect viewport = get_cull_rect_for_skp(fPath.c_str());
if (!viewport.intersect((SkRect::MakeWH(FLAGS_skpViewportSize, FLAGS_skpViewportSize)))) {
return {0, 0};
}
return viewport.roundOut().size();
}
Name SKPSrc::name() const { return SkOSPath::Basename(fPath.c_str()); }
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
BisectSrc::BisectSrc(Path path, const char* trail) : INHERITED(path), fTrail(trail) {}
Result BisectSrc::draw(SkCanvas* canvas) const {
struct FoundPath {
SkPath fPath;
SkPaint fPaint;
SkMatrix fViewMatrix;
};
// This subclass of SkCanvas just extracts all the SkPaths (drawn via drawPath) from an SKP.
class PathFindingCanvas : public SkCanvas {
public:
PathFindingCanvas(int width, int height) : SkCanvas(width, height, nullptr) {}
const TArray<FoundPath>& foundPaths() const { return fFoundPaths; }
private:
void onDrawPath(const SkPath& path, const SkPaint& paint) override {
fFoundPaths.push_back() = {path, paint, this->getTotalMatrix()};
}
TArray<FoundPath> fFoundPaths;
};
PathFindingCanvas pathFinder(canvas->getBaseLayerSize().width(),
canvas->getBaseLayerSize().height());
Result result = this->INHERITED::draw(&pathFinder);
if (!result.isOk()) {
return result;
}
int start = 0, end = pathFinder.foundPaths().size();
for (const char* ch = fTrail.c_str(); *ch; ++ch) {
int midpt = (start + end) / 2;
if ('l' == *ch) {
start = midpt;
} else if ('r' == *ch) {
end = midpt;
}
}
for (int i = start; i < end; ++i) {
const FoundPath& path = pathFinder.foundPaths()[i];
SkAutoCanvasRestore acr(canvas, true);
canvas->concat(path.fViewMatrix);
canvas->drawPath(path.fPath, path.fPaint);
}
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
#if defined(SK_ENABLE_SKOTTIE)
static DEFINE_bool(useLottieGlyphPaths, false,
"Prioritize embedded glyph paths over native fonts.");
SkottieSrc::SkottieSrc(Path path) : fPath(std::move(path)) {}
Result SkottieSrc::draw(SkCanvas* canvas) const {
auto resource_provider =
skresources::DataURIResourceProviderProxy::Make(
skresources::FileResourceProvider::Make(SkOSPath::Dirname(fPath.c_str()),
/*predecode=*/true),
/*predecode=*/true);
static constexpr char kInterceptPrefix[] = "__";
auto precomp_interceptor =
sk_make_sp<skottie_utils::ExternalAnimationPrecompInterceptor>(resource_provider,
kInterceptPrefix);
uint32_t flags = 0;
if (FLAGS_useLottieGlyphPaths) {
flags |= skottie::Animation::Builder::kPreferEmbeddedFonts;
}
auto animation = skottie::Animation::Builder(flags)
.setResourceProvider(std::move(resource_provider))
.setPrecompInterceptor(std::move(precomp_interceptor))
.makeFromFile(fPath.c_str());
if (!animation) {
return Result::Fatal("Unable to parse file: %s", fPath.c_str());
}
canvas->drawColor(SK_ColorWHITE);
const auto t_rate = 1.0f / (kTileCount * kTileCount - 1);
// Draw the frames in a shuffled order to exercise non-linear
// frame progression. The film strip will still be in order left-to-right,
// top-down, just not drawn in that order.
static constexpr int frameOrder[] = { 4, 0, 3, 1, 2 };
static_assert(std::size(frameOrder) == kTileCount, "");
for (int i = 0; i < kTileCount; ++i) {
const SkScalar y = frameOrder[i] * kTileSize;
for (int j = 0; j < kTileCount; ++j) {
const SkScalar x = frameOrder[j] * kTileSize;
SkRect dest = SkRect::MakeXYWH(x, y, kTileSize, kTileSize);
const auto t = t_rate * (frameOrder[i] * kTileCount + frameOrder[j]);
{
SkAutoCanvasRestore acr(canvas, true);
canvas->clipRect(dest, true);
canvas->concat(SkMatrix::RectToRect(SkRect::MakeSize(animation->size()), dest,
SkMatrix::kCenter_ScaleToFit));
animation->seek(t);
animation->render(canvas);
}
}
}
return Result::Ok();
}
SkISize SkottieSrc::size() const {
return SkISize::Make(kTargetSize, kTargetSize);
}
Name SkottieSrc::name() const { return SkOSPath::Basename(fPath.c_str()); }
bool SkottieSrc::veto(SinkFlags flags) const {
// No need to test to non-(raster||gpu||vector) or indirect backends.
bool type_ok = flags.type == SinkFlags::kRaster
|| flags.type == SinkFlags::kGPU
|| flags.type == SinkFlags::kVector;
return !type_ok || flags.approach != SinkFlags::kDirect;
}
#endif
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
#if defined(SK_ENABLE_SVG)
// Used when the image doesn't have an intrinsic size.
static const SkSize kDefaultSVGSize = {1000, 1000};
// Used to force-scale tiny fixed-size images.
static const SkSize kMinimumSVGSize = {128, 128};
SVGSrc::SVGSrc(Path path)
: fName(SkOSPath::Basename(path.c_str()))
, fScale(1) {
auto stream = SkStream::MakeFromFile(path.c_str());
if (!stream) {
return;
}
auto rp = skresources::DataURIResourceProviderProxy::Make(
skresources::FileResourceProvider::Make(SkOSPath::Dirname(path.c_str()),
/*predecode=*/true),
/*predecode=*/true);
fDom = SkSVGDOM::Builder().setResourceProvider(std::move(rp))
.make(*stream);
if (!fDom) {
return;
}
const SkSize& sz = fDom->containerSize();
if (sz.isEmpty()) {
// no intrinsic size
fDom->setContainerSize(kDefaultSVGSize);
} else {
fScale = std::max(1.f, std::max(kMinimumSVGSize.width() / sz.width(),
kMinimumSVGSize.height() / sz.height()));
}
}
Result SVGSrc::draw(SkCanvas* canvas) const {
if (!fDom) {
return Result::Fatal("Unable to parse file: %s", fName.c_str());
}
SkAutoCanvasRestore acr(canvas, true);
canvas->scale(fScale, fScale);
canvas->drawColor(SK_ColorWHITE);
fDom->render(canvas);
return Result::Ok();
}
SkISize SVGSrc::size() const {
if (!fDom) {
return {0, 0};
}
return SkSize{fDom->containerSize().width() * fScale, fDom->containerSize().height() * fScale}
.toRound();
}
Name SVGSrc::name() const { return fName; }
bool SVGSrc::veto(SinkFlags flags) const {
// No need to test to non-(raster||gpu||vector) or indirect backends.
bool type_ok = flags.type == SinkFlags::kRaster
|| flags.type == SinkFlags::kGPU
|| flags.type == SinkFlags::kVector;
return !type_ok || flags.approach != SinkFlags::kDirect;
}
#endif // defined(SK_ENABLE_SVG)
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
MSKPSrc::MSKPSrc(Path path) : fPath(path) {
std::unique_ptr<SkStreamAsset> stream = SkStream::MakeFromFile(fPath.c_str());
int count = SkMultiPictureDocumentReadPageCount(stream.get());
if (count > 0) {
fPages.reset(count);
(void)SkMultiPictureDocumentReadPageSizes(stream.get(), &fPages[0], fPages.size());
}
}
int MSKPSrc::pageCount() const { return fPages.size(); }
SkISize MSKPSrc::size() const { return this->size(0); }
SkISize MSKPSrc::size(int i) const {
return i >= 0 && i < fPages.size() ? fPages[i].fSize.toCeil() : SkISize{0, 0};
}
Result MSKPSrc::draw(SkCanvas* c) const {
return this->draw(0, c);
}
Result MSKPSrc::draw(int i, SkCanvas* canvas) const {
if (this->pageCount() == 0) {
return Result::Fatal("Unable to parse MultiPictureDocument file: %s", fPath.c_str());
}
if (i >= fPages.size() || i < 0) {
return Result::Fatal("MultiPictureDocument page number out of range: %d", i);
}
SkPicture* page = fPages[i].fPicture.get();
if (!page) {
std::unique_ptr<SkStreamAsset> stream = SkStream::MakeFromFile(fPath.c_str());
if (!stream) {
return Result::Fatal("Unable to open file: %s", fPath.c_str());
}
if (!SkMultiPictureDocumentRead(stream.get(), &fPages[0], fPages.size())) {
return Result::Fatal("SkMultiPictureDocument reader failed on page %d: %s", i,
fPath.c_str());
}
page = fPages[i].fPicture.get();
}
canvas->drawPicture(page);
return Result::Ok();
}
Name MSKPSrc::name() const { return SkOSPath::Basename(fPath.c_str()); }
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
Result NullSink::draw(const Src& src, SkBitmap*, SkWStream*, SkString*) const {
return src.draw(SkMakeNullCanvas().get());
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static Result compare_bitmaps(const SkBitmap& reference, const SkBitmap& bitmap) {
// The dimensions are a property of the Src only, and so should be identical.
SkASSERT(reference.computeByteSize() == bitmap.computeByteSize());
if (reference.computeByteSize() != bitmap.computeByteSize()) {
return Result::Fatal("Dimensions don't match reference");
}
// All SkBitmaps in DM are tight, so this comparison is easy.
if (0 != memcmp(reference.getPixels(), bitmap.getPixels(), reference.computeByteSize())) {
SkString encoded;
SkString errString("Pixels don't match reference");
if (BitmapToBase64DataURI(reference, &encoded)) {
errString.append("\nExpected: ");
errString.append(encoded);
} else {
errString.append("\nExpected image failed to encode: ");
errString.append(encoded);
}
if (BitmapToBase64DataURI(bitmap, &encoded)) {
errString.append("\nActual: ");
errString.append(encoded);
} else {
errString.append("\nActual image failed to encode: ");
errString.append(encoded);
}
return Result(Result::Status::Fatal, errString);
}
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static DEFINE_bool(gpuStats, false, "Append GPU stats to the log for each GPU task?");
static DEFINE_bool(preAbandonGpuContext, false,
"Test abandoning the GrContext before running the test.");
static DEFINE_bool(abandonGpuContext, false,
"Test abandoning the GrContext after running each test.");
static DEFINE_bool(releaseAndAbandonGpuContext, false,
"Test releasing all gpu resources and abandoning the GrContext "
"after running each test");
static DEFINE_bool(drawOpClip, false, "Clip each GrDrawOp to its device bounds for testing.");
static DEFINE_bool(programBinaryCache, true, "Use in-memory program binary cache");
GPUSink::GPUSink(const SkCommandLineConfigGpu* config,
const GrContextOptions& grCtxOptions)
: fContextType(config->getContextType())
, fContextOverrides(config->getContextOverrides())
, fSurfType(config->getSurfType())
, fSampleCount(config->getSamples())
, fSurfaceFlags(config->getSurfaceFlags())
, fColorType(config->getColorType())
, fAlphaType(config->getAlphaType())
, fBaseContextOptions(grCtxOptions) {
if (FLAGS_programBinaryCache) {
fBaseContextOptions.fPersistentCache = &fMemoryCache;
}
}
Result GPUSink::draw(const Src& src, SkBitmap* dst, SkWStream* dstStream, SkString* log) const {
return this->onDraw(src, dst, dstStream, log, fBaseContextOptions);
}
sk_sp<SkSurface> GPUSink::createDstSurface(GrDirectContext* context, SkISize size) const {
sk_sp<SkSurface> surface;
SkImageInfo info = SkImageInfo::Make(size, this->colorInfo());
SkSurfaceProps props(fSurfaceFlags, kRGB_H_SkPixelGeometry);
switch (fSurfType) {
case SkCommandLineConfigGpu::SurfType::kDefault:
surface = SkSurface::MakeRenderTarget(
context, skgpu::Budgeted::kNo, info, fSampleCount, &props);
break;
case SkCommandLineConfigGpu::SurfType::kBackendTexture:
surface = sk_gpu_test::MakeBackendTextureSurface(context,
info,
kTopLeft_GrSurfaceOrigin,
fSampleCount,
GrMipmapped::kNo,
GrProtected::kNo,
&props);
break;
case SkCommandLineConfigGpu::SurfType::kBackendRenderTarget:
surface = sk_gpu_test::MakeBackendRenderTargetSurface(context,
info,
kBottomLeft_GrSurfaceOrigin,
fSampleCount,
GrProtected::kNo,
&props);
break;
}
return surface;
}
bool GPUSink::readBack(SkSurface* surface, SkBitmap* dst) const {
SkCanvas* canvas = surface->getCanvas();
SkISize size = surface->imageInfo().dimensions();
SkImageInfo info = SkImageInfo::Make(size, this->colorInfo());
dst->allocPixels(info);
return canvas->readPixels(*dst, 0, 0);
}
Result GPUSink::onDraw(const Src& src, SkBitmap* dst, SkWStream*, SkString* log,
const GrContextOptions& baseOptions,
std::function<void(GrDirectContext*)> initContext,
std::function<SkCanvas*(SkCanvas*)> wrapCanvas) const {
GrContextOptions grOptions = baseOptions;
// We don't expect the src to mess with the persistent cache or the executor.
SkDEBUGCODE(auto cache = grOptions.fPersistentCache);
SkDEBUGCODE(auto exec = grOptions.fExecutor);
src.modifyGrContextOptions(&grOptions);
SkASSERT(cache == grOptions.fPersistentCache);
SkASSERT(exec == grOptions.fExecutor);
GrContextFactory factory(grOptions);
auto direct = factory.getContextInfo(fContextType, fContextOverrides).directContext();
if (initContext) {
initContext(direct);
}
const int maxDimension = direct->priv().caps()->maxTextureSize();
if (maxDimension < std::max(src.size().width(), src.size().height())) {
return Result::Skip("Src too large to create a texture.\n");
}
sk_sp<SkSurface> surface = this->createDstSurface(direct, src.size());
if (!surface) {
return Result::Fatal("Could not create a surface.");
}
if (FLAGS_preAbandonGpuContext) {
factory.abandonContexts();
}
auto canvas = surface->getCanvas();
if (wrapCanvas != nullptr) {
canvas = wrapCanvas(canvas);
}
Result result = src.draw(canvas);
if (!result.isOk()) {
return result;
}
surface->flushAndSubmit();
if (FLAGS_gpuStats) {
direct->priv().dumpCacheStats(log);
direct->priv().dumpGpuStats(log);
direct->priv().dumpContextStats(log);
}
this->readBack(surface.get(), dst);
if (FLAGS_abandonGpuContext) {
factory.abandonContexts();
} else if (FLAGS_releaseAndAbandonGpuContext) {
factory.releaseResourcesAndAbandonContexts();
}
if (grOptions.fPersistentCache) {
direct->storeVkPipelineCacheData();
}
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPUSlugSink::GPUSlugSink(const SkCommandLineConfigGpu* config, const GrContextOptions& options)
: GPUSink(config, options) {}
Result GPUSlugSink::draw(const Src& src, SkBitmap* dst, SkWStream* write, SkString* log) const {
GrContextOptions grOptions = this->baseContextOptions();
// Force padded atlas entries for slug drawing.
grOptions.fSupportBilerpFromGlyphAtlas |= true;
SkTLazy<SkTestCanvas<SkSlugTestKey>> testCanvas;
return onDraw(src, dst, write, log, grOptions, nullptr,
[&](SkCanvas* canvas){
testCanvas.init(canvas);
return testCanvas.get();
});
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPUSerializeSlugSink::GPUSerializeSlugSink(
const SkCommandLineConfigGpu* config, const GrContextOptions& options)
: GPUSink(config, options) {}
Result GPUSerializeSlugSink::draw(
const Src& src, SkBitmap* dst, SkWStream* write, SkString* log) const {
GrContextOptions grOptions = this->baseContextOptions();
// Force padded atlas entries for slug drawing.
grOptions.fSupportBilerpFromGlyphAtlas |= true;
SkTLazy<SkTestCanvas<SkSerializeSlugTestKey>> testCanvas;
return onDraw(src, dst, write, log, grOptions, nullptr,
[&](SkCanvas* canvas){
testCanvas.init(canvas);
return testCanvas.get();
});
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPURemoteSlugSink::GPURemoteSlugSink(
const SkCommandLineConfigGpu* config, const GrContextOptions& options)
: GPUSink(config, options) {}
Result GPURemoteSlugSink::draw(
const Src& src, SkBitmap* dst, SkWStream* write, SkString* log) const {
GrContextOptions grOptions = this->baseContextOptions();
// Force padded atlas entries for slug drawing.
grOptions.fSupportBilerpFromGlyphAtlas |= true;
SkTLazy<SkTestCanvas<SkRemoteSlugTestKey>> testCanvas;
return onDraw(src, dst, write, log, grOptions, nullptr,
[&](SkCanvas* canvas) {
testCanvas.init(canvas);
return testCanvas.get();
});
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPUThreadTestingSink::GPUThreadTestingSink(const SkCommandLineConfigGpu* config,
const GrContextOptions& grCtxOptions)
: INHERITED(config, grCtxOptions)
, fExecutor(SkExecutor::MakeFIFOThreadPool(FLAGS_gpuThreads)) {
SkASSERT(fExecutor);
}
Result GPUThreadTestingSink::draw(const Src& src, SkBitmap* dst, SkWStream* wStream,
SkString* log) const {
// Draw twice, once with worker threads, and once without. Verify that we get the same result.
// Also, force us to only use the software path renderer, so we really stress-test the threaded
// version of that code.
GrContextOptions contextOptions = this->baseContextOptions();
contextOptions.fGpuPathRenderers = GpuPathRenderers::kNone;
contextOptions.fExecutor = fExecutor.get();
Result result = this->onDraw(src, dst, wStream, log, contextOptions);
if (!result.isOk() || !dst) {
return result;
}
SkBitmap reference;
SkString refLog;
SkDynamicMemoryWStream refStream;
contextOptions.fExecutor = nullptr;
Result refResult = this->onDraw(src, &reference, &refStream, &refLog, contextOptions);
if (!refResult.isOk()) {
return refResult;
}
return compare_bitmaps(reference, *dst);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPUPersistentCacheTestingSink::GPUPersistentCacheTestingSink(const SkCommandLineConfigGpu* config,
const GrContextOptions& grCtxOptions)
: INHERITED(config, grCtxOptions)
, fCacheType(config->getTestPersistentCache()) {}
Result GPUPersistentCacheTestingSink::draw(const Src& src, SkBitmap* dst, SkWStream* wStream,
SkString* log) const {
// Draw twice, once with a cold cache, and again with a warm cache. Verify that we get the same
// result.
sk_gpu_test::MemoryCache memoryCache;
GrContextOptions contextOptions = this->baseContextOptions();
contextOptions.fPersistentCache = &memoryCache;
if (fCacheType == 2) {
contextOptions.fShaderCacheStrategy = GrContextOptions::ShaderCacheStrategy::kBackendSource;
}
Result result = this->onDraw(src, dst, wStream, log, contextOptions);
if (!result.isOk() || !dst) {
return result;
}
SkBitmap reference;
SkString refLog;
SkDynamicMemoryWStream refStream;
memoryCache.resetCacheStats();
Result refResult = this->onDraw(src, &reference, &refStream, &refLog, contextOptions);
if (!refResult.isOk()) {
return refResult;
}
SkASSERT(!memoryCache.numCacheMisses());
SkASSERT(!memoryCache.numCacheStores());
return compare_bitmaps(reference, *dst);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPUPrecompileTestingSink::GPUPrecompileTestingSink(const SkCommandLineConfigGpu* config,
const GrContextOptions& grCtxOptions)
: INHERITED(config, grCtxOptions) {}
Result GPUPrecompileTestingSink::draw(const Src& src, SkBitmap* dst, SkWStream* wStream,
SkString* log) const {
// Three step process:
// 1) Draw once with an SkSL cache, and store off the shader blobs.
// 2) For the second context, pre-compile the shaders to warm the cache.
// 3) Draw with the second context, ensuring that we get the same result, and no cache misses.
sk_gpu_test::MemoryCache memoryCache;
GrContextOptions contextOptions = this->baseContextOptions();
contextOptions.fPersistentCache = &memoryCache;
contextOptions.fShaderCacheStrategy = GrContextOptions::ShaderCacheStrategy::kSkSL;
Result result = this->onDraw(src, dst, wStream, log, contextOptions);
if (!result.isOk() || !dst) {
return result;
}
auto precompileShaders = [&memoryCache](GrDirectContext* dContext) {
memoryCache.foreach([dContext](sk_sp<const SkData> key,
sk_sp<SkData> data,
const SkString& /*description*/,
int /*count*/) {
SkAssertResult(dContext->precompileShader(*key, *data));
});
};
sk_gpu_test::MemoryCache replayCache;
GrContextOptions replayOptions = this->baseContextOptions();
// Ensure that the runtime cache is large enough to hold all of the shaders we pre-compile
replayOptions.fRuntimeProgramCacheSize = memoryCache.numCacheMisses();
replayOptions.fPersistentCache = &replayCache;
SkBitmap reference;
SkString refLog;
SkDynamicMemoryWStream refStream;
Result refResult = this->onDraw(src, &reference, &refStream, &refLog, replayOptions,
precompileShaders);
if (!refResult.isOk()) {
return refResult;
}
SkASSERT(!replayCache.numCacheMisses());
return compare_bitmaps(reference, *dst);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
GPUDDLSink::GPUDDLSink(const SkCommandLineConfigGpu* config, const GrContextOptions& ctxOptions)
: INHERITED(config, ctxOptions)
, fRecordingExecutor(SkExecutor::MakeLIFOThreadPool(1))
, fGPUExecutor(SkExecutor::MakeFIFOThreadPool(1, false)) {
}
Result GPUDDLSink::ddlDraw(const Src& src,
sk_sp<SkSurface> dstSurface,
SkTaskGroup* recordingTaskGroup,
SkTaskGroup* gpuTaskGroup,
sk_gpu_test::TestContext* gpuTestCtx,
GrDirectContext* dContext) const {
// We have to do this here bc characterization can hit the SkGpuDevice's thread guard (i.e.,
// leaving it until the DDLTileHelper ctor will result in multiple threads trying to use the
// same context (this thread and the gpuThread - which will be uploading textures)).
SkSurfaceCharacterization dstCharacterization;
SkAssertResult(dstSurface->characterize(&dstCharacterization));
auto size = src.size();
SkPictureRecorder recorder;
Result result = src.draw(recorder.beginRecording(SkIntToScalar(size.width()),
SkIntToScalar(size.height())));
if (!result.isOk()) {
return result;
}
sk_sp<SkPicture> inputPicture(recorder.finishRecordingAsPicture());
// this is our ultimate final drawing area/rect
SkIRect viewport = SkIRect::MakeWH(size.fWidth, size.fHeight);
SkYUVAPixmapInfo::SupportedDataTypes supportedYUVADataTypes(*dContext);
DDLPromiseImageHelper promiseImageHelper(supportedYUVADataTypes);
sk_sp<SkPicture> newSKP = promiseImageHelper.recreateSKP(dContext, inputPicture.get());
if (!newSKP) {
return Result::Fatal("GPUDDLSink: Couldn't recreate the SKP");
}
// 'gpuTestCtx/gpuThreadCtx' is being shifted to the gpuThread. Leave the main (this)
// thread w/o a context.
gpuTestCtx->makeNotCurrent();
// Job one for the GPU thread is to make 'gpuTestCtx' current!
gpuTaskGroup->add([gpuTestCtx] { gpuTestCtx->makeCurrent(); });
// TODO: move the image upload to the utility thread
promiseImageHelper.uploadAllToGPU(gpuTaskGroup, dContext);
// Care must be taken when using 'gpuThreadCtx' bc it moves between the gpu-thread and this
// one. About all it can be consistently used for is GrCaps access and 'defaultBackendFormat'
// calls.
constexpr int kNumDivisions = 3;
DDLTileHelper tiles(dContext, dstCharacterization, viewport,
kNumDivisions, kNumDivisions,
/* addRandomPaddingToDst */ false);
tiles.createBackendTextures(gpuTaskGroup, dContext);
tiles.kickOffThreadedWork(recordingTaskGroup, gpuTaskGroup, dContext, newSKP.get());
// We have to wait for the recording threads to schedule all their work on the gpu thread
// before we can schedule the composition draw and the flush. Note that the gpu thread
// is not blocked at this point and this thread is borrowing recording work.
recordingTaskGroup->wait();
// Note: at this point the recording thread(s) are stalled out w/ nothing to do.
if (FLAGS_preAbandonGpuContext) {
dContext->abandonContext();
}
// The recording threads have already scheduled the drawing of each tile's DDL on the gpu
// thread. The composition DDL must be scheduled last bc it relies on the result of all
// the tiles' rendering. Additionally, bc we're aliasing the tiles' backend textures,
// there is nothing in the DAG to automatically force the required order.
gpuTaskGroup->add([dstSurface, ddl = tiles.composeDDL()]() {
dstSurface->draw(ddl);
});
// This should be the only explicit flush for the entire DDL draw.
gpuTaskGroup->add([dContext]() {
// We need to ensure all the GPU work is finished so
// the following 'deleteAllFromGPU' call will work
// on Vulkan.
// TODO: switch over to using the promiseImage callbacks
// to free the backendTextures. This is complicated a
// bit by which thread possesses the direct context.
dContext->flush();
dContext->submit(true);
});
// The backend textures are created on the gpuThread by the 'uploadAllToGPU' call.
// It is simpler to also delete them at this point on the gpuThread.
promiseImageHelper.deleteAllFromGPU(gpuTaskGroup, dContext);
tiles.deleteBackendTextures(gpuTaskGroup, dContext);
// A flush has already been scheduled on the gpu thread along with the clean up of the backend
// textures so it is safe to schedule making 'gpuTestCtx' not current on the gpuThread.
gpuTaskGroup->add([gpuTestCtx] { gpuTestCtx->makeNotCurrent(); });
// All the work is scheduled on the gpu thread, we just need to wait
gpuTaskGroup->wait();
return Result::Ok();
}
Result GPUDDLSink::draw(const Src& src, SkBitmap* dst, SkWStream*, SkString* log) const {
GrContextOptions contextOptions = this->baseContextOptions();
src.modifyGrContextOptions(&contextOptions);
contextOptions.fPersistentCache = nullptr;
contextOptions.fExecutor = nullptr;
GrContextFactory factory(contextOptions);
// This captures the context destined to be the main gpu context
ContextInfo mainCtxInfo = factory.getContextInfo(this->contextType(), this->contextOverrides());
sk_gpu_test::TestContext* mainTestCtx = mainCtxInfo.testContext();
auto mainCtx = mainCtxInfo.directContext();
if (!mainCtx) {
return Result::Fatal("Could not create context.");
}
SkASSERT(mainCtx->priv().getGpu());
// TODO: make use of 'otherCtx' for uploads & compilation
#if 0
// This captures the context destined to be the utility context. It is in a share group
// with the main context
ContextInfo otherCtxInfo = factory.getSharedContextInfo(mainCtx);
sk_gpu_test::TestContext* otherTestCtx = otherCtxInfo.testContext();
auto otherCtx = otherCtxInfo.directContext();
if (!otherCtx) {
return Result::Fatal("Cound not create shared context.");
}
SkASSERT(otherCtx->priv().getGpu());
#endif
SkTaskGroup recordingTaskGroup(*fRecordingExecutor);
SkTaskGroup gpuTaskGroup(*fGPUExecutor);
// Make sure 'mainCtx' is current
mainTestCtx->makeCurrent();
sk_sp<SkSurface> surface = this->createDstSurface(mainCtx, src.size());
if (!surface) {
return Result::Fatal("Could not create a surface.");
}
Result result = this->ddlDraw(src, surface, &recordingTaskGroup, &gpuTaskGroup,
mainTestCtx, mainCtx);
if (!result.isOk()) {
return result;
}
// 'ddlDraw' will have made 'mainCtx' not current on the gpuThread
mainTestCtx->makeCurrent();
if (FLAGS_gpuStats) {
mainCtx->priv().dumpCacheStats(log);
mainCtx->priv().dumpGpuStats(log);
mainCtx->priv().dumpContextStats(log);
#if 0
otherCtx->priv().dumpCacheStats(log);
otherCtx->priv().dumpGpuStats(log);
otherCtx->priv().dumpContextStats(log);
#endif
}
if (!this->readBack(surface.get(), dst)) {
return Result::Fatal("Could not readback from surface.");
}
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static Result draw_skdocument(const Src& src, SkDocument* doc, SkWStream* dst) {
if (src.size().isEmpty()) {
return Result::Fatal("Source has empty dimensions");
}
SkASSERT(doc);
int pageCount = src.pageCount();
for (int i = 0; i < pageCount; ++i) {
int width = src.size(i).width(), height = src.size(i).height();
SkCanvas* canvas =
doc->beginPage(SkIntToScalar(width), SkIntToScalar(height));
if (!canvas) {
return Result::Fatal("SkDocument::beginPage(w,h) returned nullptr");
}
Result result = src.draw(i, canvas);
if (!result.isOk()) {
return result;
}
doc->endPage();
}
doc->close();
dst->flush();
return Result::Ok();
}
Result PDFSink::draw(const Src& src, SkBitmap*, SkWStream* dst, SkString*) const {
SkPDF::Metadata metadata;
metadata.fTitle = src.name();
metadata.fSubject = "rendering correctness test";
metadata.fCreator = "Skia/DM";
metadata.fRasterDPI = fRasterDpi;
metadata.fPDFA = fPDFA;
#if SK_PDF_TEST_EXECUTOR
std::unique_ptr<SkExecutor> executor = SkExecutor::MakeFIFOThreadPool();
metadata.fExecutor = executor.get();
#endif
auto doc = SkPDF::MakeDocument(dst, metadata);
if (!doc) {
return Result::Fatal("SkPDF::MakeDocument() returned nullptr");
}
return draw_skdocument(src, doc.get(), dst);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
XPSSink::XPSSink() {}
#if defined(SK_SUPPORT_XPS)
static SkTScopedComPtr<IXpsOMObjectFactory> make_xps_factory() {
IXpsOMObjectFactory* factory;
HRN(CoCreateInstance(CLSID_XpsOMObjectFactory,
nullptr,
CLSCTX_INPROC_SERVER,
IID_PPV_ARGS(&factory)));
return SkTScopedComPtr<IXpsOMObjectFactory>(factory);
}
Result XPSSink::draw(const Src& src, SkBitmap*, SkWStream* dst, SkString*) const {
SkAutoCoInitialize com;
if (!com.succeeded()) {
return Result::Fatal("Could not initialize COM.");
}
SkTScopedComPtr<IXpsOMObjectFactory> factory = make_xps_factory();
if (!factory) {
return Result::Fatal("Failed to create XPS Factory.");
}
auto doc = SkXPS::MakeDocument(dst, factory.get());
if (!doc) {
return Result::Fatal("SkXPS::MakeDocument() returned nullptr");
}
return draw_skdocument(src, doc.get(), dst);
}
#else
Result XPSSink::draw(const Src& src, SkBitmap*, SkWStream* dst, SkString*) const {
return Result::Fatal("XPS not supported on this platform.");
}
#endif
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
SKPSink::SKPSink() {}
Result SKPSink::draw(const Src& src, SkBitmap*, SkWStream* dst, SkString*) const {
auto size = SkSize::Make(src.size());
SkPictureRecorder recorder;
Result result = src.draw(recorder.beginRecording(size.width(), size.height()));
if (!result.isOk()) {
return result;
}
recorder.finishRecordingAsPicture()->serialize(dst);
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
Result DebugSink::draw(const Src& src, SkBitmap*, SkWStream* dst, SkString*) const {
DebugCanvas debugCanvas(src.size().width(), src.size().height());
Result result = src.draw(&debugCanvas);
if (!result.isOk()) {
return result;
}
std::unique_ptr<SkCanvas> nullCanvas = SkMakeNullCanvas();
UrlDataManager dataManager(SkString("data"));
SkJSONWriter writer(dst, SkJSONWriter::Mode::kPretty);
writer.beginObject(); // root
debugCanvas.toJSON(writer, dataManager, nullCanvas.get());
writer.endObject(); // root
writer.flush();
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
SVGSink::SVGSink(int pageIndex) : fPageIndex(pageIndex) {}
Result SVGSink::draw(const Src& src, SkBitmap*, SkWStream* dst, SkString*) const {
#if defined(SK_ENABLE_SVG)
if (src.pageCount() > 1) {
int pageCount = src.pageCount();
if (fPageIndex > pageCount - 1) {
return Result::Fatal("Page index %d too high for document with only %d pages.",
fPageIndex, pageCount);
}
}
return src.draw(fPageIndex,
SkSVGCanvas::Make(SkRect::MakeWH(SkIntToScalar(src.size().width()),
SkIntToScalar(src.size().height())),
dst)
.get());
#else
(void)fPageIndex;
return Result::Fatal("SVG sink is disabled.");
#endif // SK_ENABLE_SVG
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
RasterSink::RasterSink(SkColorType colorType)
: fColorType(colorType) {}
Result RasterSink::draw(const Src& src, SkBitmap* dst, SkWStream*, SkString*) const {
const SkISize size = src.size();
dst->allocPixelsFlags(SkImageInfo::Make(size, this->colorInfo()),
SkBitmap::kZeroPixels_AllocFlag);
SkCanvas canvas(*dst, SkSurfaceProps(0, kRGB_H_SkPixelGeometry));
return src.draw(&canvas);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
#if defined(SK_GRAPHITE)
GraphiteSink::GraphiteSink(const SkCommandLineConfigGraphite* config)
: fContextType(config->getContextType())
, fColorType(config->getColorType())
, fAlphaType(config->getAlphaType()) {
}
Result GraphiteSink::draw(const Src& src,
SkBitmap* dst,
SkWStream* dstStream,
SkString* log) const {
SkImageInfo ii = SkImageInfo::Make(src.size(), this->colorInfo());
skiatest::graphite::ContextFactory factory;
auto [_, context] = factory.getContextInfo(fContextType);
if (!context) {
return Result::Fatal("Could not create a context.");
}
std::unique_ptr<skgpu::graphite::Recorder> recorder =
context->makeRecorder(ToolUtils::CreateTestingRecorderOptions());
if (!recorder) {
return Result::Fatal("Could not create a recorder.");
}
dst->allocPixels(ii);
{
sk_sp<SkSurface> surface = SkSurface::MakeGraphite(recorder.get(), ii);
if (!surface) {
return Result::Fatal("Could not create a surface.");
}
Result result = src.draw(surface->getCanvas());
if (!result.isOk()) {
return result;
}
SkPixmap pm;
if (!dst->peekPixels(&pm) ||
!surface->readPixels(pm, 0, 0)) {
return Result::Fatal("Could not readback from surface.");
}
}
std::unique_ptr<skgpu::graphite::Recording> recording = recorder->snap();
if (!recording) {
return Result::Fatal("Could not create a recording.");
}
skgpu::graphite::InsertRecordingInfo info;
info.fRecording = recording.get();
if (!context->insertRecording(info)) {
return Result::Fatal("Context::insertRecording failed.");
}
context->submit(skgpu::graphite::SyncToCpu::kYes);
return Result::Ok();
}
#endif
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
// Handy for front-patching a Src. Do whatever up-front work you need, then call draw_to_canvas(),
// passing the Sink draw() arguments, a size, and a function draws into an SkCanvas.
// Several examples below.
using DrawToCanvasFn = std::function<DM::Result(SkCanvas*)>;
static Result draw_to_canvas(Sink* sink, SkBitmap* bitmap, SkWStream* stream,
SkString* log, SkISize size, const DrawToCanvasFn& draw) {
class ProxySrc : public Src {
public:
ProxySrc(SkISize size, const DrawToCanvasFn& draw) : fSize(size), fDraw(draw) {}
Result draw(SkCanvas* canvas) const override {
return fDraw(canvas);
}
Name name() const override { return "ProxySrc"; }
SkISize size() const override { return fSize; }
private:
SkISize fSize;
const DrawToCanvasFn& fDraw;
};
return sink->draw(ProxySrc(size, draw), bitmap, stream, log);
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static DEFINE_bool(check, true, "If true, have most Via- modes fail if they affect the output.");
// Is *bitmap identical to what you get drawing src into sink?
static Result check_against_reference(const SkBitmap* bitmap, const Src& src, Sink* sink) {
// We can only check raster outputs.
// (Non-raster outputs like .pdf, .skp, .svg may differ but still draw identically.)
if (FLAGS_check && bitmap) {
SkBitmap reference;
SkString log;
SkDynamicMemoryWStream wStream;
Result result = sink->draw(src, &reference, &wStream, &log);
// If we can draw into this Sink via some pipeline, we should be able to draw directly.
SkASSERT(result.isOk());
if (!result.isOk()) {
return result;
}
return compare_bitmaps(reference, *bitmap);
}
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
static SkISize auto_compute_translate(SkMatrix* matrix, int srcW, int srcH) {
SkRect bounds = SkRect::MakeIWH(srcW, srcH);
matrix->mapRect(&bounds);
matrix->postTranslate(-bounds.x(), -bounds.y());
return {SkScalarRoundToInt(bounds.width()), SkScalarRoundToInt(bounds.height())};
}
ViaMatrix::ViaMatrix(SkMatrix matrix, Sink* sink) : Via(sink), fMatrix(matrix) {}
Result ViaMatrix::draw(const Src& src, SkBitmap* bitmap, SkWStream* stream, SkString* log) const {
SkMatrix matrix = fMatrix;
SkISize size = auto_compute_translate(&matrix, src.size().width(), src.size().height());
return draw_to_canvas(fSink.get(), bitmap, stream, log, size,
[&](SkCanvas* canvas) {
canvas->concat(matrix);
return src.draw(canvas);
});
}
// Undoes any flip or 90 degree rotate without changing the scale of the bitmap.
// This should be pixel-preserving.
ViaUpright::ViaUpright(SkMatrix matrix, Sink* sink) : Via(sink), fMatrix(matrix) {}
Result ViaUpright::draw(const Src& src, SkBitmap* bitmap, SkWStream* stream, SkString* log) const {
Result result = fSink->draw(src, bitmap, stream, log);
if (!result.isOk()) {
return result;
}
SkMatrix inverse;
if (!fMatrix.rectStaysRect() || !fMatrix.invert(&inverse)) {
return Result::Fatal("Cannot upright --matrix.");
}
SkMatrix upright = SkMatrix::I();
upright.setScaleX(SkScalarSignAsScalar(inverse.getScaleX()));
upright.setScaleY(SkScalarSignAsScalar(inverse.getScaleY()));
upright.setSkewX(SkScalarSignAsScalar(inverse.getSkewX()));
upright.setSkewY(SkScalarSignAsScalar(inverse.getSkewY()));
SkBitmap uprighted;
SkISize size = auto_compute_translate(&upright, bitmap->width(), bitmap->height());
uprighted.allocPixels(bitmap->info().makeDimensions(size));
SkCanvas canvas(uprighted);
canvas.concat(upright);
SkPaint paint;
paint.setBlendMode(SkBlendMode::kSrc);
canvas.drawImage(bitmap->asImage(), 0, 0, SkSamplingOptions(), &paint);
*bitmap = uprighted;
return Result::Ok();
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
Result ViaSerialization::draw(
const Src& src, SkBitmap* bitmap, SkWStream* stream, SkString* log) const {
// Record our Src into a picture.
auto size = src.size();
SkPictureRecorder recorder;
Result result = src.draw(recorder.beginRecording(SkIntToScalar(size.width()),
SkIntToScalar(size.height())));
if (!result.isOk()) {
return result;
}
sk_sp<SkPicture> pic(recorder.finishRecordingAsPicture());
// Serialize it and then deserialize it.
sk_sp<SkPicture> deserialized(SkPicture::MakeFromData(pic->serialize().get()));
result = draw_to_canvas(fSink.get(), bitmap, stream, log, size,
[&](SkCanvas* canvas) {
canvas->drawPicture(deserialized);
return Result::Ok();
});
if (!result.isOk()) {
return result;
}
return check_against_reference(bitmap, src, fSink.get());
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
Result ViaPicture::draw(const Src& src, SkBitmap* bitmap, SkWStream* stream, SkString* log) const {
auto size = src.size();
Result result = draw_to_canvas(fSink.get(), bitmap, stream, log, size,
[&](SkCanvas* canvas) {
SkPictureRecorder recorder;
sk_sp<SkPicture> pic;
Result result = src.draw(recorder.beginRecording(SkIntToScalar(size.width()),
SkIntToScalar(size.height())));
if (!result.isOk()) {
return result;
}
pic = recorder.finishRecordingAsPicture();
canvas->drawPicture(pic);
return result;
});
if (!result.isOk()) {
return result;
}
return check_against_reference(bitmap, src, fSink.get());
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
Result ViaRuntimeBlend::draw(const Src& src,
SkBitmap* bitmap,
SkWStream* stream,
SkString* log) const {
class RuntimeBlendFilterCanvas : public SkPaintFilterCanvas {
public:
RuntimeBlendFilterCanvas(SkCanvas* canvas) : INHERITED(canvas) { }
protected:
bool onFilter(SkPaint& paint) const override {
if (std::optional<SkBlendMode> mode = paint.asBlendMode()) {
paint.setBlender(GetRuntimeBlendForBlendMode(*mode));
}
return true;
}
private:
using INHERITED = SkPaintFilterCanvas;
};
return draw_to_canvas(fSink.get(), bitmap, stream, log, src.size(),
[&](SkCanvas* canvas) {
RuntimeBlendFilterCanvas runtimeBlendCanvas{canvas};
return src.draw(&runtimeBlendCanvas);
});
}
/*~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~*/
#ifdef TEST_VIA_SVG
#include "include/svg/SkSVGCanvas.h"
#include "modules/svg/include/SkSVGDOM.h"
#include "src/xml/SkXMLWriter.h"
Result ViaSVG::draw(const Src& src, SkBitmap* bitmap, SkWStream* stream, SkString* log) const {
auto size = src.size();
return draw_to_canvas(fSink.get(), bitmap, stream, log, size,
[&](SkCanvas* canvas) -> Result {
SkDynamicMemoryWStream wstream;
SkXMLStreamWriter writer(&wstream);
Result result = src.draw(SkSVGCanvas::Make(SkRect::Make(size), &writer).get());
if (!result.isOk()) {
return result;
}
std::unique_ptr<SkStream> rstream(wstream.detachAsStream());
auto dom = SkSVGDOM::MakeFromStream(*rstream);
if (dom) {
dom->setContainerSize(SkSize::Make(size));
dom->render(canvas);
}
return Result::Ok();
});
}
#endif
} // namespace DM