AuroraMiddleware/skia2
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
9d7aa41081
skia2/tests/PathOpsCubicIntersectionTest.cpp

737 lines
28 KiB
C++
Raw Normal View History

Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
/*
* Copyright 2012 Google Inc.
*
* Use of this source code is governed by a BSD-style license that can be
* found in the LICENSE file.
*/
rewrite includes to not need so much -Ifoo Current strategy: everything from the top Things to look at first are the manual changes: - added tools/rewrite_includes.py - removed -Idirectives from BUILD.gn - various compile.sh simplifications - tweak tools/embed_resources.py - update gn/find_headers.py to write paths from the top - update gn/gn_to_bp.py SkUserConfig.h layout so that #include "include/config/SkUserConfig.h" always gets the header we want. No-Presubmit: true Change-Id: I73a4b181654e0e38d229bc456c0d0854bae3363e Reviewed-on: https://skia-review.googlesource.com/c/skia/+/209706 Commit-Queue: Mike Klein <mtklein@google.com> Reviewed-by: Hal Canary <halcanary@google.com> Reviewed-by: Brian Osman <brianosman@google.com> Reviewed-by: Florin Malita <fmalita@chromium.org>
2019-04-23 17:05:21 +00:00
#include "src/core/SkGeometry.h"
#include "src/pathops/SkIntersections.h"
#include "src/pathops/SkPathOpsRect.h"
#include "src/pathops/SkReduceOrder.h"
#include "tests/PathOpsCubicIntersectionTestData.h"
#include "tests/PathOpsTestCommon.h"
#include "tests/Test.h"
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
Remove include of stdlib.h from SkTypes.h. Unfortunately, immintrin.h (which is also included by SkTypes) includes xmmintrin.h which includes mm_malloc.h which includes stdlib.h for malloc even though, from the implementation, it is difficult to see why. Fortunately, arm_neon.h does not seem to be involved in such shenanigans, so building for Android will keep things sane. TBR=reed@google.com Doesn't change Skia API, just moves an include. Review URL: https://codereview.chromium.org/1313203003
2015-08-26 12:15:46 +00:00
#include <stdlib.h>
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
const int firstCubicIntersectionTest = 9;
static void standardTestCases(skiatest::Reporter* reporter) {
for (size_t index = firstCubicIntersectionTest; index < tests_count; ++index) {
int iIndex = static_cast<int>(index);
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic1 = tests[index][0];
const CubicPts& cubic2 = tests[index][1];
SkDCubic c1, c2;
c1.debugSet(cubic1.fPts);
c2.debugSet(cubic2.fPts);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
SkReduceOrder reduce1, reduce2;
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
int order1 = reduce1.reduce(c1, SkReduceOrder::kNo_Quadratics);
int order2 = reduce2.reduce(c2, SkReduceOrder::kNo_Quadratics);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
const bool showSkipped = false;
if (order1 < 4) {
if (showSkipped) {
SkDebugf("%s [%d] cubic1 order=%d\n", __FUNCTION__, iIndex, order1);
}
continue;
}
if (order2 < 4) {
if (showSkipped) {
SkDebugf("%s [%d] cubic2 order=%d\n", __FUNCTION__, iIndex, order2);
}
continue;
}
SkIntersections tIntersections;
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
tIntersections.intersect(c1, c2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
if (!tIntersections.used()) {
if (showSkipped) {
SkDebugf("%s [%d] no intersection\n", __FUNCTION__, iIndex);
}
continue;
}
if (tIntersections.isCoincident(0)) {
if (showSkipped) {
SkDebugf("%s [%d] coincident\n", __FUNCTION__, iIndex);
}
continue;
}
for (int pt = 0; pt < tIntersections.used(); ++pt) {
double tt1 = tIntersections[0][pt];
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDPoint xy1 = c1.ptAtT(tt1);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
double tt2 = tIntersections[1][pt];
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDPoint xy2 = c2.ptAtT(tt2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
if (!xy1.approximatelyEqual(xy2)) {
SkDebugf("%s [%d,%d] x!= t1=%g (%g,%g) t2=%g (%g,%g)\n",
__FUNCTION__, (int)index, pt, tt1, xy1.fX, xy1.fY, tt2, xy2.fX, xy2.fY);
}
REPORTER_ASSERT(reporter, xy1.approximatelyEqual(xy2));
}
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
reporter->bumpTestCount();
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
}
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
static const CubicPts testSet[] = {
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
// FIXME: uncommenting these two will cause this to fail
// this results in two curves very nearly but not exactly coincident
#if 0
{{{67.426548091427676, 37.993772624988935}, {23.483695892376684, 90.476863174921306},
{35.597065061143162, 79.872482633158796}, {75.38634169631932, 18.244890038969412}}},
{{{67.4265481, 37.9937726}, {23.4836959, 90.4768632}, {35.5970651, 79.8724826},
{75.3863417, 18.24489}}},
#endif
{{{0, 0}, {0, 1}, {1, 1}, {1, 0}}},
{{{1, 0}, {0, 0}, {0, 1}, {1, 1}}},
{{{0, 1}, {4, 5}, {1, 0}, {5, 3}}},
{{{0, 1}, {3, 5}, {1, 0}, {5, 4}}},
{{{0, 1}, {1, 6}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {1, 0}, {6, 1}}},
{{{0, 1}, {3, 4}, {1, 0}, {5, 1}}},
{{{0, 1}, {1, 5}, {1, 0}, {4, 3}}},
{{{0, 1}, {1, 2}, {1, 0}, {6, 1}}},
{{{0, 1}, {1, 6}, {1, 0}, {2, 1}}},
{{{0, 1}, {0, 5}, {1, 0}, {4, 0}}},
{{{0, 1}, {0, 4}, {1, 0}, {5, 0}}},
{{{0, 1}, {3, 4}, {1, 0}, {3, 0}}},
{{{0, 1}, {0, 3}, {1, 0}, {4, 3}}},
{{{0, 0}, {1, 2}, {3, 4}, {4, 4}}},
{{{0, 0}, {1, 2}, {3, 4}, {4, 4}}},
{{{4, 4}, {3, 4}, {1, 2}, {0, 0}}},
{{{0, 1}, {2, 3}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {1, 0}, {3, 2}}},
{{{0, 2}, {0, 1}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {2, 0}, {1, 0}}},
{{{0, 1}, {0, 2}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {1, 0}, {2, 0}}},
{{{0, 1}, {1, 6}, {1, 0}, {2, 0}}},
{{{0, 1}, {0, 2}, {1, 0}, {6, 1}}},
{{{0, 1}, {5, 6}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {1, 0}, {6, 5}}},
{{{95.837747722788592, 45.025976907939643}, {16.564570095652982, 0.72959763963222402},
{63.209855865319199, 68.047528419665767}, {57.640240647662544, 59.524565264361243}}},
{{{51.593891741518817, 38.53849970667553}, {62.34752929878772, 74.924924725166022},
{74.810149322641152, 34.17966562983564}, {29.368398119401373, 94.66719277886078}}},
{{{39.765160968417838, 33.060396198677083}, {5.1922921581157908, 66.854301452103215},
{31.619281802149157, 25.269248720849514}, {81.541621071073038, 70.025341524754353}}},
{{{46.078911165743556, 48.259962651999651}, {20.24450549867214, 49.403916182650214},
{0.26325131778756683, 24.46489805563581}, {15.915006546264051, 83.515023059917155}}},
{{{65.454505973241524, 93.881892270353575}, {45.867360264932437, 92.723972719499827},
{2.1464054482739447, 74.636369140183717}, {33.774068594804994, 40.770872887582925}}},
{{{72.963387832494163, 95.659300729473728}, {11.809496633619768, 82.209921247423594},
{13.456139067865974, 57.329313623406605}, {36.060621606214262, 70.867335643091849}}},
{{{32.484981432782945, 75.082940782924624}, {42.467313093350882, 48.131159948246157},
{3.5963115764764657, 43.208665839959245}, {79.442476890721579, 89.709102357602262}}},
{{{18.98573861410177, 93.308887208490106}, {40.405250173250792, 91.039661826118675},
{8.0467721950480584, 42.100282172719147}, {40.883324221187891, 26.030185504830527}}},
{{{7.5374809128872498, 82.441702896003477}, {22.444346930107265, 22.138854312775123},
{66.76091829629658, 50.753805856571446}, {78.193478508942519, 97.7932997968948}}},
{{{97.700573130371311, 53.53260215070685}, {87.72443481149358, 84.575876772671876},
{19.215031396232092, 47.032676472809484}, {11.989686410869325, 10.659507480757082}}},
{{{26.192053931854691, 9.8504326817814416}, {10.174241480498686, 98.476562741434464},
{21.177712558385782, 33.814968789841501}, {75.329030899018534, 55.02231980442177}}},
{{{56.222082700683771, 24.54395039218662}, {95.589995289030483, 81.050822735322086},
{28.180450866082897, 28.837706255185282}, {60.128952916771617, 87.311672180570511}}},
{{{42.449716172390481, 52.379709366885805}, {27.896043159019225, 48.797373636065686},
{92.770268299044233, 89.899302036454571}, {12.102066544863426, 99.43241951960718}}},
{{{45.77532924980639, 45.958701495993274}, {37.458701356062065, 68.393691335056758},
{37.569326692060258, 27.673713456687381}, {60.674866037757539, 62.47349659096146}}},
{{{67.426548091427676, 37.993772624988935}, {23.483695892376684, 90.476863174921306},
{35.597065061143162, 79.872482633158796}, {75.38634169631932, 18.244890038969412}}},
{{{61.336508189019057, 82.693132843213675}, {44.639380902349664, 54.074825790745592},
{16.815615499771951, 20.049704667203923}, {41.866884958868326, 56.735503699973002}}},
{{{18.1312339, 31.6473732}, {95.5711034, 63.5350219}, {92.3283165, 62.0158945},
{18.5656052, 32.1268808}}},
{{{97.402018, 35.7169972}, {33.1127443, 25.8935163}, {1.13970027, 54.9424981},
{56.4860195, 60.529264}}},
};
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
const int testSetCount = (int) SK_ARRAY_COUNT(testSet);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
static const CubicPts newTestSet[] = {
handle degenerate cubic loops fuzzer generated a cubic with a very tiny loop at one end. Breaking the cubic in two generated a piece so small it was discarded, passing the original along. Instead, use the loop detection as a hint that passing the original may fail further along. Keep the big piece, which is more linear than the original; and change its end to match the original input if the small piece isn't worth keeping. R=kjlubick Bug: skia:7480 Change-Id: If4c5cd46dbf7e16526269f2854e4f2278f427461 Reviewed-on: https://skia-review.googlesource.com/103100 Reviewed-by: Kevin Lubick <kjlubick@google.com> Commit-Queue: Cary Clark <caryclark@skia.org>
2018-02-02 20:09:49 +00:00
split tight quads and conics Tight quads and conics may nearly fold over on themselves, confusing coincidence against other curves. Split them at their max curvature early on to avoid complicating later logic. TBR=reed@google.com BUG=skia:5131 GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2357353002 Review-Url: https://codereview.chromium.org/2357353002
2016-09-23 12:47:20 +00:00
{ { { 130.0427549999999997, 11417.41309999999976 },{ 130.2331240000000037, 11418.3192999999992 },{ 131.0370790000000056, 11419 },{ 132, 11419 } } },
{ { { 132, 11419 },{ 130.8954319999999996, 11419 },{ 130, 11418.10449999999946 },{ 130, 11417 } } },
Remove trailing whitespace. Also adds a presubmit to prevent adding trailing whitespace to source code in the future. Change-Id: I41a4df81487f6f00aa19b188f0cac6a3377efde6 Reviewed-on: https://skia-review.googlesource.com/57380 Reviewed-by: Ravi Mistry <rmistry@google.com> Reviewed-by: Mike Reed <reed@google.com> Commit-Queue: Ben Wagner <bungeman@google.com>
2017-10-09 19:45:33 +00:00
Now, path ops natively intersect conics, quads, and cubics in any combination. There are still a class of cubic tests that fail and a handful of undiagnosed failures from skps and fuzz tests, but things are much better overall. Extended tests (150M+) run to completion in release in about 6 minutes; the standard test suite exceeds 100K and finishes in a few seconds on desktops. TBR=reed BUG=skia:3588 Review URL: https://codereview.chromium.org/1037953004
2015-04-20 15:31:59 +00:00
{{{1,3}, {-1.0564518,1.79032254}, {1.45265341,0.229448318}, {1.45381773,0.22913377}}},
{{{1.45381773,0.22913377}, {1.45425761,0.229014933}, {1.0967741,0.451612949}, {0,1}}},
{{{1.64551306f, 3.57876182f}, {0.298127174f, 3.70454836f}, {-0.809808373f, 6.39524937f}, {-3.66666651f, 13.333334f}}},
{{{1, 2}, {1, 2}, {-3.66666651f, 13.333334f}, {5, 6}}},
{{{0.0660428554,1.65340209}, {-0.251940489,1.43560803}, {-0.782382965,-0.196299091}, {3.33333325,-0.666666627}}},
{{{1,3}, {-1.22353387,1.09411383}, {0.319867611,0.12996155}, {0.886705518,0.107543148}}},
{{{-0.13654758,2.10514426}, {-0.585797966,1.89349782}, {-0.807703257,-0.192306399}, {6,-1}}},
{{{1,4}, {-2.25000453,1.42241001}, {1.1314013,0.0505309105}, {1.87140274,0.0363764353}}},
{{{1.3127951622009277, 2.0637707710266113}, {1.8210518360137939, 1.9148571491241455}, {1.6106204986572266, -0.68700540065765381}, {8.5, -2.5}}},
{{{3, 4}, {0.33333325386047363, 1.3333332538604736}, {3.6666667461395264, -0.66666674613952637}, {3.6666665077209473, -0.66666656732559204}}},
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
{{{980.026001,1481.276}, {980.026001,1481.276}, {980.02594,1481.27576}, {980.025879,1481.27527}}},
{{{980.025879,1481.27527}, {980.025452,1481.27222}, {980.023743,1481.26038}, {980.02179,1481.24072}}},
{{{1.80943513,3.07782435}, {1.66686702,2.16806936}, {1.68301272,0}, {3,0}}},
{{{0,1}, {0,3}, {3,2}, {5,2}}},
{{{3.4386673,2.66977954}, {4.06668949,2.17046738}, {4.78887367,1.59629118}, {6,2}}},
{{{1.71985495,3.49467373}, {2.11620402,2.7201426}, {2.91897964,1.15138781}, {6,3}}},
{{{0,1}, {0.392703831,1.78540766}, {0.219947904,2.05676103}, {0.218561709,2.05630541}}},
{{{0.218561709,2.05630541}, {0.216418028,2.05560064}, {0.624105453,1.40486407}, {4.16666651,1.00000012}}},
{{{0, 1}, {3, 5}, {2, 1}, {3, 1}}},
{{{1.01366711f, 2.21379328f}, {1.09074128f, 2.23241305f}, {1.60246587f, 0.451849401f}, {5, 3}}},
{{{0, 1}, {0.541499972f, 3.16599989f}, {1.08299994f, 2.69299984f}, {2.10083938f, 1.80391729f}}},
{{{0.806384504f, 2.85426903f}, {1.52740121f, 1.99355423f}, {2.81689167f, 0.454222918f}, {5, 1}}},
{{{0, 1}, {1.90192389f, 2.90192389f}, {2.59807634f, 2.79422879f}, {3.1076951f, 2.71539044f}}},
{{{2, 3}, {2.36602545f, 3.36602545f}, {2.330127f, 3.06217766f}, {2.28460979f, 2.67691422f}}},
{{{0, 1}, {1.90192389f, 2.90192389f}, {2.59807634f, 2.79422879f}, {3.1076951f, 2.71539044f}}},
{{{2.28460979f, 2.67691422f}, {2.20577145f, 2.00961876f}, {2.09807634f, 1.09807622f}, {4, 3}}},
{{{0, 1}, {0.8211091160774231, 2.0948121547698975}, {0.91805583238601685, 2.515404224395752}, {0.91621249914169312, 2.5146586894989014}}},
{{{0.91621249914169312, 2.5146586894989014}, {0.91132104396820068, 2.5126807689666748}, {0.21079301834106445, -0.45617169141769409}, {10.5, -1.6666665077209473}}},
{{{42.6237564,68.9841232}, {32.449646,81.963089}, {14.7713947,103.565269}, {12.6310005,105.247002}}},
{{{37.2640038,95.3540039}, {37.2640038,95.3540039}, {11.3710003,83.7339935}, {-25.0779991,124.912003}}},
{{{0,1}, {4,5}, {6,0}, {1,0}}},
{{{0,6}, {0,1}, {1,0}, {5,4}}},
{{{0,1}, {4,6}, {5,1}, {6,2}}},
{{{1,5}, {2,6}, {1,0}, {6,4}}},
{{{322, 896.04803466796875}, {314.09201049804687, 833.4376220703125}, {260.24713134765625, 785}, {195, 785}}},
{{{195, 785}, {265.14016723632812, 785}, {322, 842.30755615234375}, {322, 913}}},
{{{1, 4}, {4, 5}, {3, 2}, {6, 3}}},
{{{2, 3}, {3, 6}, {4, 1}, {5, 4}}},
{{{67, 913}, {67, 917.388916015625}, {67.224380493164063, 921.72576904296875}, {67.662384033203125, 926}}},
{{{194, 1041}, {123.85984039306641, 1041}, {67, 983.69244384765625}, {67, 913}}},
{{{1,4}, {1,5}, {6,0}, {5,1}}},
{{{0,6}, {1,5}, {4,1}, {5,1}}},
{{{0,1}, {4,5}, {6,0}, {1,0}}},
{{{0,6}, {0,1}, {1,0}, {5,4}}},
{{{0,1}, {4,6}, {2,0}, {2,0}}},
{{{0,2}, {0,2}, {1,0}, {6,4}}},
Enabling the canvas bit to turn the clip stack into a flat replace exposed around 100 failures when testing the 800K skp set generated from the top 1M web sites. This fixes all but one of those failures. Major changes include: - Replace angle indices with angle pointers. This was motivated by the need to add angles later but not renumber existing angles. - Aggressive segment chase. When the winding is known on a segment, more aggressively passing that winding to adjacent segments allows fragmented data sets to succeed. - Line segments with ends nearly the same are treated as coincident first. - Transfer partial coincidence by observing that if segment A is partially coincident to B and C then B and C may be partially coincident. TBR=reed Author: caryclark@google.com Review URL: https://codereview.chromium.org/272153002
2014-06-17 12:15:38 +00:00
{{{980.9000244140625, 1474.3280029296875}, {980.9000244140625, 1474.3280029296875}, {978.89300537109375, 1471.95703125}, {981.791015625, 1469.487060546875}}},
{{{981.791015625, 1469.487060546875}, {981.791015625, 1469.4859619140625}, {983.3580322265625, 1472.72900390625}, {980.9000244140625, 1474.3280029296875}}},
pathops work in progress BUG= Review URL: https://codereview.chromium.org/52653002 git-svn-id: http://skia.googlecode.com/svn/trunk@12089 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-11-01 17:36:03 +00:00
{{{275,532}, {277.209137,532}, {279,530.209106}, {279,528}}},
{{{278,529}, {278,530.65686}, {276.65686,532}, {275,532}}},
Sanitizing source files in Housekeeper-Nightly git-svn-id: http://skia.googlecode.com/svn/trunk@11585 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-03 07:01:37 +00:00
#if 0 // FIXME: asserts coincidence, not working yet
{{{195, 785}, {124.30755615234375, 785}, {67, 841.85986328125}, {67, 912}}},
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
{{{67, 913}, {67, 842.30755615234375}, {123.85984039306641, 785}, {194, 785}}},
#endif
pathops work in progress BUG= Review URL: https://codereview.chromium.org/52653002 git-svn-id: http://skia.googlecode.com/svn/trunk@12089 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-11-01 17:36:03 +00:00
{{{149,710.001465}, {149.000809,712.209961}, {150.791367,714}, {153,714}}},
{{{154,715}, {151.238571,715}, {149,712.761414}, {149,710}}},
{{{1,2}, {1,2}, {2,0}, {6,0}}},
{{{0,2}, {0,6}, {2,1}, {2,1}}},
{{{0,1}, {2,3}, {5,1}, {4,3}}},
{{{1,5}, {3,4}, {1,0}, {3,2}}},
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
{{{399,657}, {399,661.970581}, {403.029449,666}, {408,666}}},
{{{406,666}, {402.686279,666}, {400,663.313721}, {400,660}}},
{{{0,5}, {3,5}, {3,0}, {3,2}}},
{{{0,3}, {2,3}, {5,0}, {5,3}}},
path ops work in progress path ops work in progress BUG= Review URL: https://codereview.chromium.org/21359002 git-svn-id: http://skia.googlecode.com/svn/trunk@11291 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-09-16 15:55:01 +00:00
{{{132, 11419}, {130.89543151855469, 11419}, {130, 11418.1044921875}, {130, 11417}}},
Sanitizing source files in Housekeeper-Nightly git-svn-id: http://skia.googlecode.com/svn/trunk@10299 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-24 07:01:12 +00:00
{{{3, 4}, {1, 5}, {4, 3}, {6, 4}}},
{{{3, 4}, {4, 6}, {4, 3}, {5, 1}}},
{{{130.04275512695312, 11417.413085937500 },
{130.23312377929687, 11418.319335937500 },
{131.03707885742187, 11419.000000000000 },
{132.00000000000000, 11419.000000000000 }}},
{{{132.00000000000000, 11419.000000000000 },
{130.89543151855469, 11419.000000000000 },
{130.00000000000000, 11418.104492187500 },
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
{130.00000000000000, 11417.000000000000 }}},
Fix compiler warning/error in r10079 git-svn-id: http://skia.googlecode.com/svn/trunk@10081 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-15 15:38:55 +00:00
{{{1.0516976506771041, 2.9684399028541346 },
{1.0604363140895228, 2.9633503074444141 },
{1.0692548215065762, 2.9580354426587459 },
{1.0781560339512140, 2.9525043684031349 }}},
{{{1.0523038101345104, 2.9523755204833737 },
{1.0607035288264237, 2.9580853881628375 },
{1.0690530472271964, 2.9633896794787749 },
{1.0773566568712512, 2.9682969775000219 }}},
{{{1.0386522625066592, 2.9759024812329078 },
{1.0559713690392631, 2.9661782500838885 },
{1.0736041309019990, 2.9555348259177858 },
{1.0915734362784633, 2.9440446879826569 }}},
{{{1.0396670794879301, 2.9435062123457261 },
{1.0565690546812769, 2.9557413250983462 },
{1.0732616463413533, 2.9663369676594282 },
{1.0897791867435489, 2.9753618045797472 }}},
{{{0.8685656183311091, 3.0409266475785208 },
{0.99189542936395292, 3.0212163698184424 },
{1.1302108367493320, 2.9265646471747306 },
{1.2952305904872474, 2.7940808546473788 }}},
{{{0.85437872843682727, 2.7536036928549055 },
{1.0045584590592620, 2.9493041024831705 },
{1.1336998329885613, 3.0248027987251747 },
{1.2593809752247314, 3.0152560315809107 }}},
{{{0, 1}, {1, 6}, {1, 0}, {6, 2}}},
{{{0, 1}, {2, 6}, {1, 0}, {6, 1}}},
path ops near exact Modify line intersections to first - match exact ends - compute intersections - match near ends where the exact ends are preferred, then near matches, then computed matches. This pulls matches towards existing end points when possible, and keeps intersection distances consistent with different line/line line/quad and line/cubic computations. BUG= Review URL: https://codereview.chromium.org/19183003 git-svn-id: http://skia.googlecode.com/svn/trunk@10073 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-15 13:29:13 +00:00
path ops -- rewrite angle sort This is a major change resulting from a minor tweak. In the old code, the intersection point of two curves was shared between them, but the intersection points and end points of sorted edges was computed directly from the intersection T value. In this CL, both intersection points and sorted points are the same, and intermediate control points are computed to preserve their slope. The sort itself has been completely rewritten to be more robust and remove 'magic' checks, conditions that empirically worked but couldn't be rationalized. This CL was triggered by errors generated computing the clips of SKP files. At this point, all 73M standard tests work and at least the first troublesome SKPs work. Review URL: https://codereview.chromium.org/15338003 git-svn-id: http://skia.googlecode.com/svn/trunk@9432 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-06-04 17:59:42 +00:00
{{{134,11414}, {131.990234375,11414}, {130.32666015625,11415.482421875}, {130.04275512695312,11417.4130859375}}},
{{{132,11419}, {130.89543151855469,11419}, {130,11418.1044921875}, {130,11417}}},
{{{132,11419}, {130.89543151855469,11419}, {130,11418.1044921875}, {130,11417}}},
{{{130.04275512695312,11417.4130859375}, {130.23312377929687,11418.3193359375}, {131.03707885742187,11419}, {132,11419}}},
path ops -- fix skp bugs This fixes a series of bugs discovered by running the small set of Skia skp files through pathops to flatten the clips. Review URL: https://codereview.chromium.org/14798004 git-svn-id: http://skia.googlecode.com/svn/trunk@9042 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-05-07 18:51:31 +00:00
{{{0, 1}, {2, 3}, {5, 1}, {4, 3}}},
{{{1, 5}, {3, 4}, {1, 0}, {3, 2}}},
shape ops work in progress Try to fix the 32 bit build by making some math decisions more robust. Rewrite the cubic intersection special case that detects if only end points are shared. Rewrite the angle sort setup that computes whether a cubic bends to the left or right. git-svn-id: http://skia.googlecode.com/svn/trunk@8726 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-17 15:49:16 +00:00
{{{3, 5}, {1, 6}, {5, 0}, {3, 1}}},
{{{0, 5}, {1, 3}, {5, 3}, {6, 1}}},
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
{{{0, 1}, {1, 5}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {1, 0}, {5, 1}}},
{{{1, 3}, {5, 6}, {5, 3}, {5, 4}}},
{{{3, 5}, {4, 5}, {3, 1}, {6, 5}}},
{{{0, 5}, {0, 5}, {5, 4}, {6, 4}}},
{{{4, 5}, {4, 6}, {5, 0}, {5, 0}}},
{{{0, 4}, {1, 3}, {5, 4}, {4, 2}}},
{{{4, 5}, {2, 4}, {4, 0}, {3, 1}}},
{{{0, 2}, {1, 5}, {3, 2}, {4, 1}}},
{{{2, 3}, {1, 4}, {2, 0}, {5, 1}}},
{{{0, 2}, {2, 3}, {5, 1}, {3, 2}}},
{{{1, 5}, {2, 3}, {2, 0}, {3, 2}}},
{{{2, 6}, {4, 5}, {1, 0}, {6, 1}}},
{{{0, 1}, {1, 6}, {6, 2}, {5, 4}}},
{{{0, 1}, {1, 2}, {6, 5}, {5, 4}}},
{{{5, 6}, {4, 5}, {1, 0}, {2, 1}}},
{{{2.5119999999999996, 1.5710000000000002}, {2.6399999999999983, 1.6599999999999997},
{2.8000000000000007, 1.8000000000000003}, {3, 2}}},
{{{2.4181876227114887, 1.9849772580462195}, {2.8269904869227211, 2.009330650246834},
{3.2004679292461624, 1.9942047174679169}, {3.4986199496818058, 2.0035994597094731}}},
{{{2, 3}, {1, 4}, {1, 0}, {6, 0}}},
{{{0, 1}, {0, 6}, {3, 2}, {4, 1}}},
{{{0, 2}, {1, 5}, {1, 0}, {6, 1}}},
{{{0, 1}, {1, 6}, {2, 0}, {5, 1}}},
{{{0, 1}, {1, 5}, {2, 1}, {4, 0}}},
{{{1, 2}, {0, 4}, {1, 0}, {5, 1}}},
{{{0, 1}, {3, 5}, {2, 1}, {3, 1}}},
{{{1, 2}, {1, 3}, {1, 0}, {5, 3}}},
{{{0, 1}, {2, 5}, {6, 0}, {5, 3}}},
{{{0, 6}, {3, 5}, {1, 0}, {5, 2}}},
{{{0, 1}, {3, 6}, {1, 0}, {5, 2}}},
{{{0, 1}, {2, 5}, {1, 0}, {6, 3}}},
{{{1, 2}, {5, 6}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {2, 1}, {6, 5}}},
{{{0, 6}, {1, 2}, {1, 0}, {1, 0}}},
{{{0, 1}, {0, 1}, {6, 0}, {2, 1}}},
{{{0, 2}, {0, 1}, {3, 0}, {1, 0}}},
{{{0, 3}, {0, 1}, {2, 0}, {1, 0}}},
};
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
const int newTestSetCount = (int) SK_ARRAY_COUNT(newTestSet);
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
static void oneOff(skiatest::Reporter* reporter, const CubicPts& cubic1, const CubicPts& cubic2,
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
bool coin) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDCubic c1, c2;
c1.debugSet(cubic1.fPts);
c2.debugSet(cubic2.fPts);
SkASSERT(ValidCubic(c1));
SkASSERT(ValidCubic(c2));
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
#if ONE_OFF_DEBUG
SkDebugf("computed quadratics given\n");
path ops -- rewrite angle sort This is a major change resulting from a minor tweak. In the old code, the intersection point of two curves was shared between them, but the intersection points and end points of sorted edges was computed directly from the intersection T value. In this CL, both intersection points and sorted points are the same, and intermediate control points are computed to preserve their slope. The sort itself has been completely rewritten to be more robust and remove 'magic' checks, conditions that empirically worked but couldn't be rationalized. This CL was triggered by errors generated computing the clips of SKP files. At this point, all 73M standard tests work and at least the first troublesome SKPs work. Review URL: https://codereview.chromium.org/15338003 git-svn-id: http://skia.googlecode.com/svn/trunk@9432 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-06-04 17:59:42 +00:00
SkDebugf(" {{%1.9g,%1.9g}, {%1.9g,%1.9g}, {%1.9g,%1.9g}, {%1.9g,%1.9g}},\n",
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
cubic1[0].fX, cubic1[0].fY, cubic1[1].fX, cubic1[1].fY,
cubic1[2].fX, cubic1[2].fY, cubic1[3].fX, cubic1[3].fY);
path ops -- rewrite angle sort This is a major change resulting from a minor tweak. In the old code, the intersection point of two curves was shared between them, but the intersection points and end points of sorted edges was computed directly from the intersection T value. In this CL, both intersection points and sorted points are the same, and intermediate control points are computed to preserve their slope. The sort itself has been completely rewritten to be more robust and remove 'magic' checks, conditions that empirically worked but couldn't be rationalized. This CL was triggered by errors generated computing the clips of SKP files. At this point, all 73M standard tests work and at least the first troublesome SKPs work. Review URL: https://codereview.chromium.org/15338003 git-svn-id: http://skia.googlecode.com/svn/trunk@9432 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-06-04 17:59:42 +00:00
SkDebugf(" {{%1.9g,%1.9g}, {%1.9g,%1.9g}, {%1.9g,%1.9g}, {%1.9g,%1.9g}},\n",
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
cubic2[0].fX, cubic2[0].fY, cubic2[1].fX, cubic2[1].fY,
cubic2[2].fX, cubic2[2].fY, cubic2[3].fX, cubic2[3].fY);
#endif
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
SkIntersections intersections;
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
intersections.intersect(c1, c2);
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
#if DEBUG_T_SECT_DUMP == 3
SkDebugf("</div>\n\n");
SkDebugf("<script type=\"text/javascript\">\n\n");
SkDebugf("var testDivs = [\n");
for (int index = 1; index <= gDumpTSectNum; ++index) {
SkDebugf("sect%d,\n", index);
pathops version two R=reed@google.com marked 'no commit' to attempt to get trybots to run TBR=reed@google.com Review URL: https://codereview.chromium.org/1002693002
2015-03-24 14:28:17 +00:00
}
#endif
fix tiger b The tiger tests have uncovered numerous bugs. This CL fixes the last of them. If a pair of curves do not intersect, but have one or both ends very close to the opposite curve, consider that an intersection. TBR=reed@google.com BUG=skia:5131 GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2356363003 Review-Url: https://codereview.chromium.org/2356363003
2016-09-26 12:36:58 +00:00
if (coin && intersections.used() < 2) {
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
SkDebugf("");
pathops version two R=reed@google.com marked 'no commit' to attempt to get trybots to run TBR=reed@google.com Review URL: https://codereview.chromium.org/1002693002
2015-03-24 14:28:17 +00:00
}
fix tiger b The tiger tests have uncovered numerous bugs. This CL fixes the last of them. If a pair of curves do not intersect, but have one or both ends very close to the opposite curve, consider that an intersection. TBR=reed@google.com BUG=skia:5131 GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2356363003 Review-Url: https://codereview.chromium.org/2356363003
2016-09-26 12:36:58 +00:00
REPORTER_ASSERT(reporter, !coin || intersections.used() >= 2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
double tt1, tt2;
SkDPoint xy1, xy2;
for (int pt3 = 0; pt3 < intersections.used(); ++pt3) {
tt1 = intersections[0][pt3];
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
xy1 = c1.ptAtT(tt1);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
tt2 = intersections[1][pt3];
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
xy2 = c2.ptAtT(tt2);
call newOneOff instead of oneOff calling the latter instead of the former walked off the end of the array causing mysterious bugs TODO: safeguard against NaNs in the input BUG= Review URL: https://codereview.chromium.org/19280002 git-svn-id: http://skia.googlecode.com/svn/trunk@10093 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-15 20:50:24 +00:00
const SkDPoint& iPt = intersections.pt(pt3);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
#if ONE_OFF_DEBUG
SkDebugf("%s t1=%1.9g (%1.9g, %1.9g) (%1.9g, %1.9g) (%1.9g, %1.9g) t2=%1.9g\n",
path ops -- rewrite angle sort This is a major change resulting from a minor tweak. In the old code, the intersection point of two curves was shared between them, but the intersection points and end points of sorted edges was computed directly from the intersection T value. In this CL, both intersection points and sorted points are the same, and intermediate control points are computed to preserve their slope. The sort itself has been completely rewritten to be more robust and remove 'magic' checks, conditions that empirically worked but couldn't be rationalized. This CL was triggered by errors generated computing the clips of SKP files. At this point, all 73M standard tests work and at least the first troublesome SKPs work. Review URL: https://codereview.chromium.org/15338003 git-svn-id: http://skia.googlecode.com/svn/trunk@9432 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-06-04 17:59:42 +00:00
__FUNCTION__, tt1, xy1.fX, xy1.fY, iPt.fX,
iPt.fY, xy2.fX, xy2.fY, tt2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
#endif
path ops -- rewrite angle sort This is a major change resulting from a minor tweak. In the old code, the intersection point of two curves was shared between them, but the intersection points and end points of sorted edges was computed directly from the intersection T value. In this CL, both intersection points and sorted points are the same, and intermediate control points are computed to preserve their slope. The sort itself has been completely rewritten to be more robust and remove 'magic' checks, conditions that empirically worked but couldn't be rationalized. This CL was triggered by errors generated computing the clips of SKP files. At this point, all 73M standard tests work and at least the first troublesome SKPs work. Review URL: https://codereview.chromium.org/15338003 git-svn-id: http://skia.googlecode.com/svn/trunk@9432 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-06-04 17:59:42 +00:00
REPORTER_ASSERT(reporter, xy1.approximatelyEqual(iPt));
REPORTER_ASSERT(reporter, xy2.approximatelyEqual(iPt));
REPORTER_ASSERT(reporter, xy1.approximatelyEqual(xy2));
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
reporter->bumpTestCount();
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
static void oneOff(skiatest::Reporter* reporter, int outer, int inner) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic1 = testSet[outer];
const CubicPts& cubic2 = testSet[inner];
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
oneOff(reporter, cubic1, cubic2, false);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
static void newOneOff(skiatest::Reporter* reporter, int outer, int inner) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic1 = newTestSet[outer];
const CubicPts& cubic2 = newTestSet[inner];
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
oneOff(reporter, cubic1, cubic2, false);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
Enabling clip stack flattening exercises path ops. Iterating through the 903K skps that represent the imagable 1M top web pages triggers a number of bugs, some of which are addressed here. Some web pages trigger intersecting cubic representations of arc with their conic counterparts. This exposed a flaw in coincident detection that caused an infinite loop. The loop alternatively extended the coincident section and, determining the that the bounds of the curve pairs did not overlap, deleted the extension. Track the number of times the coincident detection is called, and if it exceeds an empirically found limit, assume that the curves are coincident and force it to be so. The loop count limit can be determined by enabling DEBUG_T_SECT_LOOP_COUNT and running all tests. The largest count is reported on completion. Another class of bugs was caused by concident detection duplicating nearly identical points that had been merged earlier. To track these bugs, the 'handle coincidence' code was duplicated as a const debug variety that reported if one of a dozen or so irregularities are present; then it is easier to see when a block of code that fixes one irregularity regresses another. Creating the debug const code version exposed some non-debug code that could be const, and some that was experimental and could be removed. Set DEBUG_COINCIDENCE to track coincidence health and handling. For running on Chrome, DEBUG_VERIFY checks the result of pathops against the same operation using SkRegion to verify that the results are nearly the same. When visualizing the pathops work using tools/pathops_visualizer.htm, set DEBUG_DUMP_ALIGNMENT to see the curves after they've been aligned for coincidence. Other bugs fixed include detecting when a section of a pair of curves have devolved into lines and are coincident. TBR=reed@google.com Review URL: https://codereview.chromium.org/1394503003
2015-10-16 16:03:38 +00:00
static void testsOneOff(skiatest::Reporter* reporter, int index) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic1 = tests[index][0];
const CubicPts& cubic2 = tests[index][1];
Enabling clip stack flattening exercises path ops. Iterating through the 903K skps that represent the imagable 1M top web pages triggers a number of bugs, some of which are addressed here. Some web pages trigger intersecting cubic representations of arc with their conic counterparts. This exposed a flaw in coincident detection that caused an infinite loop. The loop alternatively extended the coincident section and, determining the that the bounds of the curve pairs did not overlap, deleted the extension. Track the number of times the coincident detection is called, and if it exceeds an empirically found limit, assume that the curves are coincident and force it to be so. The loop count limit can be determined by enabling DEBUG_T_SECT_LOOP_COUNT and running all tests. The largest count is reported on completion. Another class of bugs was caused by concident detection duplicating nearly identical points that had been merged earlier. To track these bugs, the 'handle coincidence' code was duplicated as a const debug variety that reported if one of a dozen or so irregularities are present; then it is easier to see when a block of code that fixes one irregularity regresses another. Creating the debug const code version exposed some non-debug code that could be const, and some that was experimental and could be removed. Set DEBUG_COINCIDENCE to track coincidence health and handling. For running on Chrome, DEBUG_VERIFY checks the result of pathops against the same operation using SkRegion to verify that the results are nearly the same. When visualizing the pathops work using tools/pathops_visualizer.htm, set DEBUG_DUMP_ALIGNMENT to see the curves after they've been aligned for coincidence. Other bugs fixed include detecting when a section of a pair of curves have devolved into lines and are coincident. TBR=reed@google.com Review URL: https://codereview.chromium.org/1394503003
2015-10-16 16:03:38 +00:00
oneOff(reporter, cubic1, cubic2, false);
}
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
static void oneOffTests(skiatest::Reporter* reporter) {
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
for (int outer = 0; outer < testSetCount - 1; ++outer) {
for (int inner = outer + 1; inner < testSetCount; ++inner) {
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
oneOff(reporter, outer, inner);
}
}
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
for (int outer = 0; outer < newTestSetCount - 1; ++outer) {
for (int inner = outer + 1; inner < newTestSetCount; ++inner) {
call newOneOff instead of oneOff calling the latter instead of the former walked off the end of the array causing mysterious bugs TODO: safeguard against NaNs in the input BUG= Review URL: https://codereview.chromium.org/19280002 git-svn-id: http://skia.googlecode.com/svn/trunk@10093 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-15 20:50:24 +00:00
newOneOff(reporter, outer, inner);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
}
}
#define DEBUG_CRASH 0
static void CubicIntersection_RandTest(skiatest::Reporter* reporter) {
srand(0);
const int tests = 10000000;
shape ops test fixes - fix rand for Android - build unit test on linux - use atomic inc in test count - add casting for Android git-svn-id: http://skia.googlecode.com/svn/trunk@8610 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-11 12:33:23 +00:00
#if !defined(SK_BUILD_FOR_WIN) && !defined(SK_BUILD_FOR_ANDROID)
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
unsigned seed = 0;
#endif
for (int test = 0; test < tests; ++test) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
CubicPts cubic1, cubic2;
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
for (int i = 0; i < 4; ++i) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
cubic1.fPts[i].fX = static_cast<double>(SK_RAND(seed)) / RAND_MAX * 100;
cubic1.fPts[i].fY = static_cast<double>(SK_RAND(seed)) / RAND_MAX * 100;
cubic2.fPts[i].fX = static_cast<double>(SK_RAND(seed)) / RAND_MAX * 100;
cubic2.fPts[i].fY = static_cast<double>(SK_RAND(seed)) / RAND_MAX * 100;
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
#if DEBUG_CRASH
char str[1024];
snprintf(str, sizeof(str),
"{{{%1.9g, %1.9g}, {%1.9g, %1.9g}, {%1.9g, %1.9g}, {%1.9g, %1.9g}}},\n"
"{{{%1.9g, %1.9g}, {%1.9g, %1.9g}, {%1.9g, %1.9g}, {%1.9g, %1.9g}}},\n",
cubic1[0].fX, cubic1[0].fY, cubic1[1].fX, cubic1[1].fY, cubic1[2].fX, cubic1[2].fY,
cubic1[3].fX, cubic1[3].fY,
cubic2[0].fX, cubic2[0].fY, cubic2[1].fX, cubic2[1].fY, cubic2[2].fX, cubic2[2].fY,
cubic2[3].fX, cubic2[3].fY);
#endif
SkDRect rect1, rect2;
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDCubic c1, c2;
c1.debugSet(cubic1.fPts);
c2.debugSet(cubic2.fPts);
rect1.setBounds(c1);
rect2.setBounds(c2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
bool boundsIntersect = rect1.fLeft <= rect2.fRight && rect2.fLeft <= rect2.fRight
&& rect1.fTop <= rect2.fBottom && rect2.fTop <= rect1.fBottom;
if (test == -1) {
SkDebugf("ready...\n");
}
SkIntersections intersections2;
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
int newIntersects = intersections2.intersect(c1, c2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
if (!boundsIntersect && newIntersects) {
#if DEBUG_CRASH
SkDebugf("%s %d unexpected intersection boundsIntersect=%d "
" newIntersects=%d\n%s %s\n", __FUNCTION__, test, boundsIntersect,
newIntersects, __FUNCTION__, str);
#endif
REPORTER_ASSERT(reporter, 0);
}
for (int pt = 0; pt < intersections2.used(); ++pt) {
double tt1 = intersections2[0][pt];
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDPoint xy1 = c1.ptAtT(tt1);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
double tt2 = intersections2[1][pt];
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDPoint xy2 = c2.ptAtT(tt2);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
REPORTER_ASSERT(reporter, xy1.approximatelyEqual(xy2));
}
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
reporter->bumpTestCount();
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
}
static void intersectionFinder(int index0, int index1, double t1Seed, double t2Seed,
double t1Step, double t2Step) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic1 = newTestSet[index0];
const CubicPts& cubic2 = newTestSet[index1];
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
SkDPoint t1[3], t2[3];
bool toggle = true;
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDCubic c1, c2;
c1.debugSet(cubic1.fPts);
c2.debugSet(cubic2.fPts);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
do {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
t1[0] = c1.ptAtT(t1Seed - t1Step);
t1[1] = c1.ptAtT(t1Seed);
t1[2] = c1.ptAtT(t1Seed + t1Step);
t2[0] = c2.ptAtT(t2Seed - t2Step);
t2[1] = c2.ptAtT(t2Seed);
t2[2] = c2.ptAtT(t2Seed + t2Step);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
double dist[3][3];
dist[1][1] = t1[1].distance(t2[1]);
int best_i = 1, best_j = 1;
for (int i = 0; i < 3; ++i) {
for (int j = 0; j < 3; ++j) {
if (i == 1 && j == 1) {
continue;
}
dist[i][j] = t1[i].distance(t2[j]);
if (dist[best_i][best_j] > dist[i][j]) {
best_i = i;
best_j = j;
}
}
}
if (best_i == 0) {
t1Seed -= t1Step;
} else if (best_i == 2) {
t1Seed += t1Step;
}
if (best_j == 0) {
t2Seed -= t2Step;
} else if (best_j == 2) {
t2Seed += t2Step;
}
if (best_i == 1 && best_j == 1) {
if ((toggle ^= true)) {
t1Step /= 2;
} else {
t2Step /= 2;
}
}
} while (!t1[1].approximatelyEqual(t2[1]));
t1Step = t2Step = 0.1;
double t10 = t1Seed - t1Step * 2;
double t12 = t1Seed + t1Step * 2;
double t20 = t2Seed - t2Step * 2;
double t22 = t2Seed + t2Step * 2;
SkDPoint test;
while (!approximately_zero(t1Step)) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
test = c1.ptAtT(t10);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
t10 += t1[1].approximatelyEqual(test) ? -t1Step : t1Step;
t1Step /= 2;
}
t1Step = 0.1;
while (!approximately_zero(t1Step)) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
test = c1.ptAtT(t12);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
t12 -= t1[1].approximatelyEqual(test) ? -t1Step : t1Step;
t1Step /= 2;
}
while (!approximately_zero(t2Step)) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
test = c2.ptAtT(t20);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
t20 += t2[1].approximatelyEqual(test) ? -t2Step : t2Step;
t2Step /= 2;
}
t2Step = 0.1;
while (!approximately_zero(t2Step)) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
test = c2.ptAtT(t22);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
t22 -= t2[1].approximatelyEqual(test) ? -t2Step : t2Step;
t2Step /= 2;
}
#if ONE_OFF_DEBUG
SkDebugf("%s t1=(%1.9g<%1.9g<%1.9g) t2=(%1.9g<%1.9g<%1.9g)\n", __FUNCTION__,
t10, t1Seed, t12, t20, t2Seed, t22);
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDPoint p10 = c1.ptAtT(t10);
SkDPoint p1Seed = c1.ptAtT(t1Seed);
SkDPoint p12 = c1.ptAtT(t12);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
SkDebugf("%s p1=(%1.9g,%1.9g)<(%1.9g,%1.9g)<(%1.9g,%1.9g)\n", __FUNCTION__,
p10.fX, p10.fY, p1Seed.fX, p1Seed.fY, p12.fX, p12.fY);
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
SkDPoint p20 = c2.ptAtT(t20);
SkDPoint p2Seed = c2.ptAtT(t2Seed);
SkDPoint p22 = c2.ptAtT(t22);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
SkDebugf("%s p2=(%1.9g,%1.9g)<(%1.9g,%1.9g)<(%1.9g,%1.9g)\n", __FUNCTION__,
p20.fX, p20.fY, p2Seed.fX, p2Seed.fY, p22.fX, p22.fY);
#endif
}
static void CubicIntersection_IntersectionFinder() {
// double t1Seed = 0.87;
// double t2Seed = 0.87;
double t1Step = 0.000001;
double t2Step = 0.000001;
intersectionFinder(0, 1, 0.855895664, 0.864850875, t1Step, t2Step);
intersectionFinder(0, 1, 0.865207906, 0.865207887, t1Step, t2Step);
intersectionFinder(0, 1, 0.865213351, 0.865208087, t1Step, t2Step);
}
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
static const CubicPts selfSet[] = {
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
{{{2, 3}, {0, 4}, {3, 2}, {5, 3}}},
{{{3, 6}, {2, 3}, {4, 0}, {3, 2}}},
{{{0, 2}, {2, 3}, {5, 1}, {3, 2}}},
{{{0, 2}, {3, 5}, {5, 0}, {4, 2}}},
{{{3.34, 8.98}, {1.95, 10.27}, {3.76, 7.65}, {4.96, 10.64}}},
{{{3.13, 2.74}, {1.08, 4.62}, {3.71, 0.94}, {2.01, 3.81}}},
{{{6.71, 3.14}, {7.99, 2.75}, {8.27, 1.96}, {6.35, 3.57}}},
{{{12.81, 7.27}, {7.22, 6.98}, {12.49, 8.97}, {11.42, 6.18}}},
};
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
int selfSetCount = (int) SK_ARRAY_COUNT(selfSet);
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
static void selfOneOff(skiatest::Reporter* reporter, int index) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic = selfSet[index];
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
SkPoint c[4];
for (int i = 0; i < 4; ++i) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
c[i] = cubic.fPts[i].asSkPoint();
Revert of pathops version two (patchset #16 id:150001 of https://codereview.chromium.org/1002693002/) Reason for revert: ASAN investigation Original issue's description: > pathops version two > > R=reed@google.com > > marked 'no commit' to attempt to get trybots to run > > TBR=reed@google.com > > Committed: https://skia.googlesource.com/skia/+/ccec0f958ffc71a9986d236bc2eb335cb2111119 TBR=caryclark@google.com NOPRESUBMIT=true NOTREECHECKS=true NOTRY=true Review URL: https://codereview.chromium.org/1029993002
2015-03-24 20:55:33 +00:00
}
simplify bug The path contains a cubic with a very tight curve. Split the cubic into pieces so that the individual curves are better behaved. Use both inflections and max curvature to potentially split cubics. Since this may require a bit of work, preflight to ignore cubics that monotonically change in x and y. Only one of the three tests referred to by the bug below repro'd. Use path.dumpHex() instead of path.dump() to capture the crashing data. TBR=reed@google.com BUG=skia:6041 Change-Id: I29a264f87242cacc7c421e7685b90aca81621c74 Reviewed-on: https://skia-review.googlesource.com/5702 Reviewed-by: Cary Clark <caryclark@google.com> Commit-Queue: Cary Clark <caryclark@google.com>
2016-12-08 19:36:32 +00:00
SkScalar loopT[3];
Improve cubic KLM accuracy Moves cubic root finding logic out of GrPathUtils and PathOpsCubicIntersectionTest, and unifies it in SkGeometry. "Normalizes" the homogeneous parameter values of the roots, rather than the cubic inflection function. Does this normalization by twiddling the exponents instead of division (which causes a loss of precision). Abandons the built-in derivatives in GrCubicEffect. These don't have high enough precision on many mobile gpus. Instead we pass the KLM matrix to the vertex shader via uniform, where we can use it to set up new linear functionals from which the fragment shader can calculate the gradient of the implicit function. Bug: skia:4410 Change-Id: Ibd64e999520adc8cdef7803a492d3699995aef5a Reviewed-on: https://skia-review.googlesource.com/19017 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Chris Dalton <csmartdalton@google.com>
2017-06-08 19:12:02 +00:00
SkCubicType cubicType = SkClassifyCubic(c);
simplify bug The path contains a cubic with a very tight curve. Split the cubic into pieces so that the individual curves are better behaved. Use both inflections and max curvature to potentially split cubics. Since this may require a bit of work, preflight to ignore cubics that monotonically change in x and y. Only one of the three tests referred to by the bug below repro'd. Use path.dumpHex() instead of path.dump() to capture the crashing data. TBR=reed@google.com BUG=skia:6041 Change-Id: I29a264f87242cacc7c421e7685b90aca81621c74 Reviewed-on: https://skia-review.googlesource.com/5702 Reviewed-by: Cary Clark <caryclark@google.com> Commit-Queue: Cary Clark <caryclark@google.com>
2016-12-08 19:36:32 +00:00
int breaks = SkDCubic::ComplexBreak(c, loopT);
SkASSERT(breaks < 2);
Improve accuracy of cubic classification - Updates the logic to reflect the Loop-Blinn paper instead of the GPU gems website. - Removes the threshold for detecting local cusps. The serpentine codepath works for these cusps anyway, so what we really want to know is whether the discriminant is negative. - Makes sure to not scale the inflection function by 1/0. - Shifts the inflection function coefficients in d[] so they match the paper. - Stores the cubic discriminant in d[0]. Bug: skia: Change-Id: I909a522a0fd27c9c8dfbc27d968bc43eeb7a416f Reviewed-on: https://skia-review.googlesource.com/13304 Reviewed-by: Greg Daniel <egdaniel@google.com> Commit-Queue: Chris Dalton <csmartdalton@google.com>
2017-04-13 20:26:00 +00:00
if (breaks && cubicType == SkCubicType::kLoop) {
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
SkIntersections i;
SkPoint twoCubics[7];
simplify bug The path contains a cubic with a very tight curve. Split the cubic into pieces so that the individual curves are better behaved. Use both inflections and max curvature to potentially split cubics. Since this may require a bit of work, preflight to ignore cubics that monotonically change in x and y. Only one of the three tests referred to by the bug below repro'd. Use path.dumpHex() instead of path.dump() to capture the crashing data. TBR=reed@google.com BUG=skia:6041 Change-Id: I29a264f87242cacc7c421e7685b90aca81621c74 Reviewed-on: https://skia-review.googlesource.com/5702 Reviewed-by: Cary Clark <caryclark@google.com> Commit-Queue: Cary Clark <caryclark@google.com>
2016-12-08 19:36:32 +00:00
SkChopCubicAt(c, twoCubics, loopT[0]);
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
SkDCubic chopped[2];
chopped[0].set(&twoCubics[0]);
chopped[1].set(&twoCubics[3]);
int result = i.intersect(chopped[0], chopped[1]);
REPORTER_ASSERT(reporter, result == 2);
REPORTER_ASSERT(reporter, i.used() == 2);
for (int index = 0; index < result; ++index) {
SkDPoint pt1 = chopped[0].ptAtT(i[0][index]);
SkDPoint pt2 = chopped[1].ptAtT(i[1][index]);
REPORTER_ASSERT(reporter, pt1.approximatelyEqual(pt2));
reporter->bumpTestCount();
}
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
}
}
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
static void cubicIntersectionSelfTest(skiatest::Reporter* reporter) {
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
int firstFail = 0;
for (int index = firstFail; index < selfSetCount; ++index) {
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
selfOneOff(reporter, index);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
}
}
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
static const CubicPts coinSet[] = {
Enabling clip stack flattening exercises path ops. Iterating through the 903K skps that represent the imagable 1M top web pages triggers a number of bugs, some of which are addressed here. Some web pages trigger intersecting cubic representations of arc with their conic counterparts. This exposed a flaw in coincident detection that caused an infinite loop. The loop alternatively extended the coincident section and, determining the that the bounds of the curve pairs did not overlap, deleted the extension. Track the number of times the coincident detection is called, and if it exceeds an empirically found limit, assume that the curves are coincident and force it to be so. The loop count limit can be determined by enabling DEBUG_T_SECT_LOOP_COUNT and running all tests. The largest count is reported on completion. Another class of bugs was caused by concident detection duplicating nearly identical points that had been merged earlier. To track these bugs, the 'handle coincidence' code was duplicated as a const debug variety that reported if one of a dozen or so irregularities are present; then it is easier to see when a block of code that fixes one irregularity regresses another. Creating the debug const code version exposed some non-debug code that could be const, and some that was experimental and could be removed. Set DEBUG_COINCIDENCE to track coincidence health and handling. For running on Chrome, DEBUG_VERIFY checks the result of pathops against the same operation using SkRegion to verify that the results are nearly the same. When visualizing the pathops work using tools/pathops_visualizer.htm, set DEBUG_DUMP_ALIGNMENT to see the curves after they've been aligned for coincidence. Other bugs fixed include detecting when a section of a pair of curves have devolved into lines and are coincident. TBR=reed@google.com Review URL: https://codereview.chromium.org/1394503003
2015-10-16 16:03:38 +00:00
{{{72.350448608398438, 27.966041564941406}, {72.58441162109375, 27.861515045166016},
{72.818222045898437, 27.756658554077148}, {73.394996643066406, 27.49799919128418}}},
{{{73.394996643066406, 27.49799919128418}, {72.818222045898437, 27.756658554077148},
{72.58441162109375, 27.861515045166016}, {72.350448608398438, 27.966041564941406}}},
fix pathops fuzz failures If a curve has the identical start and control points, the initial or final tangent can't be trivally determined. The perpendicular to the tangent is used to measure coincidence. Add logic for cubics, quadratics, and conics, to use the secondary control points or the end points if the initial control point alone can't determine the tangent. Add debugging (currently untriggered by exhaustive testing) to detect zero-length tangents which are not at the curve endpoints. Increase the number of temporary intersecions gathered from 10 to 12 but reduce the max passed in by cubic intersection from 27 to 12. Also, add checks if the max passed exceeds the storage allocated. When cleaning up parallel lines, choose the intersection which is on the end of both segments over the intersection which is on the end of a single segment. TBR=reed@google.com BUG=425140,516266 Review URL: https://codereview.chromium.org/1288863004
2015-08-18 14:12:43 +00:00
{{{297.04998779296875, 43.928997039794922}, {297.04998779296875, 43.928997039794922},
{300.69699096679688, 45.391998291015625}, {306.92498779296875, 43.08599853515625}}},
{{{297.04998779296875, 43.928997039794922}, {297.04998779296875, 43.928997039794922},
{300.69699096679688, 45.391998291015625}, {306.92498779296875, 43.08599853515625}}},
cumulative pathops patch Replace the implicit curve intersection with a geometric curve intersection. The implicit intersection proved mathematically unstable and took a long time to zero in on an answer. Use pointers instead of indices to refer to parts of curves. Indices required awkward renumbering. Unify t and point values so that small intervals can be eliminated in one pass. Break cubics up front to eliminate loops and cusps. Make the Simplify and Op code more regular and eliminate arbitrary differences. Add a builder that takes an array of paths and operators. Delete unused code. BUG=skia:3588 R=reed@google.com Review URL: https://codereview.chromium.org/1037573004
2015-03-26 14:52:43 +00:00
{{{2, 3}, {0, 4}, {3, 2}, {5, 3}}},
{{{2, 3}, {0, 4}, {3, 2}, {5, 3}}},
Sanitizing source files in Housekeeper-Nightly git-svn-id: http://skia.googlecode.com/svn/trunk@11585 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-03 07:01:37 +00:00
{{{317, 711}, {322.52285766601562, 711}, {327, 715.4771728515625}, {327, 721}}},
{{{324.07107543945312, 713.928955078125}, {324.4051513671875, 714.26300048828125},
{324.71566772460937, 714.62060546875}, {325, 714.9990234375}}},
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
};
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
static int coinSetCount = (int) SK_ARRAY_COUNT(coinSet);
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
static void coinOneOff(skiatest::Reporter* reporter, int index) {
fix fuzzers Many old pathops-related fuzz failures have built up while the codebase was under a state a flux. Now that the code is stable, address these failures. Most of the CL plumbs the debug global state to downstream routines so that, if the data is not trusted (ala fuzzed) the function can safely exit without asserting. TBR=reed@google.com GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2426173002 Review-Url: https://chromiumcodereview.appspot.com/2426173002
2016-10-20 15:32:18 +00:00
const CubicPts& cubic1 = coinSet[index];
const CubicPts& cubic2 = coinSet[index + 1];
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
oneOff(reporter, cubic1, cubic2, true);
}
static void cubicIntersectionCoinTest(skiatest::Reporter* reporter) {
Mike R: please sanity check SkPostConfig.h Mike K: please sanity check Test.cpp and skia_test.cpp Feel free to look at the rest, but I don't expect any in depth review of path ops innards. Path Ops first iteration used QuickSort to order segments radiating from an intersection to compute the winding rule. This revision uses a circular sort instead. Breaking out the circular sort into its own long-lived structure (SkOpAngle) allows doing less work and provides a home for caching additional sorting data. The circle sort is more stable than the former sort, has a robust ordering and fewer exceptions. It finds unsortable ordering less often. It is less reliant on the initial curve tangent, using convex hulls instead whenever it can. Additional debug validation makes sure that the computed structures are self-consistent. A new visualization tool helps verify that the angle ordering is correct. The 70+M tests pass with this change on Windows, Mac, Linux 32 and Linux 64 in debug and release. R=mtklein@google.com, reed@google.com Author: caryclark@google.com Review URL: https://codereview.chromium.org/131103009 git-svn-id: http://skia.googlecode.com/svn/trunk@14183 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-04-14 17:08:59 +00:00
int firstFail = 0;
for (int index = firstFail; index < coinSetCount; index += 2) {
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
coinOneOff(reporter, index);
}
}
Get rid of DEFINE_TESTCLASS_SHORT() macro. Instead tests should be written using DEF_TEST() macro, which is much nicer and simplifies the process of setting up an unit test. BUG=None TEST=skpskgr_test, pathops_unittest R=mtklein@google.com Review URL: https://codereview.chromium.org/117863005 git-svn-id: http://skia.googlecode.com/svn/trunk@12870 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-01-02 21:45:03 +00:00
DEF_TEST(PathOpsCubicCoinOneOff, reporter) {
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
coinOneOff(reporter, 0);
}
Get rid of DEFINE_TESTCLASS_SHORT() macro. Instead tests should be written using DEF_TEST() macro, which is much nicer and simplifies the process of setting up an unit test. BUG=None TEST=skpskgr_test, pathops_unittest R=mtklein@google.com Review URL: https://codereview.chromium.org/117863005 git-svn-id: http://skia.googlecode.com/svn/trunk@12870 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-01-02 21:45:03 +00:00
DEF_TEST(PathOpsCubicIntersectionOneOff, reporter) {
split tight quads and conics Tight quads and conics may nearly fold over on themselves, confusing coincidence against other curves. Split them at their max curvature early on to avoid complicating later logic. TBR=reed@google.com BUG=skia:5131 GOLD_TRYBOT_URL= https://gold.skia.org/search?issue=2357353002 Review-Url: https://codereview.chromium.org/2357353002
2016-09-23 12:47:20 +00:00
newOneOff(reporter, 0, 1);
Enabling clip stack flattening exercises path ops. Iterating through the 903K skps that represent the imagable 1M top web pages triggers a number of bugs, some of which are addressed here. Some web pages trigger intersecting cubic representations of arc with their conic counterparts. This exposed a flaw in coincident detection that caused an infinite loop. The loop alternatively extended the coincident section and, determining the that the bounds of the curve pairs did not overlap, deleted the extension. Track the number of times the coincident detection is called, and if it exceeds an empirically found limit, assume that the curves are coincident and force it to be so. The loop count limit can be determined by enabling DEBUG_T_SECT_LOOP_COUNT and running all tests. The largest count is reported on completion. Another class of bugs was caused by concident detection duplicating nearly identical points that had been merged earlier. To track these bugs, the 'handle coincidence' code was duplicated as a const debug variety that reported if one of a dozen or so irregularities are present; then it is easier to see when a block of code that fixes one irregularity regresses another. Creating the debug const code version exposed some non-debug code that could be const, and some that was experimental and could be removed. Set DEBUG_COINCIDENCE to track coincidence health and handling. For running on Chrome, DEBUG_VERIFY checks the result of pathops against the same operation using SkRegion to verify that the results are nearly the same. When visualizing the pathops work using tools/pathops_visualizer.htm, set DEBUG_DUMP_ALIGNMENT to see the curves after they've been aligned for coincidence. Other bugs fixed include detecting when a section of a pair of curves have devolved into lines and are coincident. TBR=reed@google.com Review URL: https://codereview.chromium.org/1394503003
2015-10-16 16:03:38 +00:00
}
DEF_TEST(PathOpsCubicIntersectionTestsOneOff, reporter) {
testsOneOff(reporter, 10);
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
}
Get rid of DEFINE_TESTCLASS_SHORT() macro. Instead tests should be written using DEF_TEST() macro, which is much nicer and simplifies the process of setting up an unit test. BUG=None TEST=skpskgr_test, pathops_unittest R=mtklein@google.com Review URL: https://codereview.chromium.org/117863005 git-svn-id: http://skia.googlecode.com/svn/trunk@12870 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-01-02 21:45:03 +00:00
DEF_TEST(PathOpsCubicSelfOneOff, reporter) {
turn off debugging printfs fix pathops issues 1417, 1418 be more rigorous about pulling intersections of lines to end points rewrite cubic/line and quad/line intersections to share style BUG= Review URL: https://codereview.chromium.org/19543005 git-svn-id: http://skia.googlecode.com/svn/trunk@10270 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-23 15:27:41 +00:00
selfOneOff(reporter, 0);
path ops near exact Modify line intersections to first - match exact ends - compute intersections - match near ends where the exact ends are preferred, then near matches, then computed matches. This pulls matches towards existing end points when possible, and keeps intersection distances consistent with different line/line line/quad and line/cubic computations. BUG= Review URL: https://codereview.chromium.org/19183003 git-svn-id: http://skia.googlecode.com/svn/trunk@10073 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-07-15 13:29:13 +00:00
}
Get rid of DEFINE_TESTCLASS_SHORT() macro. Instead tests should be written using DEF_TEST() macro, which is much nicer and simplifies the process of setting up an unit test. BUG=None TEST=skpskgr_test, pathops_unittest R=mtklein@google.com Review URL: https://codereview.chromium.org/117863005 git-svn-id: http://skia.googlecode.com/svn/trunk@12870 2bbb7eff-a529-9590-31e7-b0007b416f81
2014-01-02 21:45:03 +00:00
DEF_TEST(PathOpsCubicIntersection, reporter) {
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
oneOffTests(reporter);
cubicIntersectionSelfTest(reporter);
path ops work in progress make more skps work remove edit files BUG= Review URL: https://codereview.chromium.org/23542056 git-svn-id: http://skia.googlecode.com/svn/trunk@11570 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-10-02 14:49:34 +00:00
cubicIntersectionCoinTest(reporter);
Add intersections for path ops This CL depends on https://codereview.chromium.org/12827020/ "Add base types for path ops" The intersection of a line, quadratic, or cubic with another curve (or with itself) is found by solving the implicit equation for the curve pair. The curves are first reduced to find the simplest form that will describe the original, and to detect degenerate or special-case data like horizontal and vertical lines. For cubic self-intersection, and for a pair of cubics, the intersection is found by recursively approximating the cubic with a series of quadratics. The implicit solutions depend on the root finding contained in the DCubic and DQuad structs, and the quartic root finder included here. Review URL: https://codereview.chromium.org/12880016 git-svn-id: http://skia.googlecode.com/svn/trunk@8552 2bbb7eff-a529-9590-31e7-b0007b416f81
2013-04-08 11:50:00 +00:00
standardTestCases(reporter);
if (false) CubicIntersection_IntersectionFinder();
if (false) CubicIntersection_RandTest(reporter);
}
Reference in New Issue Copy Permalink