gecko-dev/dom/base/DOMMatrix.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

1028 строки
32 KiB
C++
Исходник Обычный вид История

/* -*- Mode: C++; tab-width: 8; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* vim: set ts=8 sts=2 et sw=2 tw=80: */
/* This Source Code Form is subject to the terms of the Mozilla Public
* License, v. 2.0. If a copy of the MPL was not distributed with this
* file, You can obtain one at http://mozilla.org/MPL/2.0/. */
#include "mozilla/dom/DOMMatrix.h"
#include <cmath>
#include <cstdint>
#include <new>
#include "ErrorList.h"
#include "js/Conversions.h"
#include "js/Equality.h"
#include "js/StructuredClone.h"
#include "js/Value.h"
#include "mozilla/Casting.h"
#include "mozilla/ErrorResult.h"
#include "mozilla/FloatingPoint.h"
#include "mozilla/MacroForEach.h"
#include "mozilla/RefPtr.h"
#include "mozilla/ServoCSSParser.h"
#include "mozilla/dom/BindingDeclarations.h"
#include "mozilla/dom/DOMMatrixBinding.h"
#include "mozilla/dom/DOMPoint.h"
#include "mozilla/dom/DOMPointBinding.h"
#include "mozilla/dom/ToJSValue.h"
#include "mozilla/gfx/BasePoint.h"
#include "mozilla/gfx/BasePoint4D.h"
#include "mozilla/gfx/Point.h"
#include "nsIGlobalObject.h"
#include "nsPIDOMWindow.h"
#include "nsString.h"
#include "nsStringFlags.h"
#include "nsTArray.h"
#include "nsTLiteralString.h"
namespace mozilla::dom {
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
template <typename T>
static void SetDataInMatrix(DOMMatrixReadOnly* aMatrix, const T* aData,
int aLength, ErrorResult& aRv);
static const double radPerDegree = 2.0 * M_PI / 360.0;
NS_IMPL_CYCLE_COLLECTION_WRAPPERCACHE(DOMMatrixReadOnly, mParent)
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
JSObject* DOMMatrixReadOnly::WrapObject(JSContext* aCx,
JS::Handle<JSObject*> aGivenProto) {
return DOMMatrixReadOnly_Binding::Wrap(aCx, this, aGivenProto);
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
}
// https://drafts.fxtf.org/geometry/#matrix-validate-and-fixup-2d
static bool ValidateAndFixupMatrix2DInit(DOMMatrix2DInit& aMatrixInit,
ErrorResult& aRv) {
#define ValidateAliases(field, alias, fieldName, aliasName) \
if ((field).WasPassed() && (alias).WasPassed() && \
!JS::SameValueZero((field).Value(), (alias).Value())) { \
aRv.ThrowTypeError<MSG_MATRIX_INIT_CONFLICTING_VALUE>((fieldName), \
(aliasName)); \
return false; \
}
#define SetFromAliasOrDefault(field, alias, defaultValue) \
if (!(field).WasPassed()) { \
if ((alias).WasPassed()) { \
(field).Construct((alias).Value()); \
} else { \
(field).Construct(defaultValue); \
} \
}
#define ValidateAndSet(field, alias, fieldName, aliasName, defaultValue) \
ValidateAliases((field), (alias), fieldName, aliasName); \
SetFromAliasOrDefault((field), (alias), (defaultValue));
ValidateAndSet(aMatrixInit.mM11, aMatrixInit.mA, "m11", "a", 1);
ValidateAndSet(aMatrixInit.mM12, aMatrixInit.mB, "m12", "b", 0);
ValidateAndSet(aMatrixInit.mM21, aMatrixInit.mC, "m21", "c", 0);
ValidateAndSet(aMatrixInit.mM22, aMatrixInit.mD, "m22", "d", 1);
ValidateAndSet(aMatrixInit.mM41, aMatrixInit.mE, "m41", "e", 0);
ValidateAndSet(aMatrixInit.mM42, aMatrixInit.mF, "m42", "f", 0);
return true;
#undef ValidateAliases
#undef SetFromAliasOrDefault
#undef ValidateAndSet
}
// https://drafts.fxtf.org/geometry/#matrix-validate-and-fixup
static bool ValidateAndFixupMatrixInit(DOMMatrixInit& aMatrixInit,
ErrorResult& aRv) {
#define Check3DField(field, fieldName, defaultValue) \
if ((field) != (defaultValue)) { \
if (!aMatrixInit.mIs2D.WasPassed()) { \
aMatrixInit.mIs2D.Construct(false); \
return true; \
} \
if (aMatrixInit.mIs2D.Value()) { \
aRv.ThrowTypeError<MSG_MATRIX_INIT_EXCEEDS_2D>(fieldName); \
return false; \
} \
}
if (!ValidateAndFixupMatrix2DInit(aMatrixInit, aRv)) {
return false;
}
Check3DField(aMatrixInit.mM13, "m13", 0);
Check3DField(aMatrixInit.mM14, "m14", 0);
Check3DField(aMatrixInit.mM23, "m23", 0);
Check3DField(aMatrixInit.mM24, "m24", 0);
Check3DField(aMatrixInit.mM31, "m31", 0);
Check3DField(aMatrixInit.mM32, "m32", 0);
Check3DField(aMatrixInit.mM34, "m34", 0);
Check3DField(aMatrixInit.mM43, "m43", 0);
Check3DField(aMatrixInit.mM33, "m33", 1);
Check3DField(aMatrixInit.mM44, "m44", 1);
if (!aMatrixInit.mIs2D.WasPassed()) {
aMatrixInit.mIs2D.Construct(true);
}
return true;
#undef Check3DField
}
void DOMMatrixReadOnly::SetDataFromMatrix2DInit(
const DOMMatrix2DInit& aMatrixInit) {
MOZ_ASSERT(Is2D());
mMatrix2D->_11 = aMatrixInit.mM11.Value();
mMatrix2D->_12 = aMatrixInit.mM12.Value();
mMatrix2D->_21 = aMatrixInit.mM21.Value();
mMatrix2D->_22 = aMatrixInit.mM22.Value();
mMatrix2D->_31 = aMatrixInit.mM41.Value();
mMatrix2D->_32 = aMatrixInit.mM42.Value();
}
void DOMMatrixReadOnly::SetDataFromMatrixInit(
const DOMMatrixInit& aMatrixInit) {
const bool is2D = aMatrixInit.mIs2D.Value();
MOZ_ASSERT(is2D == Is2D());
if (is2D) {
SetDataFromMatrix2DInit(aMatrixInit);
} else {
mMatrix3D->_11 = aMatrixInit.mM11.Value();
mMatrix3D->_12 = aMatrixInit.mM12.Value();
mMatrix3D->_13 = aMatrixInit.mM13;
mMatrix3D->_14 = aMatrixInit.mM14;
mMatrix3D->_21 = aMatrixInit.mM21.Value();
mMatrix3D->_22 = aMatrixInit.mM22.Value();
mMatrix3D->_23 = aMatrixInit.mM23;
mMatrix3D->_24 = aMatrixInit.mM24;
mMatrix3D->_31 = aMatrixInit.mM31;
mMatrix3D->_32 = aMatrixInit.mM32;
mMatrix3D->_33 = aMatrixInit.mM33;
mMatrix3D->_34 = aMatrixInit.mM34;
mMatrix3D->_41 = aMatrixInit.mM41.Value();
mMatrix3D->_42 = aMatrixInit.mM42.Value();
mMatrix3D->_43 = aMatrixInit.mM43;
mMatrix3D->_44 = aMatrixInit.mM44;
}
}
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::FromMatrix(
nsISupports* aParent, const DOMMatrix2DInit& aMatrixInit,
ErrorResult& aRv) {
DOMMatrix2DInit matrixInit(aMatrixInit);
if (!ValidateAndFixupMatrix2DInit(matrixInit, aRv)) {
return nullptr;
};
RefPtr<DOMMatrixReadOnly> matrix =
new DOMMatrixReadOnly(aParent, /* is2D */ true);
matrix->SetDataFromMatrix2DInit(matrixInit);
return matrix.forget();
}
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::FromMatrix(
nsISupports* aParent, const DOMMatrixInit& aMatrixInit, ErrorResult& aRv) {
DOMMatrixInit matrixInit(aMatrixInit);
if (!ValidateAndFixupMatrixInit(matrixInit, aRv)) {
return nullptr;
};
RefPtr<DOMMatrixReadOnly> rval =
new DOMMatrixReadOnly(aParent, matrixInit.mIs2D.Value());
rval->SetDataFromMatrixInit(matrixInit);
return rval.forget();
}
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::FromMatrix(
const GlobalObject& aGlobal, const DOMMatrixInit& aMatrixInit,
ErrorResult& aRv) {
RefPtr<DOMMatrixReadOnly> matrix =
FromMatrix(aGlobal.GetAsSupports(), aMatrixInit, aRv);
return matrix.forget();
}
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::FromFloat32Array(
const GlobalObject& aGlobal, const Float32Array& aArray32,
ErrorResult& aRv) {
aArray32.ComputeState();
const int length = aArray32.Length();
const bool is2D = length == 6;
RefPtr<DOMMatrixReadOnly> obj =
new DOMMatrixReadOnly(aGlobal.GetAsSupports(), is2D);
SetDataInMatrix(obj, aArray32.Data(), length, aRv);
return obj.forget();
}
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::FromFloat64Array(
const GlobalObject& aGlobal, const Float64Array& aArray64,
ErrorResult& aRv) {
aArray64.ComputeState();
const int length = aArray64.Length();
const bool is2D = length == 6;
RefPtr<DOMMatrixReadOnly> obj =
new DOMMatrixReadOnly(aGlobal.GetAsSupports(), is2D);
SetDataInMatrix(obj, aArray64.Data(), length, aRv);
return obj.forget();
}
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::Constructor(
const GlobalObject& aGlobal,
const Optional<UTF8StringOrUnrestrictedDoubleSequenceOrDOMMatrixReadOnly>&
aArg,
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
ErrorResult& aRv) {
if (!aArg.WasPassed()) {
RefPtr<DOMMatrixReadOnly> rval =
new DOMMatrixReadOnly(aGlobal.GetAsSupports());
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
return rval.forget();
}
const auto& arg = aArg.Value();
if (arg.IsUTF8String()) {
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
nsCOMPtr<nsPIDOMWindowInner> win =
do_QueryInterface(aGlobal.GetAsSupports());
if (!win) {
aRv.ThrowTypeError<MSG_ILLEGAL_CONSTRUCTOR>();
return nullptr;
}
RefPtr<DOMMatrixReadOnly> rval =
new DOMMatrixReadOnly(aGlobal.GetAsSupports());
rval->SetMatrixValue(arg.GetAsUTF8String(), aRv);
return rval.forget();
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
}
if (arg.IsDOMMatrixReadOnly()) {
RefPtr<DOMMatrixReadOnly> obj = new DOMMatrixReadOnly(
aGlobal.GetAsSupports(), arg.GetAsDOMMatrixReadOnly());
return obj.forget();
}
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
const auto& sequence = arg.GetAsUnrestrictedDoubleSequence();
const int length = sequence.Length();
const bool is2D = length == 6;
RefPtr<DOMMatrixReadOnly> rval =
new DOMMatrixReadOnly(aGlobal.GetAsSupports(), is2D);
SetDataInMatrix(rval, sequence.Elements(), length, aRv);
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
return rval.forget();
}
already_AddRefed<DOMMatrixReadOnly> DOMMatrixReadOnly::ReadStructuredClone(
JSContext* aCx, nsIGlobalObject* aGlobal,
JSStructuredCloneReader* aReader) {
uint8_t is2D;
if (!JS_ReadBytes(aReader, &is2D, 1)) {
return nullptr;
}
RefPtr<DOMMatrixReadOnly> rval = new DOMMatrixReadOnly(aGlobal, is2D);
if (!ReadStructuredCloneElements(aReader, rval)) {
return nullptr;
};
return rval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::Translate(double aTx, double aTy,
double aTz) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->TranslateSelf(aTx, aTy, aTz);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::Scale(
double aScaleX, const Optional<double>& aScaleY, double aScaleZ,
double aOriginX, double aOriginY, double aOriginZ) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->ScaleSelf(aScaleX, aScaleY, aScaleZ, aOriginX, aOriginY, aOriginZ);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::Scale3d(double aScale,
double aOriginX,
double aOriginY,
double aOriginZ) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->Scale3dSelf(aScale, aOriginX, aOriginY, aOriginZ);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::ScaleNonUniform(
double aScaleX, double aScaleY) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->ScaleSelf(aScaleX, Optional<double>(aScaleY), 1, 0, 0, 0);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::Rotate(
double aRotX, const Optional<double>& aRotY,
const Optional<double>& aRotZ) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->RotateSelf(aRotX, aRotY, aRotZ);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::RotateFromVector(
double x, double y) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->RotateFromVectorSelf(x, y);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::RotateAxisAngle(
double aX, double aY, double aZ, double aAngle) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->RotateAxisAngleSelf(aX, aY, aZ, aAngle);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::SkewX(double aSx) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->SkewXSelf(aSx);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::SkewY(double aSy) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->SkewYSelf(aSy);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::Multiply(
const DOMMatrixInit& other, ErrorResult& aRv) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->MultiplySelf(other, aRv);
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::FlipX() const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
if (mMatrix3D) {
gfx::Matrix4x4Double m;
m._11 = -1;
retval->mMatrix3D = MakeUnique<gfx::Matrix4x4Double>(m * *mMatrix3D);
} else {
gfx::MatrixDouble m;
m._11 = -1;
retval->mMatrix2D =
MakeUnique<gfx::MatrixDouble>(mMatrix2D ? m * *mMatrix2D : m);
}
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::FlipY() const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
if (mMatrix3D) {
gfx::Matrix4x4Double m;
m._22 = -1;
retval->mMatrix3D = MakeUnique<gfx::Matrix4x4Double>(m * *mMatrix3D);
} else {
gfx::MatrixDouble m;
m._22 = -1;
retval->mMatrix2D =
MakeUnique<gfx::MatrixDouble>(mMatrix2D ? m * *mMatrix2D : m);
}
return retval.forget();
}
already_AddRefed<DOMMatrix> DOMMatrixReadOnly::Inverse() const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMMatrix> retval = new DOMMatrix(mParent, *this);
retval->InvertSelf();
return retval.forget();
}
bool DOMMatrixReadOnly::Is2D() const { return !mMatrix3D; }
bool DOMMatrixReadOnly::IsIdentity() const {
if (mMatrix3D) {
return mMatrix3D->IsIdentity();
}
return mMatrix2D->IsIdentity();
}
already_AddRefed<DOMPoint> DOMMatrixReadOnly::TransformPoint(
const DOMPointInit& point) const {
Bug 1207245 - part 6 - rename nsRefPtr<T> to RefPtr<T>; r=ehsan; a=Tomcat The bulk of this commit was generated with a script, executed at the top level of a typical source code checkout. The only non-machine-generated part was modifying MFBT's moz.build to reflect the new naming. CLOSED TREE makes big refactorings like this a piece of cake. # The main substitution. find . -name '*.cpp' -o -name '*.cc' -o -name '*.h' -o -name '*.mm' -o -name '*.idl'| \ xargs perl -p -i -e ' s/nsRefPtr\.h/RefPtr\.h/g; # handle includes s/nsRefPtr ?</RefPtr</g; # handle declarations and variables ' # Handle a special friend declaration in gfx/layers/AtomicRefCountedWithFinalize.h. perl -p -i -e 's/::nsRefPtr;/::RefPtr;/' gfx/layers/AtomicRefCountedWithFinalize.h # Handle nsRefPtr.h itself, a couple places that define constructors # from nsRefPtr, and code generators specially. We do this here, rather # than indiscriminantly s/nsRefPtr/RefPtr/, because that would rename # things like nsRefPtrHashtable. perl -p -i -e 's/nsRefPtr/RefPtr/g' \ mfbt/nsRefPtr.h \ xpcom/glue/nsCOMPtr.h \ xpcom/base/OwningNonNull.h \ ipc/ipdl/ipdl/lower.py \ ipc/ipdl/ipdl/builtin.py \ dom/bindings/Codegen.py \ python/lldbutils/lldbutils/utils.py # In our indiscriminate substitution above, we renamed # nsRefPtrGetterAddRefs, the class behind getter_AddRefs. Fix that up. find . -name '*.cpp' -o -name '*.h' -o -name '*.idl' | \ xargs perl -p -i -e 's/nsRefPtrGetterAddRefs/RefPtrGetterAddRefs/g' if [ -d .git ]; then git mv mfbt/nsRefPtr.h mfbt/RefPtr.h else hg mv mfbt/nsRefPtr.h mfbt/RefPtr.h fi --HG-- rename : mfbt/nsRefPtr.h => mfbt/RefPtr.h
2015-10-18 08:24:48 +03:00
RefPtr<DOMPoint> retval = new DOMPoint(mParent);
if (mMatrix3D) {
gfx::Point4D transformedPoint;
transformedPoint.x = point.mX;
transformedPoint.y = point.mY;
transformedPoint.z = point.mZ;
transformedPoint.w = point.mW;
transformedPoint = mMatrix3D->TransformPoint(transformedPoint);
retval->SetX(transformedPoint.x);
retval->SetY(transformedPoint.y);
retval->SetZ(transformedPoint.z);
retval->SetW(transformedPoint.w);
} else if (point.mZ != 0 || point.mW != 1.0) {
gfx::Matrix4x4Double tempMatrix(gfx::Matrix4x4Double::From2D(*mMatrix2D));
gfx::PointDouble4D transformedPoint;
transformedPoint.x = point.mX;
transformedPoint.y = point.mY;
transformedPoint.z = point.mZ;
transformedPoint.w = point.mW;
transformedPoint = tempMatrix.TransformPoint(transformedPoint);
retval->SetX(transformedPoint.x);
retval->SetY(transformedPoint.y);
retval->SetZ(transformedPoint.z);
retval->SetW(transformedPoint.w);
} else {
gfx::PointDouble transformedPoint;
transformedPoint.x = point.mX;
transformedPoint.y = point.mY;
transformedPoint = mMatrix2D->TransformPoint(transformedPoint);
retval->SetX(transformedPoint.x);
retval->SetY(transformedPoint.y);
retval->SetZ(point.mZ);
retval->SetW(point.mW);
}
return retval.forget();
}
template <typename T>
void GetDataFromMatrix(const DOMMatrixReadOnly* aMatrix, T* aData) {
aData[0] = static_cast<T>(aMatrix->M11());
aData[1] = static_cast<T>(aMatrix->M12());
aData[2] = static_cast<T>(aMatrix->M13());
aData[3] = static_cast<T>(aMatrix->M14());
aData[4] = static_cast<T>(aMatrix->M21());
aData[5] = static_cast<T>(aMatrix->M22());
aData[6] = static_cast<T>(aMatrix->M23());
aData[7] = static_cast<T>(aMatrix->M24());
aData[8] = static_cast<T>(aMatrix->M31());
aData[9] = static_cast<T>(aMatrix->M32());
aData[10] = static_cast<T>(aMatrix->M33());
aData[11] = static_cast<T>(aMatrix->M34());
aData[12] = static_cast<T>(aMatrix->M41());
aData[13] = static_cast<T>(aMatrix->M42());
aData[14] = static_cast<T>(aMatrix->M43());
aData[15] = static_cast<T>(aMatrix->M44());
}
void DOMMatrixReadOnly::ToFloat32Array(JSContext* aCx,
JS::MutableHandle<JSObject*> aResult,
ErrorResult& aRv) const {
AutoTArray<float, 16> arr;
arr.SetLength(16);
GetDataFromMatrix(this, arr.Elements());
JS::Rooted<JS::Value> value(aCx);
if (!ToJSValue(aCx, TypedArrayCreator<Float32Array>(arr), &value)) {
aRv.Throw(NS_ERROR_OUT_OF_MEMORY);
return;
}
aResult.set(&value.toObject());
}
void DOMMatrixReadOnly::ToFloat64Array(JSContext* aCx,
JS::MutableHandle<JSObject*> aResult,
ErrorResult& aRv) const {
AutoTArray<double, 16> arr;
arr.SetLength(16);
GetDataFromMatrix(this, arr.Elements());
JS::Rooted<JS::Value> value(aCx);
if (!ToJSValue(aCx, TypedArrayCreator<Float64Array>(arr), &value)) {
aRv.Throw(NS_ERROR_OUT_OF_MEMORY);
return;
}
aResult.set(&value.toObject());
}
void DOMMatrixReadOnly::Stringify(nsAString& aResult, ErrorResult& aRv) {
char cbuf[JS::MaximumNumberToStringLength];
nsAutoString matrixStr;
auto AppendDouble = [&aRv, &cbuf, &matrixStr](double d,
bool isLastItem = false) {
if (!mozilla::IsFinite(d)) {
aRv.ThrowInvalidStateError(
"Matrix with a non-finite element cannot be stringified.");
return false;
}
JS::NumberToString(d, cbuf);
matrixStr.AppendASCII(cbuf);
if (!isLastItem) {
matrixStr.AppendLiteral(", ");
}
return true;
};
if (mMatrix3D) {
// We can't use AppendPrintf here, because it does locale-specific
// formatting of floating-point values.
matrixStr.AssignLiteral("matrix3d(");
if (!AppendDouble(M11()) || !AppendDouble(M12()) || !AppendDouble(M13()) ||
!AppendDouble(M14()) || !AppendDouble(M21()) || !AppendDouble(M22()) ||
!AppendDouble(M23()) || !AppendDouble(M24()) || !AppendDouble(M31()) ||
!AppendDouble(M32()) || !AppendDouble(M33()) || !AppendDouble(M34()) ||
!AppendDouble(M41()) || !AppendDouble(M42()) || !AppendDouble(M43()) ||
!AppendDouble(M44(), true)) {
return;
}
matrixStr.AppendLiteral(")");
} else {
// We can't use AppendPrintf here, because it does locale-specific
// formatting of floating-point values.
matrixStr.AssignLiteral("matrix(");
if (!AppendDouble(A()) || !AppendDouble(B()) || !AppendDouble(C()) ||
!AppendDouble(D()) || !AppendDouble(E()) || !AppendDouble(F(), true)) {
return;
}
matrixStr.AppendLiteral(")");
}
aResult = matrixStr;
}
// https://drafts.fxtf.org/geometry/#structured-serialization
bool DOMMatrixReadOnly::WriteStructuredClone(
JSContext* aCx, JSStructuredCloneWriter* aWriter) const {
const uint8_t is2D = Is2D();
if (!JS_WriteBytes(aWriter, &is2D, 1)) {
return false;
}
if (is2D == 1) {
return JS_WriteDouble(aWriter, mMatrix2D->_11) &&
JS_WriteDouble(aWriter, mMatrix2D->_12) &&
JS_WriteDouble(aWriter, mMatrix2D->_21) &&
JS_WriteDouble(aWriter, mMatrix2D->_22) &&
JS_WriteDouble(aWriter, mMatrix2D->_31) &&
JS_WriteDouble(aWriter, mMatrix2D->_32);
}
return JS_WriteDouble(aWriter, mMatrix3D->_11) &&
JS_WriteDouble(aWriter, mMatrix3D->_12) &&
JS_WriteDouble(aWriter, mMatrix3D->_13) &&
JS_WriteDouble(aWriter, mMatrix3D->_14) &&
JS_WriteDouble(aWriter, mMatrix3D->_21) &&
JS_WriteDouble(aWriter, mMatrix3D->_22) &&
JS_WriteDouble(aWriter, mMatrix3D->_23) &&
JS_WriteDouble(aWriter, mMatrix3D->_24) &&
JS_WriteDouble(aWriter, mMatrix3D->_31) &&
JS_WriteDouble(aWriter, mMatrix3D->_32) &&
JS_WriteDouble(aWriter, mMatrix3D->_33) &&
JS_WriteDouble(aWriter, mMatrix3D->_34) &&
JS_WriteDouble(aWriter, mMatrix3D->_41) &&
JS_WriteDouble(aWriter, mMatrix3D->_42) &&
JS_WriteDouble(aWriter, mMatrix3D->_43) &&
JS_WriteDouble(aWriter, mMatrix3D->_44);
}
bool DOMMatrixReadOnly::ReadStructuredCloneElements(
JSStructuredCloneReader* aReader, DOMMatrixReadOnly* matrix) {
if (matrix->Is2D() == 1) {
JS_ReadDouble(aReader, &(matrix->mMatrix2D->_11));
JS_ReadDouble(aReader, &(matrix->mMatrix2D->_12));
JS_ReadDouble(aReader, &(matrix->mMatrix2D->_21));
JS_ReadDouble(aReader, &(matrix->mMatrix2D->_22));
JS_ReadDouble(aReader, &(matrix->mMatrix2D->_31));
JS_ReadDouble(aReader, &(matrix->mMatrix2D->_32));
} else {
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_11));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_12));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_13));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_14));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_21));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_22));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_23));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_24));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_31));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_32));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_33));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_34));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_41));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_42));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_43));
JS_ReadDouble(aReader, &(matrix->mMatrix3D->_44));
}
return true;
}
already_AddRefed<DOMMatrix> DOMMatrix::FromMatrix(
nsISupports* aParent, const DOMMatrixInit& aMatrixInit, ErrorResult& aRv) {
DOMMatrixInit matrixInit(aMatrixInit);
if (!ValidateAndFixupMatrixInit(matrixInit, aRv)) {
return nullptr;
};
RefPtr<DOMMatrix> matrix = new DOMMatrix(aParent, matrixInit.mIs2D.Value());
matrix->SetDataFromMatrixInit(matrixInit);
return matrix.forget();
}
already_AddRefed<DOMMatrix> DOMMatrix::FromMatrix(
const GlobalObject& aGlobal, const DOMMatrixInit& aMatrixInit,
ErrorResult& aRv) {
RefPtr<DOMMatrix> matrix =
FromMatrix(aGlobal.GetAsSupports(), aMatrixInit, aRv);
return matrix.forget();
}
already_AddRefed<DOMMatrix> DOMMatrix::FromFloat32Array(
const GlobalObject& aGlobal, const Float32Array& aArray32,
ErrorResult& aRv) {
aArray32.ComputeState();
const int length = aArray32.Length();
const bool is2D = length == 6;
RefPtr<DOMMatrix> obj = new DOMMatrix(aGlobal.GetAsSupports(), is2D);
SetDataInMatrix(obj, aArray32.Data(), length, aRv);
return obj.forget();
}
already_AddRefed<DOMMatrix> DOMMatrix::FromFloat64Array(
const GlobalObject& aGlobal, const Float64Array& aArray64,
ErrorResult& aRv) {
aArray64.ComputeState();
const int length = aArray64.Length();
const bool is2D = length == 6;
RefPtr<DOMMatrix> obj = new DOMMatrix(aGlobal.GetAsSupports(), is2D);
SetDataInMatrix(obj, aArray64.Data(), length, aRv);
return obj.forget();
}
already_AddRefed<DOMMatrix> DOMMatrix::Constructor(
const GlobalObject& aGlobal,
const Optional<UTF8StringOrUnrestrictedDoubleSequenceOrDOMMatrixReadOnly>&
aArg,
ErrorResult& aRv) {
if (!aArg.WasPassed()) {
RefPtr<DOMMatrix> rval = new DOMMatrix(aGlobal.GetAsSupports());
return rval.forget();
}
const auto& arg = aArg.Value();
if (arg.IsUTF8String()) {
nsCOMPtr<nsPIDOMWindowInner> win =
do_QueryInterface(aGlobal.GetAsSupports());
if (!win) {
aRv.ThrowTypeError<MSG_ILLEGAL_CONSTRUCTOR>();
return nullptr;
}
RefPtr<DOMMatrix> rval = new DOMMatrix(aGlobal.GetAsSupports());
rval->SetMatrixValue(arg.GetAsUTF8String(), aRv);
return rval.forget();
}
if (arg.IsDOMMatrixReadOnly()) {
RefPtr<DOMMatrix> obj =
new DOMMatrix(aGlobal.GetAsSupports(), arg.GetAsDOMMatrixReadOnly());
return obj.forget();
}
const auto& sequence = arg.GetAsUnrestrictedDoubleSequence();
const int length = sequence.Length();
const bool is2D = length == 6;
RefPtr<DOMMatrix> rval = new DOMMatrix(aGlobal.GetAsSupports(), is2D);
SetDataInMatrix(rval, sequence.Elements(), length, aRv);
return rval.forget();
}
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
template <typename T>
static void SetDataInMatrix(DOMMatrixReadOnly* aMatrix, const T* aData,
int aLength, ErrorResult& aRv) {
if (aLength == 16) {
aMatrix->SetM11(aData[0]);
aMatrix->SetM12(aData[1]);
aMatrix->SetM13(aData[2]);
aMatrix->SetM14(aData[3]);
aMatrix->SetM21(aData[4]);
aMatrix->SetM22(aData[5]);
aMatrix->SetM23(aData[6]);
aMatrix->SetM24(aData[7]);
aMatrix->SetM31(aData[8]);
aMatrix->SetM32(aData[9]);
aMatrix->SetM33(aData[10]);
aMatrix->SetM34(aData[11]);
aMatrix->SetM41(aData[12]);
aMatrix->SetM42(aData[13]);
aMatrix->SetM43(aData[14]);
aMatrix->SetM44(aData[15]);
} else if (aLength == 6) {
aMatrix->SetA(aData[0]);
aMatrix->SetB(aData[1]);
aMatrix->SetC(aData[2]);
aMatrix->SetD(aData[3]);
aMatrix->SetE(aData[4]);
aMatrix->SetF(aData[5]);
} else {
nsAutoCString lengthStr;
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
lengthStr.AppendInt(aLength);
aRv.ThrowTypeError<MSG_MATRIX_INIT_LENGTH_WRONG>(lengthStr);
}
}
already_AddRefed<DOMMatrix> DOMMatrix::ReadStructuredClone(
JSContext* aCx, nsIGlobalObject* aGlobal,
JSStructuredCloneReader* aReader) {
uint8_t is2D;
if (!JS_ReadBytes(aReader, &is2D, 1)) {
return nullptr;
}
RefPtr<DOMMatrix> rval = new DOMMatrix(aGlobal, is2D);
if (!ReadStructuredCloneElements(aReader, rval)) {
return nullptr;
};
return rval.forget();
}
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
void DOMMatrixReadOnly::Ensure3DMatrix() {
if (!mMatrix3D) {
mMatrix3D = MakeUnique<gfx::Matrix4x4Double>(
gfx::Matrix4x4Double::From2D(*mMatrix2D));
mMatrix2D = nullptr;
}
}
DOMMatrix* DOMMatrix::MultiplySelf(const DOMMatrixInit& aOtherInit,
ErrorResult& aRv) {
RefPtr<DOMMatrix> other = FromMatrix(mParent, aOtherInit, aRv);
if (aRv.Failed()) {
return nullptr;
}
MOZ_ASSERT(other);
if (other->IsIdentity()) {
return this;
}
if (other->Is2D()) {
if (mMatrix3D) {
*mMatrix3D = gfx::Matrix4x4Double::From2D(*other->mMatrix2D) * *mMatrix3D;
} else {
*mMatrix2D = *other->mMatrix2D * *mMatrix2D;
}
} else {
Ensure3DMatrix();
*mMatrix3D = *other->mMatrix3D * *mMatrix3D;
}
return this;
}
DOMMatrix* DOMMatrix::PreMultiplySelf(const DOMMatrixInit& aOtherInit,
ErrorResult& aRv) {
RefPtr<DOMMatrix> other = FromMatrix(mParent, aOtherInit, aRv);
if (aRv.Failed()) {
return nullptr;
}
MOZ_ASSERT(other);
if (other->IsIdentity()) {
return this;
}
if (other->Is2D()) {
if (mMatrix3D) {
*mMatrix3D = *mMatrix3D * gfx::Matrix4x4Double::From2D(*other->mMatrix2D);
} else {
*mMatrix2D = *mMatrix2D * *other->mMatrix2D;
}
} else {
Ensure3DMatrix();
*mMatrix3D = *mMatrix3D * *other->mMatrix3D;
}
return this;
}
DOMMatrix* DOMMatrix::TranslateSelf(double aTx, double aTy, double aTz) {
if (aTx == 0 && aTy == 0 && aTz == 0) {
return this;
}
if (mMatrix3D || aTz != 0) {
Ensure3DMatrix();
mMatrix3D->PreTranslate(aTx, aTy, aTz);
} else {
mMatrix2D->PreTranslate(aTx, aTy);
}
return this;
}
DOMMatrix* DOMMatrix::ScaleSelf(double aScaleX, const Optional<double>& aScaleY,
double aScaleZ, double aOriginX,
double aOriginY, double aOriginZ) {
const double scaleY = aScaleY.WasPassed() ? aScaleY.Value() : aScaleX;
TranslateSelf(aOriginX, aOriginY, aOriginZ);
if (mMatrix3D || aScaleZ != 1.0) {
Ensure3DMatrix();
gfx::Matrix4x4Double m;
m._11 = aScaleX;
m._22 = scaleY;
m._33 = aScaleZ;
*mMatrix3D = m * *mMatrix3D;
} else {
gfx::MatrixDouble m;
m._11 = aScaleX;
m._22 = scaleY;
*mMatrix2D = m * *mMatrix2D;
}
TranslateSelf(-aOriginX, -aOriginY, -aOriginZ);
return this;
}
DOMMatrix* DOMMatrix::Scale3dSelf(double aScale, double aOriginX,
double aOriginY, double aOriginZ) {
ScaleSelf(aScale, Optional<double>(aScale), aScale, aOriginX, aOriginY,
aOriginZ);
return this;
}
DOMMatrix* DOMMatrix::RotateFromVectorSelf(double aX, double aY) {
const double angle = (aX == 0.0 && aY == 0.0) ? 0 : atan2(aY, aX);
if (fmod(angle, 2 * M_PI) == 0) {
return this;
}
if (mMatrix3D) {
RotateAxisAngleSelf(0, 0, 1, angle / radPerDegree);
} else {
*mMatrix2D = mMatrix2D->PreRotate(angle);
}
return this;
}
DOMMatrix* DOMMatrix::RotateSelf(double aRotX, const Optional<double>& aRotY,
const Optional<double>& aRotZ) {
double rotY;
double rotZ;
if (!aRotY.WasPassed() && !aRotZ.WasPassed()) {
rotZ = aRotX;
aRotX = 0;
rotY = 0;
} else {
rotY = aRotY.WasPassed() ? aRotY.Value() : 0;
rotZ = aRotZ.WasPassed() ? aRotZ.Value() : 0;
}
if (aRotX != 0 || rotY != 0) {
Ensure3DMatrix();
}
if (mMatrix3D) {
if (fmod(rotZ, 360) != 0) {
mMatrix3D->RotateZ(rotZ * radPerDegree);
}
if (fmod(rotY, 360) != 0) {
mMatrix3D->RotateY(rotY * radPerDegree);
}
if (fmod(aRotX, 360) != 0) {
mMatrix3D->RotateX(aRotX * radPerDegree);
}
} else if (fmod(rotZ, 360) != 0) {
*mMatrix2D = mMatrix2D->PreRotate(rotZ * radPerDegree);
}
return this;
}
DOMMatrix* DOMMatrix::RotateAxisAngleSelf(double aX, double aY, double aZ,
double aAngle) {
if (fmod(aAngle, 360) == 0) {
return this;
}
aAngle *= radPerDegree;
Ensure3DMatrix();
gfx::Matrix4x4Double m;
m.SetRotateAxisAngle(aX, aY, aZ, aAngle);
*mMatrix3D = m * *mMatrix3D;
return this;
}
DOMMatrix* DOMMatrix::SkewXSelf(double aSx) {
if (fmod(aSx, 360) == 0) {
return this;
}
if (mMatrix3D) {
gfx::Matrix4x4Double m;
m._21 = tan(aSx * radPerDegree);
*mMatrix3D = m * *mMatrix3D;
} else {
gfx::MatrixDouble m;
m._21 = tan(aSx * radPerDegree);
*mMatrix2D = m * *mMatrix2D;
}
return this;
}
DOMMatrix* DOMMatrix::SkewYSelf(double aSy) {
if (fmod(aSy, 360) == 0) {
return this;
}
if (mMatrix3D) {
gfx::Matrix4x4Double m;
m._12 = tan(aSy * radPerDegree);
*mMatrix3D = m * *mMatrix3D;
} else {
gfx::MatrixDouble m;
m._12 = tan(aSy * radPerDegree);
*mMatrix2D = m * *mMatrix2D;
}
return this;
}
DOMMatrix* DOMMatrix::InvertSelf() {
if (mMatrix3D) {
if (!mMatrix3D->Invert()) {
mMatrix3D->SetNAN();
}
} else if (!mMatrix2D->Invert()) {
mMatrix2D = nullptr;
mMatrix3D = MakeUnique<gfx::Matrix4x4Double>();
mMatrix3D->SetNAN();
}
return this;
}
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
DOMMatrixReadOnly* DOMMatrixReadOnly::SetMatrixValue(
const nsACString& aTransformList, ErrorResult& aRv) {
// An empty string is a no-op.
if (aTransformList.IsEmpty()) {
return this;
}
gfx::Matrix4x4 transform;
bool contains3dTransform = false;
if (!ServoCSSParser::ParseTransformIntoMatrix(
aTransformList, contains3dTransform, transform)) {
aRv.Throw(NS_ERROR_DOM_SYNTAX_ERR);
return nullptr;
}
if (!contains3dTransform) {
mMatrix3D = nullptr;
if (!mMatrix2D) {
mMatrix2D = MakeUnique<gfx::MatrixDouble>();
}
SetA(transform._11);
SetB(transform._12);
SetC(transform._21);
SetD(transform._22);
SetE(transform._41);
SetF(transform._42);
} else {
mMatrix3D = MakeUnique<gfx::Matrix4x4Double>(transform);
mMatrix2D = nullptr;
}
return this;
}
DOMMatrix* DOMMatrix::SetMatrixValue(const nsACString& aTransformList,
Bug 1186265 - Partially update DOMPoint, DOMQuad, DOMRect, DOMMatrix. r=bz Some notes: this does not fully bring us to compliance to the current spec. Instead, these are the fixes that I needed to make in order to make css/geometry/interfaces.html pass with the DOMPoint changes in the previous patches. I don't fully understand why that patch caused the test to fail the way it did, but it ended up being easier to fix our code than understand why the harness was falling over. The DOMQuad::QuadBounds class was the source of some confusion for me. Now that DOMRectReadOnly is a concrete class with members, I wanted to avoid wasting them. However, the spec is unclear as to whether a DOMQuad's bound's should be live -- that is because DOMQuad exposes DOMPoint, we can set its points after retrieving a QuadBounds object. Our current code is live, setting the points changes the QuadBounds. Chromium's current behavior is to never update the QuadBounds object. I've left our behavior untouched in this patch (and waste 4 doubles per QuadBounds object), but I am intending to file a bug to understand what the intent of the spec is. I wonder if the author intended the points to be DOMPointReadOnly instead. If so, we could simplify the DOMRectReadOnly code and get rid of the virtual getters, which would be nice. I also wasn't thrilled to put the DOMMatrix setters on the DOMMatrixReadOnly class, but for brevity and simplicity of implementation, I've made them public. I briefly considered making the setters protected on the ReadOnly version of the class, but I'm not convinced that having to explicitly make them public on the derived class is worth the extra copies of the names. MozReview-Commit-ID: CjdW4Nbnc6A --HG-- extra : rebase_source : 44489693afebff571a415b487e29fa6153288421
2018-03-30 02:19:31 +03:00
ErrorResult& aRv) {
DOMMatrixReadOnly::SetMatrixValue(aTransformList, aRv);
return this;
}
Bug 1117172 part 3. Change the wrappercached WrapObject methods to allow passing in aGivenProto. r=peterv The only manual changes here are to BindingUtils.h, BindingUtils.cpp, Codegen.py, Element.cpp, IDBFileRequest.cpp, IDBObjectStore.cpp, dom/workers/Navigator.cpp, WorkerPrivate.cpp, DeviceStorageRequestChild.cpp, Notification.cpp, nsGlobalWindow.cpp, MessagePort.cpp, nsJSEnvironment.cpp, Sandbox.cpp, XPCConvert.cpp, ExportHelpers.cpp, and DataStoreService.cpp. The rest of this diff was generated by running the following commands: find . -name "*.h" -o -name "*.cpp" | xargs perl -pi -e 'BEGIN { $/ = undef } s/(WrapObjectInternal\(JSContext *\* *(?:aCx|cx|aContext|aCtx|js))\)/\1, JS::Handle<JSObject*> aGivenProto)/g' find . -name "*.h" -o -name "*.cpp" | xargs perl -pi -e 'BEGIN { $/ = undef } s/(WrapObjectInternal\((?:aCx|cx|aContext|aCtx|js))\)/\1, aGivenProto)/g' find . -name "*.h" -o -name "*.cpp" | xargs perl -pi -e 'BEGIN { $/ = undef } s/(WrapNode\(JSContext *\* *(?:aCx|cx|aContext|aCtx|js))\)/\1, JS::Handle<JSObject*> aGivenProto)/g' find . -name "*.h" -o -name "*.cpp" | xargs perl -pi -e 'BEGIN { $/ = undef } s/(WrapNode\((?:aCx|cx|aContext|aCtx|js))\)/\1, aGivenProto)/g' find . -name "*.h" -o -name "*.cpp" | xargs perl -pi -e 'BEGIN { $/ = undef } s/(WrapObject\(JSContext *\* *(?:aCx|cx|aContext|aCtx|js))\)/\1, JS::Handle<JSObject*> aGivenProto)/g' find . -name "*.h" -o -name "*.cpp" | xargs perl -pi -e 'BEGIN { $/ = undef } s/(Binding(?:_workers)?::Wrap\((?:aCx|cx|aContext|aCtx|js), [^,)]+)\)/\1, aGivenProto)/g'
2015-03-19 17:13:33 +03:00
JSObject* DOMMatrix::WrapObject(JSContext* aCx,
JS::Handle<JSObject*> aGivenProto) {
return DOMMatrix_Binding::Wrap(aCx, this, aGivenProto);
}
} // namespace mozilla::dom