blob: 0f0d01b9cdb060b2fcf737ed3f6910d6b2fda2f9 [file] [log] [blame]
/*
* Copyright (C) 2007-2019 Apple Inc. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY APPLE INC. AND ITS CONTRIBUTORS ``AS IS'' AND ANY
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL APPLE INC. OR ITS CONTRIBUTORS BE LIABLE FOR ANY
* DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
* ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include "config.h"
#include "ComplexTextController.h"
#include "CharacterProperties.h"
#include "FloatSize.h"
#include "FontCascade.h"
#include "RenderBlock.h"
#include "RenderText.h"
#include "TextRun.h"
#include <unicode/ubrk.h>
#include <wtf/StdLibExtras.h>
#include <wtf/text/TextBreakIterator.h>
#include <wtf/unicode/CharacterNames.h>
#if PLATFORM(IOS_FAMILY)
#include <CoreText/CoreText.h>
#endif
namespace WebCore {
class TextLayout {
WTF_MAKE_FAST_ALLOCATED;
public:
static bool isNeeded(RenderText& text, const FontCascade& font)
{
TextRun run = RenderBlock::constructTextRun(text, text.style());
return font.codePath(run) == FontCascade::CodePath::Complex;
}
TextLayout(RenderText& text, const FontCascade& font, float xPos)
: m_font(font)
, m_run(constructTextRun(text, xPos))
, m_controller(makeUnique<ComplexTextController>(m_font, m_run, true))
{
}
float width(unsigned from, unsigned len, HashSet<const Font*>* fallbackFonts)
{
m_controller->advance(from, 0, ByWholeGlyphs, fallbackFonts);
float beforeWidth = m_controller->runWidthSoFar();
if (m_font.wordSpacing() && from && FontCascade::treatAsSpace(m_run[from]))
beforeWidth += m_font.wordSpacing();
m_controller->advance(from + len, 0, ByWholeGlyphs, fallbackFonts);
float afterWidth = m_controller->runWidthSoFar();
return afterWidth - beforeWidth;
}
private:
static TextRun constructTextRun(RenderText& text, float xPos)
{
TextRun run = RenderBlock::constructTextRun(text, text.style());
run.setXPos(xPos);
return run;
}
// ComplexTextController has only references to its FontCascade and TextRun so they must be kept alive here.
FontCascade m_font;
TextRun m_run;
std::unique_ptr<ComplexTextController> m_controller;
};
void TextLayoutDeleter::operator()(TextLayout* layout) const
{
delete layout;
}
std::unique_ptr<TextLayout, TextLayoutDeleter> FontCascade::createLayout(RenderText& text, float xPos, bool collapseWhiteSpace) const
{
if (!collapseWhiteSpace || !TextLayout::isNeeded(text, *this))
return nullptr;
return std::unique_ptr<TextLayout, TextLayoutDeleter>(new TextLayout(text, *this, xPos));
}
float FontCascade::width(TextLayout& layout, unsigned from, unsigned len, HashSet<const Font*>* fallbackFonts)
{
return layout.width(from, len, fallbackFonts);
}
void ComplexTextController::computeExpansionOpportunity()
{
if (!m_expansion)
m_expansionPerOpportunity = 0;
else {
unsigned expansionOpportunityCount = FontCascade::expansionOpportunityCount(m_run.text(), m_run.ltr() ? TextDirection::LTR : TextDirection::RTL, m_run.expansionBehavior()).first;
if (!expansionOpportunityCount)
m_expansionPerOpportunity = 0;
else
m_expansionPerOpportunity = m_expansion / expansionOpportunityCount;
}
}
ComplexTextController::ComplexTextController(const FontCascade& font, const TextRun& run, bool mayUseNaturalWritingDirection, HashSet<const Font*>* fallbackFonts, bool forTextEmphasis)
: m_fallbackFonts(fallbackFonts)
, m_font(font)
, m_run(run)
, m_end(run.length())
, m_expansion(run.expansion())
, m_mayUseNaturalWritingDirection(mayUseNaturalWritingDirection)
, m_forTextEmphasis(forTextEmphasis)
{
computeExpansionOpportunity();
collectComplexTextRuns();
finishConstruction();
}
ComplexTextController::ComplexTextController(const FontCascade& font, const TextRun& run, Vector<Ref<ComplexTextRun>>& runs)
: m_font(font)
, m_run(run)
, m_end(run.length())
, m_expansion(run.expansion())
{
computeExpansionOpportunity();
for (auto& run : runs)
m_complexTextRuns.append(run.ptr());
finishConstruction();
}
void ComplexTextController::finishConstruction()
{
adjustGlyphsAndAdvances();
if (!m_isLTROnly) {
unsigned length = m_complexTextRuns.size();
m_runIndices.reserveInitialCapacity(length);
for (unsigned i = 0; i < length; ++i)
m_runIndices.uncheckedAppend(length - i - 1);
std::sort(m_runIndices.data(), m_runIndices.data() + length,
[this](auto a, auto b) {
return stringBegin(*m_complexTextRuns[a]) < stringBegin(*m_complexTextRuns[b]);
});
m_glyphCountFromStartToIndex.reserveInitialCapacity(length);
unsigned glyphCountSoFar = 0;
for (unsigned i = 0; i < length; ++i) {
m_glyphCountFromStartToIndex.uncheckedAppend(glyphCountSoFar);
glyphCountSoFar += m_complexTextRuns[i]->glyphCount();
}
}
}
unsigned ComplexTextController::offsetForPosition(float h, bool includePartialGlyphs)
{
if (h >= m_totalAdvance.width())
return m_run.ltr() ? m_end : 0;
if (h < 0)
return m_run.ltr() ? 0 : m_end;
float x = h;
size_t runCount = m_complexTextRuns.size();
unsigned offsetIntoAdjustedGlyphs = 0;
for (size_t r = 0; r < runCount; ++r) {
const ComplexTextRun& complexTextRun = *m_complexTextRuns[r];
for (unsigned j = 0; j < complexTextRun.glyphCount(); ++j) {
unsigned index = offsetIntoAdjustedGlyphs + j;
float adjustedAdvance = m_adjustedBaseAdvances[index].width();
bool hit = m_run.ltr() ? x < adjustedAdvance : (x <= adjustedAdvance && adjustedAdvance);
if (hit) {
unsigned hitGlyphStart = complexTextRun.indexAt(j);
unsigned hitGlyphEnd;
if (m_run.ltr())
hitGlyphEnd = std::max(hitGlyphStart, j + 1 < complexTextRun.glyphCount() ? complexTextRun.indexAt(j + 1) : complexTextRun.indexEnd());
else
hitGlyphEnd = std::max(hitGlyphStart, j > 0 ? complexTextRun.indexAt(j - 1) : complexTextRun.indexEnd());
// FIXME: Instead of dividing the glyph's advance equally between the characters, this
// could use the glyph's "ligature carets". This is available in CoreText via CTFontGetLigatureCaretPositions().
unsigned hitIndex;
if (m_run.ltr())
hitIndex = hitGlyphStart + (hitGlyphEnd - hitGlyphStart) * (x / adjustedAdvance);
else {
if (hitGlyphStart == hitGlyphEnd)
hitIndex = hitGlyphStart;
else if (x)
hitIndex = hitGlyphEnd - (hitGlyphEnd - hitGlyphStart) * (x / adjustedAdvance);
else
hitIndex = hitGlyphEnd - 1;
}
unsigned stringLength = complexTextRun.stringLength();
CachedTextBreakIterator cursorPositionIterator(StringView(complexTextRun.characters(), stringLength), TextBreakIterator::Mode::Caret, nullAtom());
unsigned clusterStart;
if (cursorPositionIterator.isBoundary(hitIndex))
clusterStart = hitIndex;
else
clusterStart = cursorPositionIterator.preceding(hitIndex).value_or(0);
if (!includePartialGlyphs)
return complexTextRun.stringLocation() + clusterStart;
unsigned clusterEnd = cursorPositionIterator.following(hitIndex).value_or(stringLength);
float clusterWidth;
// FIXME: The search stops at the boundaries of complexTextRun. In theory, it should go on into neighboring ComplexTextRuns
// derived from the same CTLine. In practice, we do not expect there to be more than one CTRun in a CTLine, as no
// reordering and no font fallback should occur within a CTLine.
if (clusterEnd - clusterStart > 1) {
clusterWidth = adjustedAdvance;
if (j) {
unsigned firstGlyphBeforeCluster = j - 1;
while (complexTextRun.indexAt(firstGlyphBeforeCluster) >= clusterStart && complexTextRun.indexAt(firstGlyphBeforeCluster) < clusterEnd) {
float width = m_adjustedBaseAdvances[offsetIntoAdjustedGlyphs + firstGlyphBeforeCluster].width();
clusterWidth += width;
x += width;
if (!firstGlyphBeforeCluster)
break;
firstGlyphBeforeCluster--;
}
}
unsigned firstGlyphAfterCluster = j + 1;
while (firstGlyphAfterCluster < complexTextRun.glyphCount() && complexTextRun.indexAt(firstGlyphAfterCluster) >= clusterStart && complexTextRun.indexAt(firstGlyphAfterCluster) < clusterEnd) {
clusterWidth += m_adjustedBaseAdvances[offsetIntoAdjustedGlyphs + firstGlyphAfterCluster].width();
firstGlyphAfterCluster++;
}
} else {
clusterWidth = adjustedAdvance / (hitGlyphEnd - hitGlyphStart);
x -= clusterWidth * (m_run.ltr() ? hitIndex - hitGlyphStart : hitGlyphEnd - hitIndex - 1);
}
if (x <= clusterWidth / 2)
return complexTextRun.stringLocation() + (m_run.ltr() ? clusterStart : clusterEnd);
return complexTextRun.stringLocation() + (m_run.ltr() ? clusterEnd : clusterStart);
}
x -= adjustedAdvance;
}
offsetIntoAdjustedGlyphs += complexTextRun.glyphCount();
}
ASSERT_NOT_REACHED();
return 0;
}
// FIXME: We should consider reimplementing this function using ICU to advance by grapheme.
// The current implementation only considers explicitly emoji sequences and emoji variations.
static bool advanceByCombiningCharacterSequence(const UChar*& iterator, const UChar* end, UChar32& baseCharacter, unsigned& markCount)
{
ASSERT(iterator < end);
markCount = 0;
unsigned i = 0;
unsigned remainingCharacters = end - iterator;
U16_NEXT(iterator, i, remainingCharacters, baseCharacter);
iterator = iterator + i;
if (U_IS_SURROGATE(baseCharacter))
return false;
// Consume marks.
bool sawEmojiGroupCandidate = isEmojiGroupCandidate(baseCharacter);
bool sawJoiner = false;
bool sawRegionalIndicator = isEmojiRegionalIndicator(baseCharacter);
while (iterator < end) {
UChar32 nextCharacter;
unsigned markLength = 0;
bool shouldContinue = false;
ASSERT(end >= iterator);
U16_NEXT(iterator, markLength, static_cast<unsigned>(end - iterator), nextCharacter);
if (isVariationSelector(nextCharacter) || isEmojiFitzpatrickModifier(nextCharacter))
shouldContinue = true;
if (sawRegionalIndicator && isEmojiRegionalIndicator(nextCharacter)) {
shouldContinue = true;
sawRegionalIndicator = false;
}
if (sawJoiner && isEmojiGroupCandidate(nextCharacter))
shouldContinue = true;
sawJoiner = false;
if (sawEmojiGroupCandidate && nextCharacter == zeroWidthJoiner) {
sawJoiner = true;
shouldContinue = true;
}
if (!shouldContinue && !(U_GET_GC_MASK(nextCharacter) & U_GC_M_MASK))
break;
markCount += markLength;
iterator += markLength;
}
return true;
}
// FIXME: Capitalization is language-dependent and context-dependent and should operate on grapheme clusters instead of codepoints.
static inline std::optional<UChar32> capitalized(UChar32 baseCharacter)
{
if (U_GET_GC_MASK(baseCharacter) & U_GC_M_MASK)
return std::nullopt;
UChar32 uppercaseCharacter = u_toupper(baseCharacter);
ASSERT(uppercaseCharacter == baseCharacter || (U_IS_BMP(baseCharacter) == U_IS_BMP(uppercaseCharacter)));
if (uppercaseCharacter != baseCharacter)
return uppercaseCharacter;
return std::nullopt;
}
static bool shouldSynthesize(bool dontSynthesizeSmallCaps, const Font* nextFont, UChar32 baseCharacter, std::optional<UChar32> capitalizedBase, FontVariantCaps fontVariantCaps, bool engageAllSmallCapsProcessing)
{
if (dontSynthesizeSmallCaps)
return false;
if (!nextFont || nextFont == Font::systemFallback())
return false;
if (engageAllSmallCapsProcessing && isASCIISpace(baseCharacter))
return false;
if (!engageAllSmallCapsProcessing && !capitalizedBase)
return false;
return !nextFont->variantCapsSupportsCharacterForSynthesis(fontVariantCaps, baseCharacter);
}
void ComplexTextController::collectComplexTextRuns()
{
if (!m_end || !m_font.size())
return;
// We break up glyph run generation for the string by Font.
const UChar* cp;
if (m_run.is8Bit()) {
String stringFor8BitRun = String::make16BitFrom8BitSource(m_run.characters8(), m_run.length());
m_stringsFor8BitRuns.append(WTFMove(stringFor8BitRun));
cp = m_stringsFor8BitRuns.last().characters16();
} else
cp = m_run.characters16();
auto fontVariantCaps = m_font.fontDescription().variantCaps();
bool dontSynthesizeSmallCaps = !static_cast<bool>(m_font.fontDescription().fontSynthesis() & FontSynthesisSmallCaps);
bool engageAllSmallCapsProcessing = fontVariantCaps == FontVariantCaps::AllSmall || fontVariantCaps == FontVariantCaps::AllPetite;
bool engageSmallCapsProcessing = engageAllSmallCapsProcessing || fontVariantCaps == FontVariantCaps::Small || fontVariantCaps == FontVariantCaps::Petite;
if (engageAllSmallCapsProcessing || engageSmallCapsProcessing)
m_smallCapsBuffer.resize(m_end);
unsigned indexOfFontTransition = 0;
const UChar* curr = cp;
const UChar* end = cp + m_end;
const Font* font;
const Font* nextFont;
const Font* synthesizedFont = nullptr;
const Font* smallSynthesizedFont = nullptr;
unsigned markCount;
UChar32 baseCharacter;
if (!advanceByCombiningCharacterSequence(curr, end, baseCharacter, markCount))
return;
nextFont = m_font.fontForCombiningCharacterSequence(cp, curr - cp);
bool isSmallCaps = false;
bool nextIsSmallCaps = false;
auto capitalizedBase = capitalized(baseCharacter);
if (shouldSynthesize(dontSynthesizeSmallCaps, nextFont, baseCharacter, capitalizedBase, fontVariantCaps, engageAllSmallCapsProcessing)) {
synthesizedFont = &nextFont->noSynthesizableFeaturesFont();
smallSynthesizedFont = synthesizedFont->smallCapsFont(m_font.fontDescription());
UChar32 characterToWrite = capitalizedBase ? capitalizedBase.value() : cp[0];
unsigned characterIndex = 0;
U16_APPEND_UNSAFE(m_smallCapsBuffer, characterIndex, characterToWrite);
for (unsigned i = characterIndex; cp + i < curr; ++i)
m_smallCapsBuffer[i] = cp[i];
nextIsSmallCaps = true;
}
while (curr < end) {
font = nextFont;
isSmallCaps = nextIsSmallCaps;
unsigned index = curr - cp;
if (!advanceByCombiningCharacterSequence(curr, end, baseCharacter, markCount))
return;
if (synthesizedFont) {
if (auto capitalizedBase = capitalized(baseCharacter)) {
unsigned characterIndex = index;
U16_APPEND_UNSAFE(m_smallCapsBuffer, characterIndex, capitalizedBase.value());
for (unsigned i = 0; i < markCount; ++i)
m_smallCapsBuffer[i + characterIndex] = cp[i + characterIndex];
nextIsSmallCaps = true;
} else {
if (engageAllSmallCapsProcessing) {
for (unsigned i = 0; i < curr - cp - index; ++i)
m_smallCapsBuffer[index + i] = cp[index + i];
}
nextIsSmallCaps = engageAllSmallCapsProcessing;
}
}
if (baseCharacter == zeroWidthJoiner)
nextFont = font;
else
nextFont = m_font.fontForCombiningCharacterSequence(cp + index, curr - cp - index);
capitalizedBase = capitalized(baseCharacter);
if (!synthesizedFont && shouldSynthesize(dontSynthesizeSmallCaps, nextFont, baseCharacter, capitalizedBase, fontVariantCaps, engageAllSmallCapsProcessing)) {
// Rather than synthesize each character individually, we should synthesize the entire "run" if any character requires synthesis.
synthesizedFont = &nextFont->noSynthesizableFeaturesFont();
smallSynthesizedFont = synthesizedFont->smallCapsFont(m_font.fontDescription());
nextIsSmallCaps = true;
curr = cp + indexOfFontTransition;
continue;
}
if (nextFont != font || nextIsSmallCaps != isSmallCaps) {
unsigned itemLength = index - indexOfFontTransition;
if (itemLength) {
unsigned itemStart = indexOfFontTransition;
if (synthesizedFont) {
if (isSmallCaps)
collectComplexTextRunsForCharacters(m_smallCapsBuffer.data() + itemStart, itemLength, itemStart, smallSynthesizedFont);
else
collectComplexTextRunsForCharacters(cp + itemStart, itemLength, itemStart, synthesizedFont);
} else
collectComplexTextRunsForCharacters(cp + itemStart, itemLength, itemStart, font);
if (nextFont != font) {
synthesizedFont = nullptr;
smallSynthesizedFont = nullptr;
nextIsSmallCaps = false;
}
}
indexOfFontTransition = index;
}
}
ASSERT(m_end >= indexOfFontTransition);
unsigned itemLength = m_end - indexOfFontTransition;
if (itemLength) {
unsigned itemStart = indexOfFontTransition;
if (synthesizedFont) {
if (nextIsSmallCaps)
collectComplexTextRunsForCharacters(m_smallCapsBuffer.data() + itemStart, itemLength, itemStart, smallSynthesizedFont);
else
collectComplexTextRunsForCharacters(cp + itemStart, itemLength, itemStart, synthesizedFont);
} else
collectComplexTextRunsForCharacters(cp + itemStart, itemLength, itemStart, nextFont);
}
if (!m_run.ltr())
m_complexTextRuns.reverse();
}
unsigned ComplexTextController::ComplexTextRun::indexAt(unsigned i) const
{
ASSERT(i < m_glyphCount);
return m_coreTextIndices[i];
}
void ComplexTextController::ComplexTextRun::setIsNonMonotonic()
{
ASSERT(m_isMonotonic);
m_isMonotonic = false;
Vector<bool, 64> mappedIndices(m_stringLength, false);
for (unsigned i = 0; i < m_glyphCount; ++i) {
ASSERT(indexAt(i) < m_stringLength);
mappedIndices[indexAt(i)] = true;
}
m_glyphEndOffsets.grow(m_glyphCount);
for (unsigned i = 0; i < m_glyphCount; ++i) {
unsigned nextMappedIndex = m_indexEnd;
for (unsigned j = indexAt(i) + 1; j < m_stringLength; ++j) {
if (mappedIndices[j]) {
nextMappedIndex = j;
break;
}
}
m_glyphEndOffsets[i] = nextMappedIndex;
}
}
unsigned ComplexTextController::indexOfCurrentRun(unsigned& leftmostGlyph)
{
leftmostGlyph = 0;
size_t runCount = m_complexTextRuns.size();
if (m_currentRun >= runCount)
return runCount;
if (m_isLTROnly) {
for (unsigned i = 0; i < m_currentRun; ++i)
leftmostGlyph += m_complexTextRuns[i]->glyphCount();
return m_currentRun;
}
unsigned currentRunIndex = m_runIndices[m_currentRun];
leftmostGlyph = m_glyphCountFromStartToIndex[currentRunIndex];
return currentRunIndex;
}
unsigned ComplexTextController::incrementCurrentRun(unsigned& leftmostGlyph)
{
if (m_isLTROnly) {
leftmostGlyph += m_complexTextRuns[m_currentRun++]->glyphCount();
return m_currentRun;
}
m_currentRun++;
leftmostGlyph = 0;
return indexOfCurrentRun(leftmostGlyph);
}
float ComplexTextController::runWidthSoFarFraction(unsigned glyphStartOffset, unsigned glyphEndOffset, unsigned oldCharacterInCurrentGlyph, GlyphIterationStyle iterationStyle) const
{
// FIXME: Instead of dividing the glyph's advance equally between the characters, this
// could use the glyph's "ligature carets". This is available in CoreText via CTFontGetLigatureCaretPositions().
if (glyphStartOffset == glyphEndOffset) {
// When there are multiple glyphs per character we need to advance by the full width of the glyph.
ASSERT(m_characterInCurrentGlyph == oldCharacterInCurrentGlyph);
return 1;
}
if (iterationStyle == ByWholeGlyphs) {
if (!oldCharacterInCurrentGlyph)
return 1;
return 0;
}
return static_cast<float>(m_characterInCurrentGlyph - oldCharacterInCurrentGlyph) / (glyphEndOffset - glyphStartOffset);
}
void ComplexTextController::advance(unsigned offset, GlyphBuffer* glyphBuffer, GlyphIterationStyle iterationStyle, HashSet<const Font*>* fallbackFonts)
{
if (offset > m_end)
offset = m_end;
if (offset < m_currentCharacter) {
m_runWidthSoFar = 0;
m_numGlyphsSoFar = 0;
m_currentRun = 0;
m_glyphInCurrentRun = 0;
m_characterInCurrentGlyph = 0;
}
m_currentCharacter = offset;
size_t runCount = m_complexTextRuns.size();
unsigned indexOfLeftmostGlyphInCurrentRun = 0; // Relative to the beginning of ComplexTextController.
unsigned currentRunIndex = indexOfCurrentRun(indexOfLeftmostGlyphInCurrentRun);
while (m_currentRun < runCount) {
const ComplexTextRun& complexTextRun = *m_complexTextRuns[currentRunIndex];
bool ltr = complexTextRun.isLTR();
unsigned glyphCount = complexTextRun.glyphCount();
unsigned glyphIndexIntoCurrentRun = ltr ? m_glyphInCurrentRun : glyphCount - 1 - m_glyphInCurrentRun;
unsigned glyphIndexIntoComplexTextController = indexOfLeftmostGlyphInCurrentRun + glyphIndexIntoCurrentRun;
if (fallbackFonts && &complexTextRun.font() != &m_font.primaryFont())
fallbackFonts->add(&complexTextRun.font());
// We must store the initial advance for the first glyph we are going to draw.
// When leftmostGlyph is 0, it represents the first glyph to draw, taking into
// account the text direction.
if (!indexOfLeftmostGlyphInCurrentRun && glyphBuffer)
glyphBuffer->setInitialAdvance(makeGlyphBufferAdvance(complexTextRun.initialAdvance()));
while (m_glyphInCurrentRun < glyphCount) {
unsigned glyphStartOffset = complexTextRun.indexAt(glyphIndexIntoCurrentRun);
unsigned glyphEndOffset;
if (complexTextRun.isMonotonic()) {
if (ltr)
glyphEndOffset = std::max(glyphStartOffset, glyphIndexIntoCurrentRun + 1 < glyphCount ? complexTextRun.indexAt(glyphIndexIntoCurrentRun + 1) : complexTextRun.indexEnd());
else
glyphEndOffset = std::max(glyphStartOffset, glyphIndexIntoCurrentRun > 0 ? complexTextRun.indexAt(glyphIndexIntoCurrentRun - 1) : complexTextRun.indexEnd());
} else
glyphEndOffset = complexTextRun.endOffsetAt(glyphIndexIntoCurrentRun);
FloatSize adjustedBaseAdvance = m_adjustedBaseAdvances[glyphIndexIntoComplexTextController];
if (glyphStartOffset + complexTextRun.stringLocation() >= m_currentCharacter)
return;
if (glyphBuffer && !m_characterInCurrentGlyph) {
auto currentGlyphOrigin = glyphOrigin(glyphIndexIntoComplexTextController);
GlyphBufferAdvance paintAdvance = makeGlyphBufferAdvance(adjustedBaseAdvance);
if (!glyphIndexIntoCurrentRun) {
// The first layout advance of every run includes the "initial layout advance." However, here, we need
// paint advances, so subtract it out before transforming the layout advance into a paint advance.
setWidth(paintAdvance, width(paintAdvance) - (complexTextRun.initialAdvance().width() - currentGlyphOrigin.x()));
setHeight(paintAdvance, height(paintAdvance) - (complexTextRun.initialAdvance().height() - currentGlyphOrigin.y()));
}
setWidth(paintAdvance, width(paintAdvance) + glyphOrigin(glyphIndexIntoComplexTextController + 1).x() - currentGlyphOrigin.x());
setHeight(paintAdvance, height(paintAdvance) + glyphOrigin(glyphIndexIntoComplexTextController + 1).y() - currentGlyphOrigin.y());
if (glyphIndexIntoCurrentRun == glyphCount - 1 && currentRunIndex + 1 < runCount) {
// Our paint advance points to the end of the run. However, the next run may have an
// initial advance, and our paint advance needs to point to the location of the next
// glyph. So, we need to add in the next run's initial advance.
setWidth(paintAdvance, width(paintAdvance) - glyphOrigin(glyphIndexIntoComplexTextController + 1).x() + m_complexTextRuns[currentRunIndex + 1]->initialAdvance().width());
setHeight(paintAdvance, height(paintAdvance) - glyphOrigin(glyphIndexIntoComplexTextController + 1).y() + m_complexTextRuns[currentRunIndex + 1]->initialAdvance().height());
}
setHeight(paintAdvance, -height(paintAdvance)); // Increasing y points down
glyphBuffer->add(m_adjustedGlyphs[glyphIndexIntoComplexTextController], complexTextRun.font(), paintAdvance, complexTextRun.indexAt(m_glyphInCurrentRun));
}
unsigned oldCharacterInCurrentGlyph = m_characterInCurrentGlyph;
m_characterInCurrentGlyph = std::min(m_currentCharacter - complexTextRun.stringLocation(), glyphEndOffset) - glyphStartOffset;
m_runWidthSoFar += adjustedBaseAdvance.width() * runWidthSoFarFraction(glyphStartOffset, glyphEndOffset, oldCharacterInCurrentGlyph, iterationStyle);
if (glyphEndOffset + complexTextRun.stringLocation() > m_currentCharacter)
return;
m_numGlyphsSoFar++;
m_glyphInCurrentRun++;
m_characterInCurrentGlyph = 0;
if (ltr) {
glyphIndexIntoCurrentRun++;
glyphIndexIntoComplexTextController++;
} else {
glyphIndexIntoCurrentRun--;
glyphIndexIntoComplexTextController--;
}
}
currentRunIndex = incrementCurrentRun(indexOfLeftmostGlyphInCurrentRun);
m_glyphInCurrentRun = 0;
}
}
static inline std::pair<bool, bool> expansionLocation(bool ideograph, bool treatAsSpace, bool ltr, bool isAfterExpansion, bool forbidLeftExpansion, bool forbidRightExpansion, bool forceLeftExpansion, bool forceRightExpansion)
{
bool expandLeft = ideograph;
bool expandRight = ideograph;
if (treatAsSpace) {
if (ltr)
expandRight = true;
else
expandLeft = true;
}
if (isAfterExpansion)
expandLeft = false;
ASSERT(!forbidLeftExpansion || !forceLeftExpansion);
ASSERT(!forbidRightExpansion || !forceRightExpansion);
if (forbidLeftExpansion)
expandLeft = false;
if (forbidRightExpansion)
expandRight = false;
if (forceLeftExpansion)
expandLeft = true;
if (forceRightExpansion)
expandRight = true;
return std::make_pair(expandLeft, expandRight);
}
void ComplexTextController::adjustGlyphsAndAdvances()
{
bool afterExpansion = m_run.expansionBehavior().left == ExpansionBehavior::Behavior::Forbid;
size_t runCount = m_complexTextRuns.size();
bool hasExtraSpacing = (m_font.letterSpacing() || m_font.wordSpacing() || m_expansion) && !m_run.spacingDisabled();
bool runForcesLeftExpansion = m_run.expansionBehavior().left == ExpansionBehavior::Behavior::Force;
bool runForcesRightExpansion = m_run.expansionBehavior().right == ExpansionBehavior::Behavior::Force;
bool runForbidsLeftExpansion = m_run.expansionBehavior().left == ExpansionBehavior::Behavior::Forbid;
bool runForbidsRightExpansion = m_run.expansionBehavior().right == ExpansionBehavior::Behavior::Forbid;
// We are iterating in glyph order, not string order. Compare this to WidthIterator::advanceInternal()
for (size_t runIndex = 0; runIndex < runCount; ++runIndex) {
ComplexTextRun& complexTextRun = *m_complexTextRuns[runIndex];
unsigned glyphCount = complexTextRun.glyphCount();
const Font& font = complexTextRun.font();
if (!complexTextRun.isLTR())
m_isLTROnly = false;
const CGGlyph* glyphs = complexTextRun.glyphs();
const FloatSize* advances = complexTextRun.baseAdvances();
// Lower in this function, synthetic bold is blanket-applied to everything, so no need to double-apply it here.
float spaceWidth = font.spaceWidth(Font::SyntheticBoldInclusion::Exclude);
const UChar* cp = complexTextRun.characters();
FloatPoint glyphOrigin;
unsigned lastCharacterIndex = m_run.ltr() ? std::numeric_limits<unsigned>::min() : std::numeric_limits<unsigned>::max();
bool isMonotonic = true;
for (unsigned i = 0; i < glyphCount; i++) {
unsigned characterIndex = complexTextRun.indexAt(i);
if (m_run.ltr()) {
if (characterIndex < lastCharacterIndex)
isMonotonic = false;
} else {
if (characterIndex > lastCharacterIndex)
isMonotonic = false;
}
UChar ch = *(cp + characterIndex);
bool treatAsSpace = FontCascade::treatAsSpace(ch);
CGGlyph glyph = glyphs[i];
FloatSize advance = treatAsSpace ? FloatSize(spaceWidth, advances[i].height()) : advances[i];
if (ch == tabCharacter && m_run.allowTabs())
advance.setWidth(m_font.tabWidth(font, m_run.tabSize(), m_run.xPos() + m_totalAdvance.width(), Font::SyntheticBoldInclusion::Exclude));
else if (FontCascade::treatAsZeroWidthSpace(ch) && !treatAsSpace) {
advance.setWidth(0);
glyph = font.spaceGlyph();
}
if (!i) {
advance.expand(complexTextRun.initialAdvance().width(), complexTextRun.initialAdvance().height());
if (auto* origins = complexTextRun.glyphOrigins())
advance.expand(-origins[0].x(), -origins[0].y());
}
advance.expand(font.syntheticBoldOffset(), 0);
if (hasExtraSpacing) {
// If we're a glyph with an advance, add in letter-spacing.
// That way we weed out zero width lurkers. This behavior matches the fast text code path.
if (advance.width())
advance.expand(m_font.letterSpacing(), 0);
unsigned characterIndexInRun = characterIndex + complexTextRun.stringLocation();
bool isFirstCharacter = !(characterIndex + complexTextRun.stringLocation());
bool isLastCharacter = characterIndexInRun + 1 == m_run.length() || (U16_IS_LEAD(ch) && characterIndexInRun + 2 == m_run.length() && U16_IS_TRAIL(*(cp + characterIndex + 1)));
bool forceLeftExpansion = false; // On the left, regardless of m_run.ltr()
bool forceRightExpansion = false; // On the right, regardless of m_run.ltr()
bool forbidLeftExpansion = false;
bool forbidRightExpansion = false;
if (runForcesLeftExpansion)
forceLeftExpansion = m_run.ltr() ? isFirstCharacter : isLastCharacter;
if (runForcesRightExpansion)
forceRightExpansion = m_run.ltr() ? isLastCharacter : isFirstCharacter;
if (runForbidsLeftExpansion)
forbidLeftExpansion = m_run.ltr() ? isFirstCharacter : isLastCharacter;
if (runForbidsRightExpansion)
forbidRightExpansion = m_run.ltr() ? isLastCharacter : isFirstCharacter;
// Handle justification and word-spacing.
bool ideograph = FontCascade::canExpandAroundIdeographsInComplexText() && FontCascade::isCJKIdeographOrSymbol(ch);
if (treatAsSpace || ideograph || forceLeftExpansion || forceRightExpansion) {
// Distribute the run's total expansion evenly over all expansion opportunities in the run.
if (m_expansion) {
auto [expandLeft, expandRight] = expansionLocation(ideograph, treatAsSpace, m_run.ltr(), afterExpansion, forbidLeftExpansion, forbidRightExpansion, forceLeftExpansion, forceRightExpansion);
if (expandLeft) {
m_expansion -= m_expansionPerOpportunity;
// Increase previous width
if (m_adjustedBaseAdvances.isEmpty()) {
advance.expand(m_expansionPerOpportunity, 0);
complexTextRun.growInitialAdvanceHorizontally(m_expansionPerOpportunity);
} else {
m_adjustedBaseAdvances.last().expand(m_expansionPerOpportunity, 0);
m_totalAdvance.expand(m_expansionPerOpportunity, 0);
}
}
if (expandRight) {
m_expansion -= m_expansionPerOpportunity;
advance.expand(m_expansionPerOpportunity, 0);
afterExpansion = true;
}
} else
afterExpansion = false;
// Account for word-spacing.
if (treatAsSpace && (ch != '\t' || !m_run.allowTabs()) && (characterIndex > 0 || runIndex > 0 || ch == noBreakSpace) && m_font.wordSpacing())
advance.expand(m_font.wordSpacing(), 0);
} else
afterExpansion = false;
}
m_totalAdvance += advance;
// FIXME: Combining marks should receive a text emphasis mark if they are combine with a space.
if (m_forTextEmphasis && (!FontCascade::canReceiveTextEmphasis(ch) || (U_GET_GC_MASK(ch) & U_GC_M_MASK)))
glyph = 0;
m_adjustedBaseAdvances.append(advance);
if (auto* origins = complexTextRun.glyphOrigins()) {
ASSERT(m_glyphOrigins.size() < m_adjustedBaseAdvances.size());
m_glyphOrigins.grow(m_adjustedBaseAdvances.size());
m_glyphOrigins[m_glyphOrigins.size() - 1] = origins[i];
ASSERT(m_glyphOrigins.size() == m_adjustedBaseAdvances.size());
}
m_adjustedGlyphs.append(glyph);
FloatRect glyphBounds = font.boundsForGlyph(glyph);
glyphBounds.move(glyphOrigin.x(), glyphOrigin.y());
m_minGlyphBoundingBoxX = std::min(m_minGlyphBoundingBoxX, glyphBounds.x());
m_maxGlyphBoundingBoxX = std::max(m_maxGlyphBoundingBoxX, glyphBounds.maxX());
m_minGlyphBoundingBoxY = std::min(m_minGlyphBoundingBoxY, glyphBounds.y());
m_maxGlyphBoundingBoxY = std::max(m_maxGlyphBoundingBoxY, glyphBounds.maxY());
glyphOrigin.move(advance);
lastCharacterIndex = characterIndex;
}
if (!isMonotonic)
complexTextRun.setIsNonMonotonic();
}
}
// Missing glyphs run constructor. Core Text will not generate a run of missing glyphs, instead falling back on
// glyphs from LastResort. We want to use the primary font's missing glyph in order to match the fast text code path.
ComplexTextController::ComplexTextRun::ComplexTextRun(const Font& font, const UChar* characters, unsigned stringLocation, unsigned stringLength, unsigned indexBegin, unsigned indexEnd, bool ltr)
: m_font(font)
, m_characters(characters)
, m_stringLength(stringLength)
, m_indexBegin(indexBegin)
, m_indexEnd(indexEnd)
, m_stringLocation(stringLocation)
, m_isLTR(ltr)
{
auto runLengthInCodeUnits = m_indexEnd - m_indexBegin;
m_coreTextIndices.reserveInitialCapacity(runLengthInCodeUnits);
unsigned r = m_indexBegin;
while (r < m_indexEnd) {
auto currentIndex = r;
UChar32 character;
U16_NEXT(m_characters, r, m_stringLength, character);
// https://drafts.csswg.org/css-text-3/#white-space-processing
// "Unsupported Default_ignorable characters must be ignored for text rendering."
if (!FontCascade::isCharacterWhoseGlyphsShouldBeDeletedForTextRendering(character))
m_coreTextIndices.uncheckedAppend(currentIndex);
}
m_glyphCount = m_coreTextIndices.size();
if (!ltr) {
for (unsigned r = 0, end = m_glyphCount - 1; r < m_glyphCount / 2; ++r, --end)
std::swap(m_coreTextIndices[r], m_coreTextIndices[end]);
}
// Synthesize a run of missing glyphs.
m_glyphs.fill(0, m_glyphCount);
// Synthetic bold will be handled later in adjustGlyphsAndAdvances().
m_baseAdvances.fill(FloatSize(m_font.widthForGlyph(0, Font::SyntheticBoldInclusion::Exclude), 0), m_glyphCount);
}
ComplexTextController::ComplexTextRun::ComplexTextRun(const Vector<FloatSize>& advances, const Vector<FloatPoint>& origins, const Vector<Glyph>& glyphs, const Vector<unsigned>& stringIndices, FloatSize initialAdvance, const Font& font, const UChar* characters, unsigned stringLocation, unsigned stringLength, unsigned indexBegin, unsigned indexEnd, bool ltr)
: m_baseAdvances(advances)
, m_glyphOrigins(origins)
, m_glyphs(glyphs)
, m_coreTextIndices(stringIndices)
, m_initialAdvance(initialAdvance)
, m_font(font)
, m_characters(characters)
, m_stringLength(stringLength)
, m_indexBegin(indexBegin)
, m_indexEnd(indexEnd)
, m_glyphCount(glyphs.size())
, m_stringLocation(stringLocation)
, m_isLTR(ltr)
{
}
} // namespace WebCore