uniscribecontroller.cpp

linux下开源浏览器WebKit的源码,市面上的很多商用浏览器都是移植自WebKit

/*
 * Copyright (C) 2007 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. 
 * 3.  Neither the name of Apple Computer, Inc. ("Apple") nor the names of
 *     its contributors may be used to endorse or promote products derived
 *     from this software without specific prior written permission. 
 *
 * THIS SOFTWARE IS PROVIDED BY APPLE 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 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 "UniscribeController.h"
#include "Font.h"
#include "SimpleFontData.h"
#include <wtf/MathExtras.h>

namespace WebCore {

// FIXME: Rearchitect this to be more like WidthIterator in Font.cpp.  Have an advance() method
// that does stuff in that method instead of doing everything in the constructor.  Have advance()
// take the GlyphBuffer as an arg so that we don't have to populate the glyph buffer when
// measuring.
UniscribeController::UniscribeController(const Font* font, const TextRun& run)
: m_font(*font)
, m_run(run)
, m_end(run.length())
, m_currentCharacter(0)
, m_runWidthSoFar(0)
, m_computingOffsetPosition(false)
, m_includePartialGlyphs(false)
, m_offsetX(0)
, m_offsetPosition(0)
{
    m_padding = m_run.padding();
    if (!m_padding)
        m_padPerSpace = 0;
    else {
        float numSpaces = 0;
        for (int s = 0; s < m_run.length(); s++)
            if (Font::treatAsSpace(m_run[s]))
                numSpaces++;

        if (numSpaces == 0)
            m_padPerSpace = 0;
        else
            m_padPerSpace = ceilf(m_run.padding() / numSpaces);
    }

    // Null out our uniscribe structs
    resetControlAndState();
}

int UniscribeController::offsetForPosition(int x, bool includePartialGlyphs)
{
    m_computingOffsetPosition = true;
    m_includePartialGlyphs = includePartialGlyphs;
    m_offsetX = x;
    m_offsetPosition = 0;
    advance(m_run.length());
    if (m_computingOffsetPosition) {
        // The point is to the left or to the right of the entire run.
        if (m_offsetX >= m_runWidthSoFar && m_run.ltr() || m_offsetX < 0 && m_run.rtl())
            m_offsetPosition = m_end;
    }
    m_computingOffsetPosition = false;
    return m_offsetPosition;
}

void UniscribeController::advance(unsigned offset, GlyphBuffer* glyphBuffer)
{
    // FIXME: We really want to be using a newer version of Uniscribe that supports the new OpenType
    // functions.  Those functions would allow us to turn off kerning and ligatures.  Without being able
    // to do that, we will have buggy line breaking and metrics when simple and complex text are close
    // together (the complex code path will narrow the text because of kerning and ligatures and then
    // when bidi processing splits into multiple runs, the simple portions will get wider and cause us to
    // spill off the edge of a line).
    if (static_cast<int>(offset) > m_end)
        offset = m_end;

    // Itemize the string.
    const UChar* cp = m_run.data(m_currentCharacter);
    int length = offset - m_currentCharacter;
    if (length <= 0)
        return;

    unsigned baseCharacter = m_currentCharacter;

    // We break up itemization of the string by fontData and (if needed) the use of small caps.

    // FIXME: It's inconsistent that we use logical order when itemizing, since this
    // does not match normal RTL.

    // FIXME: This function should decode surrogate pairs. Currently it makes little difference that
    // it does not because the font cache on Windows does not support non-BMP characters.
    Vector<UChar, 256> smallCapsBuffer;
    if (m_font.isSmallCaps())
        smallCapsBuffer.resize(length);

    unsigned indexOfFontTransition = m_run.rtl() ? length - 1 : 0;
    const UChar* curr = m_run.rtl() ? cp + length  - 1 : cp;
    const UChar* end = m_run.rtl() ? cp - 1 : cp + length;

    const SimpleFontData* fontData;
    const SimpleFontData* nextFontData = m_font.glyphDataForCharacter(*curr, false).fontData;

    UChar newC = 0;

    bool isSmallCaps;
    bool nextIsSmallCaps = m_font.isSmallCaps() && !(U_GET_GC_MASK(*curr) & U_GC_M_MASK) && (newC = u_toupper(*curr)) != *curr;

    if (nextIsSmallCaps)
        smallCapsBuffer[curr - cp] = newC;

    while (true) {
        curr = m_run.rtl() ? curr - 1 : curr + 1;
        if (curr == end)
            break;

        fontData = nextFontData;
        isSmallCaps = nextIsSmallCaps;
        int index = curr - cp;
        UChar c = *curr;

        bool forceSmallCaps = isSmallCaps && (U_GET_GC_MASK(c) & U_GC_M_MASK);
        nextFontData = m_font.glyphDataForCharacter(*curr, false, forceSmallCaps).fontData;
        if (m_font.isSmallCaps()) {
            nextIsSmallCaps = forceSmallCaps || (newC = u_toupper(c)) != c;
            if (nextIsSmallCaps)
                smallCapsBuffer[index] = forceSmallCaps ? c : newC;
        }

        if (nextFontData != fontData || nextIsSmallCaps != isSmallCaps) {
            int itemStart = m_run.rtl() ? index + 1 : indexOfFontTransition;
            int itemLength = m_run.rtl() ? indexOfFontTransition - index : index - indexOfFontTransition;
            m_currentCharacter = baseCharacter + itemStart;
            itemizeShapeAndPlace((isSmallCaps ? smallCapsBuffer.data() : cp) + itemStart, itemLength, fontData, glyphBuffer);
            indexOfFontTransition = index;
        }
    }
    
    int itemLength = m_run.rtl() ? indexOfFontTransition + 1 : length - indexOfFontTransition;
    if (itemLength) {
        int itemStart = m_run.rtl() ? 0 : indexOfFontTransition;
        m_currentCharacter = baseCharacter + itemStart;
        itemizeShapeAndPlace((nextIsSmallCaps ? smallCapsBuffer.data() : cp) + itemStart, itemLength, nextFontData, glyphBuffer);
    }

    m_currentCharacter = baseCharacter + length;
}

void UniscribeController::itemizeShapeAndPlace(const UChar* cp, unsigned length, const SimpleFontData* fontData, GlyphBuffer* glyphBuffer)
{
    // ScriptItemize (in Windows XP versions prior to SP2) can overflow by 1.  This is why there is an extra empty item
    // hanging out at the end of the array
    m_items.resize(6);
    int numItems = 0;
    while (ScriptItemize(cp, length, m_items.size() - 1, &m_control, &m_state, m_items.data(), &numItems) == E_OUTOFMEMORY) {
        m_items.resize(m_items.size() * 2);
        resetControlAndState();
    }
    m_items.resize(numItems + 1);

    if (m_run.rtl()) {
        for (int i = m_items.size() - 2; i >= 0; i--) {
            if (!shapeAndPlaceItem(cp, i, fontData, glyphBuffer))
                return;
        }
    } else {
        for (unsigned i = 0; i < m_items.size() - 1; i++) {
            if (!shapeAndPlaceItem(cp, i, fontData, glyphBuffer))
                return;
        }
    }
}

void UniscribeController::resetControlAndState()
{
    memset(&m_control, 0, sizeof(SCRIPT_CONTROL));
    memset(&m_state, 0, sizeof(SCRIPT_STATE));

    // Set up the correct direction for the run.
    m_state.uBidiLevel = m_run.rtl();
    
    // Lock the correct directional override.
    m_state.fOverrideDirection = m_run.directionalOverride();
}

bool UniscribeController::shapeAndPlaceItem(const UChar* cp, unsigned i, const SimpleFontData* fontData, GlyphBuffer* glyphBuffer)
{
    // Determine the string for this item.
    const UChar* str = cp + m_items[i].iCharPos;
    int len = m_items[i+1].iCharPos - m_items[i].iCharPos;
    SCRIPT_ITEM item = m_items[i];

    // Set up buffers to hold the results of shaping the item.
    Vector<WORD> glyphs;
    Vector<WORD> clusters;
    Vector<SCRIPT_VISATTR> visualAttributes;
    clusters.resize(len);