path: root/gfx/thebes/gfxHarfBuzzShaper.cpp

/* -*- Mode: C++; tab-width: 20; indent-tabs-mode: nil; c-basic-offset: 2 -*- */
/* 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 "nsString.h"
#include "gfxContext.h"
#include "gfxFontConstants.h"
#include "gfxHarfBuzzShaper.h"
#include "gfxFontUtils.h"
#include "gfxTextRun.h"
#include "mozilla/Sprintf.h"
#include "nsUnicodeProperties.h"
#include "nsUnicodeScriptCodes.h"

#include "harfbuzz/hb.h"
#include "harfbuzz/hb-ot.h"

#include "unicode/unorm.h"
#include "unicode/utext.h"

static const UNormalizer2* sNormalizer = nullptr;

#include <algorithm>

#define FloatToFixed(f) (65536 * (f))
#define FixedToFloat(f) ((f) * (1.0 / 65536.0))
// Right shifts of negative (signed) integers are undefined, as are overflows
// when converting unsigned to negative signed integers.
// (If speed were an issue we could make some 2's complement assumptions.)
#define FixedToIntRound(f) \
  ((f) > 0 ? ((32768 + (f)) >> 16) : -((32767 - (f)) >> 16))

using namespace mozilla;           // for AutoSwap_* types
using namespace mozilla::unicode;  // for Unicode property lookup

/*
 * Creation and destruction; on deletion, release any font tables we're holding
 */

gfxHarfBuzzShaper::gfxHarfBuzzShaper(gfxFont* aFont)
    : gfxFontShaper(aFont),
      mHBFont(nullptr),
      mBuffer(nullptr),
      mCallbackData(),
      mKernTable(nullptr),
      mHmtxTable(nullptr),
      mVmtxTable(nullptr),
      mVORGTable(nullptr),
      mLocaTable(nullptr),
      mGlyfTable(nullptr),
      mCmapTable(nullptr),
      mCmapFormat(-1),
      mSubtableOffset(0),
      mUVSTableOffset(0),
      mNumLongHMetrics(0),
      mNumLongVMetrics(0),
      mDefaultVOrg(-1.0),
      mUseFontGetGlyph(aFont->ProvidesGetGlyph()),
      mUseFontGlyphWidths(aFont->ProvidesGlyphWidths()),
      mInitialized(false),
      mVerticalInitialized(false),
      mUseVerticalPresentationForms(false),
      mLoadedLocaGlyf(false),
      mLocaLongOffsets(false) {}

gfxHarfBuzzShaper::~gfxHarfBuzzShaper() {
  // hb_*_destroy functions are safe to call on nullptr
  hb_blob_destroy(mCmapTable);
  hb_blob_destroy(mHmtxTable);
  hb_blob_destroy(mKernTable);
  hb_blob_destroy(mVmtxTable);
  hb_blob_destroy(mVORGTable);
  hb_blob_destroy(mLocaTable);
  hb_blob_destroy(mGlyfTable);
  hb_font_destroy(mHBFont);
  hb_buffer_destroy(mBuffer);
}

#define UNICODE_BMP_LIMIT 0x10000

hb_codepoint_t gfxHarfBuzzShaper::GetNominalGlyph(
    hb_codepoint_t unicode) const {
  hb_codepoint_t gid = 0;

  if (mUseFontGetGlyph) {
    gid = mFont->GetGlyph(unicode, 0);
  } else {
    // we only instantiate a harfbuzz shaper if there's a cmap available
    NS_ASSERTION(mFont->GetFontEntry()->HasCmapTable(),
                 "we cannot be using this font!");

    NS_ASSERTION(mCmapTable && (mCmapFormat > 0) && (mSubtableOffset > 0),
                 "cmap data not correctly set up, expect disaster");

    uint32_t length;
    const uint8_t* data = (const uint8_t*)hb_blob_get_data(mCmapTable, &length);

    switch (mCmapFormat) {
      case 4:
        gid =
            unicode < UNICODE_BMP_LIMIT
                ? gfxFontUtils::MapCharToGlyphFormat4(
                      data + mSubtableOffset, length - mSubtableOffset, unicode)
                : 0;
        break;
      case 10:
        gid = gfxFontUtils::MapCharToGlyphFormat10(data + mSubtableOffset,
                                                   unicode);
        break;
      case 12:
      case 13:
        gid = gfxFontUtils::MapCharToGlyphFormat12or13(data + mSubtableOffset,
                                                       unicode);
        break;
      default:
        NS_WARNING("unsupported cmap format, glyphs will be missing");
        break;
    }
  }

  if (!gid) {
    switch (unicode) {
      case 0xA0:
        // if there's no glyph for &nbsp;, just use the space glyph instead.
        gid = mFont->GetSpaceGlyph();
        break;
      case 0x2010:
      case 0x2011:
        // For Unicode HYPHEN and NON-BREAKING HYPHEN, fall back to the ASCII
        // HYPHEN-MINUS as a substitute.
        gid = GetNominalGlyph('-');
        break;
    }
  }

  return gid;
}

hb_codepoint_t gfxHarfBuzzShaper::GetVariationGlyph(
    hb_codepoint_t unicode, hb_codepoint_t variation_selector) const {
  if (mUseFontGetGlyph) {
    return mFont->GetGlyph(unicode, variation_selector);
  }

  NS_ASSERTION(mFont->GetFontEntry()->HasCmapTable(),
               "we cannot be using this font!");
  NS_ASSERTION(mCmapTable && (mCmapFormat > 0) && (mSubtableOffset > 0),
               "cmap data not correctly set up, expect disaster");

  uint32_t length;
  const uint8_t* data = (const uint8_t*)hb_blob_get_data(mCmapTable, &length);

  if (mUVSTableOffset) {
    hb_codepoint_t gid = gfxFontUtils::MapUVSToGlyphFormat14(
        data + mUVSTableOffset, unicode, variation_selector);
    if (gid) {
      return gid;
    }
  }

  uint32_t compat = gfxFontUtils::GetUVSFallback(unicode, variation_selector);
  if (compat) {
    switch (mCmapFormat) {
      case 4:
        if (compat < UNICODE_BMP_LIMIT) {
          return gfxFontUtils::MapCharToGlyphFormat4(
              data + mSubtableOffset, length - mSubtableOffset, compat);
        }
        break;
      case 10:
        return gfxFontUtils::MapCharToGlyphFormat10(data + mSubtableOffset,
                                                    compat);
        break;
      case 12:
      case 13:
        return gfxFontUtils::MapCharToGlyphFormat12or13(data + mSubtableOffset,
                                                        compat);
        break;
    }
  }

  return 0;
}

static int VertFormsGlyphCompare(const void* aKey, const void* aElem) {
  return int(*((hb_codepoint_t*)(aKey))) - int(*((uint16_t*)(aElem)));
}

// Return a vertical presentation-form codepoint corresponding to the
// given Unicode value, or 0 if no such form is available.
hb_codepoint_t gfxHarfBuzzShaper::GetVerticalPresentationForm(
    hb_codepoint_t aUnicode) {
  static const uint16_t sVerticalForms[][2] = {
      {0x2013, 0xfe32},  // EN DASH
      {0x2014, 0xfe31},  // EM DASH
      {0x2025, 0xfe30},  // TWO DOT LEADER
      {0x2026, 0xfe19},  // HORIZONTAL ELLIPSIS
      {0x3001, 0xfe11},  // IDEOGRAPHIC COMMA
      {0x3002, 0xfe12},  // IDEOGRAPHIC FULL STOP
      {0x3008, 0xfe3f},  // LEFT ANGLE BRACKET
      {0x3009, 0xfe40},  // RIGHT ANGLE BRACKET
      {0x300a, 0xfe3d},  // LEFT DOUBLE ANGLE BRACKET
      {0x300b, 0xfe3e},  // RIGHT DOUBLE ANGLE BRACKET
      {0x300c, 0xfe41},  // LEFT CORNER BRACKET
      {0x300d, 0xfe42},  // RIGHT CORNER BRACKET
      {0x300e, 0xfe43},  // LEFT WHITE CORNER BRACKET
      {0x300f, 0xfe44},  // RIGHT WHITE CORNER BRACKET
      {0x3010, 0xfe3b},  // LEFT BLACK LENTICULAR BRACKET
      {0x3011, 0xfe3c},  // RIGHT BLACK LENTICULAR BRACKET
      {0x3014, 0xfe39},  // LEFT TORTOISE SHELL BRACKET
      {0x3015, 0xfe3a},  // RIGHT TORTOISE SHELL BRACKET
      {0x3016, 0xfe17},  // LEFT WHITE LENTICULAR BRACKET
      {0x3017, 0xfe18},  // RIGHT WHITE LENTICULAR BRACKET
      {0xfe4f, 0xfe34},  // WAVY LOW LINE
      {0xff01, 0xfe15},  // FULLWIDTH EXCLAMATION MARK
      {0xff08, 0xfe35},  // FULLWIDTH LEFT PARENTHESIS
      {0xff09, 0xfe36},  // FULLWIDTH RIGHT PARENTHESIS
      {0xff0c, 0xfe10},  // FULLWIDTH COMMA
      {0xff1a, 0xfe13},  // FULLWIDTH COLON
      {0xff1b, 0xfe14},  // FULLWIDTH SEMICOLON
      {0xff1f, 0xfe16},  // FULLWIDTH QUESTION MARK
      {0xff3b, 0xfe47},  // FULLWIDTH LEFT SQUARE BRACKET
      {0xff3d, 0xfe48},  // FULLWIDTH RIGHT SQUARE BRACKET
      {0xff3f, 0xfe33},  // FULLWIDTH LOW LINE
      {0xff5b, 0xfe37},  // FULLWIDTH LEFT CURLY BRACKET
      {0xff5d, 0xfe38}   // FULLWIDTH RIGHT CURLY BRACKET
  };
  const uint16_t* charPair = static_cast<const uint16_t*>(
      bsearch(&aUnicode, sVerticalForms, ArrayLength(sVerticalForms),
              sizeof(sVerticalForms[0]), VertFormsGlyphCompare));
  return charPair ? charPair[1] : 0;
}

static hb_bool_t HBGetNominalGlyph(hb_font_t* font, void* font_data,
                                   hb_codepoint_t unicode,
                                   hb_codepoint_t* glyph, void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);

  if (fcd->mShaper->UseVerticalPresentationForms()) {
    hb_codepoint_t verticalForm =
        gfxHarfBuzzShaper::GetVerticalPresentationForm(unicode);
    if (verticalForm) {
      *glyph = fcd->mShaper->GetNominalGlyph(verticalForm);
      if (*glyph != 0) {
        return true;
      }
    }
    // fall back to the non-vertical form if we didn't find an alternate
  }

  *glyph = fcd->mShaper->GetNominalGlyph(unicode);
  return *glyph != 0;
}

static hb_bool_t HBGetVariationGlyph(hb_font_t* font, void* font_data,
                                     hb_codepoint_t unicode,
                                     hb_codepoint_t variation_selector,
                                     hb_codepoint_t* glyph, void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);

  if (fcd->mShaper->UseVerticalPresentationForms()) {
    hb_codepoint_t verticalForm =
        gfxHarfBuzzShaper::GetVerticalPresentationForm(unicode);
    if (verticalForm) {
      *glyph =
          fcd->mShaper->GetVariationGlyph(verticalForm, variation_selector);
      if (*glyph != 0) {
        return true;
      }
    }
    // fall back to the non-vertical form if we didn't find an alternate
  }

  *glyph = fcd->mShaper->GetVariationGlyph(unicode, variation_selector);
  return *glyph != 0;
}

// Glyph metrics structures, shared (with appropriate reinterpretation of
// field names) by horizontal and vertical metrics tables.
struct LongMetric {
  AutoSwap_PRUint16 advanceWidth;  // or advanceHeight, when vertical
  AutoSwap_PRInt16 lsb;            // or tsb, when vertical
};

struct GlyphMetrics {
  LongMetric metrics[1];  // actually numberOfLongMetrics
  // the variable-length metrics[] array is immediately followed by:
  //  AutoSwap_PRUint16    leftSideBearing[];
};

hb_position_t gfxHarfBuzzShaper::GetGlyphHAdvance(hb_codepoint_t glyph) const {
  // font did not implement GetGlyphWidth, so get an unhinted value
  // directly from the font tables

  NS_ASSERTION((mNumLongHMetrics > 0) && mHmtxTable != nullptr,
               "font is lacking metrics, we shouldn't be here");

  if (glyph >= uint32_t(mNumLongHMetrics)) {
    glyph = mNumLongHMetrics - 1;
  }

  // glyph must be valid now, because we checked during initialization
  // that mNumLongHMetrics is > 0, and that the metrics table is large enough
  // to contain mNumLongHMetrics records
  const ::GlyphMetrics* metrics = reinterpret_cast<const ::GlyphMetrics*>(
      hb_blob_get_data(mHmtxTable, nullptr));
  return FloatToFixed(mFont->FUnitsToDevUnitsFactor() *
                      uint16_t(metrics->metrics[glyph].advanceWidth));
}

hb_position_t gfxHarfBuzzShaper::GetGlyphVAdvance(hb_codepoint_t glyph) const {
  if (!mVmtxTable) {
    // Must be a "vertical" font that doesn't actually have vertical metrics;
    // use a fixed advance.
    return FloatToFixed(
        mFont->GetMetrics(nsFontMetrics::eVertical).aveCharWidth);
  }

  NS_ASSERTION(mNumLongVMetrics > 0,
               "font is lacking metrics, we shouldn't be here");

  if (glyph >= uint32_t(mNumLongVMetrics)) {
    glyph = mNumLongVMetrics - 1;
  }

  // glyph must be valid now, because we checked during initialization
  // that mNumLongVMetrics is > 0, and that the metrics table is large enough
  // to contain mNumLongVMetrics records
  const ::GlyphMetrics* metrics = reinterpret_cast<const ::GlyphMetrics*>(
      hb_blob_get_data(mVmtxTable, nullptr));
  return FloatToFixed(mFont->FUnitsToDevUnitsFactor() *
                      uint16_t(metrics->metrics[glyph].advanceWidth));
}

/* static */
hb_position_t gfxHarfBuzzShaper::HBGetGlyphHAdvance(hb_font_t* font,
                                                    void* font_data,
                                                    hb_codepoint_t glyph,
                                                    void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);
  const gfxHarfBuzzShaper* shaper = fcd->mShaper;
  if (shaper->mUseFontGlyphWidths) {
    return shaper->GetFont()->GetGlyphWidth(glyph);
  }
  return shaper->GetGlyphHAdvance(glyph);
}

/* static */
hb_position_t gfxHarfBuzzShaper::HBGetGlyphVAdvance(hb_font_t* font,
                                                    void* font_data,
                                                    hb_codepoint_t glyph,
                                                    void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);
  // Currently, we don't offer gfxFont subclasses a method to override this
  // and provide hinted platform-specific vertical advances (analogous to the
  // GetGlyphWidth method for horizontal advances). If that proves necessary,
  // we'll add a new gfxFont method and call it from here.
  //
  // We negate the value from GetGlyphVAdvance here because harfbuzz shapes
  // with a coordinate system where positive is upwards, whereas the inline
  // direction in which glyphs advance is downwards.
  return -fcd->mShaper->GetGlyphVAdvance(glyph);
}

struct VORG {
  AutoSwap_PRUint16 majorVersion;
  AutoSwap_PRUint16 minorVersion;
  AutoSwap_PRInt16 defaultVertOriginY;
  AutoSwap_PRUint16 numVertOriginYMetrics;
};

struct VORGrec {
  AutoSwap_PRUint16 glyphIndex;
  AutoSwap_PRInt16 vertOriginY;
};

/* static */
hb_bool_t gfxHarfBuzzShaper::HBGetGlyphVOrigin(hb_font_t* font, void* font_data,
                                               hb_codepoint_t glyph,
                                               hb_position_t* x,
                                               hb_position_t* y,
                                               void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);
  fcd->mShaper->GetGlyphVOrigin(glyph, x, y);
  return true;
}

void gfxHarfBuzzShaper::GetGlyphVOrigin(hb_codepoint_t aGlyph,
                                        hb_position_t* aX,
                                        hb_position_t* aY) const {
  *aX = 0.5 * (mUseFontGlyphWidths ? mFont->GetGlyphWidth(aGlyph)
                                   : GetGlyphHAdvance(aGlyph));

  if (mVORGTable) {
    // We checked in Initialize() that the VORG table is safely readable,
    // so no length/bounds-check needed here.
    const VORG* vorg =
        reinterpret_cast<const VORG*>(hb_blob_get_data(mVORGTable, nullptr));

    const VORGrec* lo = reinterpret_cast<const VORGrec*>(vorg + 1);
    const VORGrec* hi = lo + uint16_t(vorg->numVertOriginYMetrics);
    const VORGrec* limit = hi;
    while (lo < hi) {
      const VORGrec* mid = lo + (hi - lo) / 2;
      if (uint16_t(mid->glyphIndex) < aGlyph) {
        lo = mid + 1;
      } else {
        hi = mid;
      }
    }

    if (lo < limit && uint16_t(lo->glyphIndex) == aGlyph) {
      *aY = FloatToFixed(GetFont()->FUnitsToDevUnitsFactor() *
                         int16_t(lo->vertOriginY));
    } else {
      *aY = FloatToFixed(GetFont()->FUnitsToDevUnitsFactor() *
                         int16_t(vorg->defaultVertOriginY));
    }
    return;
  }

  if (mVmtxTable) {
    bool emptyGlyf;
    const Glyf* glyf = FindGlyf(aGlyph, &emptyGlyf);
    if (glyf) {
      if (emptyGlyf) {
        *aY = 0;
        return;
      }

      const ::GlyphMetrics* metrics = reinterpret_cast<const ::GlyphMetrics*>(
          hb_blob_get_data(mVmtxTable, nullptr));
      int16_t lsb;
      if (aGlyph < hb_codepoint_t(mNumLongVMetrics)) {
        // Glyph is covered by the first (advance & sidebearing) array
        lsb = int16_t(metrics->metrics[aGlyph].lsb);
      } else {
        // Glyph is covered by the second (sidebearing-only) array
        const AutoSwap_PRInt16* sidebearings =
            reinterpret_cast<const AutoSwap_PRInt16*>(
                &metrics->metrics[mNumLongVMetrics]);
        lsb = int16_t(sidebearings[aGlyph - mNumLongVMetrics]);
      }
      *aY = FloatToFixed(mFont->FUnitsToDevUnitsFactor() *
                         (lsb + int16_t(glyf->yMax)));
      return;
    } else {
      // XXX TODO: not a truetype font; need to get glyph extents
      // via some other API?
      // For now, fall through to default code below.
    }
  }

  if (mDefaultVOrg < 0.0) {
    // XXX should we consider using OS/2 sTypo* metrics if available?

    gfxFontEntry::AutoTable hheaTable(GetFont()->GetFontEntry(),
                                      TRUETYPE_TAG('h', 'h', 'e', 'a'));
    if (hheaTable) {
      uint32_t len;
      const MetricsHeader* hhea = reinterpret_cast<const MetricsHeader*>(
          hb_blob_get_data(hheaTable, &len));
      if (len >= sizeof(MetricsHeader)) {
        // divide up the default advance we're using (1em) in proportion
        // to ascender:descender from the hhea table
        int16_t a = int16_t(hhea->ascender);
        int16_t d = int16_t(hhea->descender);
        mDefaultVOrg = FloatToFixed(GetFont()->GetAdjustedSize() * a / (a - d));
      }
    }

    if (mDefaultVOrg < 0.0) {
      // Last resort, for non-sfnt fonts: get the horizontal metrics and
      // compute a default VOrg from their ascent and descent.
      const gfxFont::Metrics& mtx = mFont->GetHorizontalMetrics();
      gfxFloat advance =
          mFont->GetMetrics(nsFontMetrics::eVertical).aveCharWidth;
      gfxFloat ascent = mtx.emAscent;
      gfxFloat height = ascent + mtx.emDescent;
      // vOrigin that will place the glyph so that its origin is shifted
      // down most of the way within overall (vertical) advance, in
      // proportion to the font ascent as a part of the overall font
      // height.
      mDefaultVOrg = FloatToFixed(advance * ascent / height);
    }
  }

  *aY = mDefaultVOrg;
}

static hb_bool_t HBGetGlyphExtents(hb_font_t* font, void* font_data,
                                   hb_codepoint_t glyph,
                                   hb_glyph_extents_t* extents,
                                   void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);
  return fcd->mShaper->GetGlyphExtents(glyph, extents);
}

// Find the data for glyph ID |aGlyph| in the 'glyf' table, if present.
// Returns null if not found, otherwise pointer to the beginning of the
// glyph's data. Sets aEmptyGlyf true if there is no actual data;
// otherwise, it's guaranteed that we can read at least the bounding box.
const gfxHarfBuzzShaper::Glyf* gfxHarfBuzzShaper::FindGlyf(
    hb_codepoint_t aGlyph, bool* aEmptyGlyf) const {
  if (!mLoadedLocaGlyf) {
    mLoadedLocaGlyf = true;  // only try this once; if it fails, this
                             // isn't a truetype font
    gfxFontEntry* entry = mFont->GetFontEntry();
    uint32_t len;
    gfxFontEntry::AutoTable headTable(entry, TRUETYPE_TAG('h', 'e', 'a', 'd'));
    if (!headTable) {
      return nullptr;
    }
    const HeadTable* head =
        reinterpret_cast<const HeadTable*>(hb_blob_get_data(headTable, &len));
    if (len < sizeof(HeadTable)) {
      return nullptr;
    }
    mLocaLongOffsets = int16_t(head->indexToLocFormat) > 0;
    mLocaTable = entry->GetFontTable(TRUETYPE_TAG('l', 'o', 'c', 'a'));
    mGlyfTable = entry->GetFontTable(TRUETYPE_TAG('g', 'l', 'y', 'f'));
  }

  if (!mLocaTable || !mGlyfTable) {
    // it's not a truetype font
    return nullptr;
  }

  uint32_t offset;  // offset of glyph record in the 'glyf' table
  uint32_t len;
  const char* data = hb_blob_get_data(mLocaTable, &len);
  if (mLocaLongOffsets) {
    if ((aGlyph + 1) * sizeof(AutoSwap_PRUint32) > len) {
      return nullptr;
    }
    const AutoSwap_PRUint32* offsets =
        reinterpret_cast<const AutoSwap_PRUint32*>(data);
    offset = offsets[aGlyph];
    *aEmptyGlyf = (offset == uint16_t(offsets[aGlyph + 1]));
  } else {
    if ((aGlyph + 1) * sizeof(AutoSwap_PRUint16) > len) {
      return nullptr;
    }
    const AutoSwap_PRUint16* offsets =
        reinterpret_cast<const AutoSwap_PRUint16*>(data);
    offset = uint16_t(offsets[aGlyph]);
    *aEmptyGlyf = (offset == uint16_t(offsets[aGlyph + 1]));
    offset *= 2;
  }

  data = hb_blob_get_data(mGlyfTable, &len);
  if (offset + sizeof(Glyf) > len) {
    return nullptr;
  }

  return reinterpret_cast<const Glyf*>(data + offset);
}

hb_bool_t gfxHarfBuzzShaper::GetGlyphExtents(
    hb_codepoint_t aGlyph, hb_glyph_extents_t* aExtents) const {
  bool emptyGlyf;
  const Glyf* glyf = FindGlyf(aGlyph, &emptyGlyf);
  if (!glyf) {
    // TODO: for non-truetype fonts, get extents some other way?
    return false;
  }

  if (emptyGlyf) {
    aExtents->x_bearing = 0;
    aExtents->y_bearing = 0;
    aExtents->width = 0;
    aExtents->height = 0;
    return true;
  }

  double f = mFont->FUnitsToDevUnitsFactor();
  aExtents->x_bearing = FloatToFixed(int16_t(glyf->xMin) * f);
  aExtents->width =
      FloatToFixed((int16_t(glyf->xMax) - int16_t(glyf->xMin)) * f);

  // Our y-coordinates are positive-downwards, whereas harfbuzz assumes
  // positive-upwards; hence the apparently-reversed subtractions here.
  aExtents->y_bearing = FloatToFixed(int16_t(glyf->yMax) * f -
                                     mFont->GetHorizontalMetrics().emAscent);
  aExtents->height =
      FloatToFixed((int16_t(glyf->yMin) - int16_t(glyf->yMax)) * f);

  return true;
}

static hb_bool_t HBGetContourPoint(hb_font_t* font, void* font_data,
                                   unsigned int point_index,
                                   hb_codepoint_t glyph, hb_position_t* x,
                                   hb_position_t* y, void* user_data) {
  /* not yet implemented - no support for used of hinted contour points
     to fine-tune anchor positions in GPOS AnchorFormat2 */
  return false;
}

struct KernHeaderFmt0 {
  AutoSwap_PRUint16 nPairs;
  AutoSwap_PRUint16 searchRange;
  AutoSwap_PRUint16 entrySelector;
  AutoSwap_PRUint16 rangeShift;
};

struct KernPair {
  AutoSwap_PRUint16 left;
  AutoSwap_PRUint16 right;
  AutoSwap_PRInt16 value;
};

// Find a kern pair in a Format 0 subtable.
// The aSubtable parameter points to the subtable itself, NOT its header,
// as the header structure differs between Windows and Mac (v0 and v1.0)
// versions of the 'kern' table.
// aSubtableLen is the length of the subtable EXCLUDING its header.
// If the pair <aFirstGlyph,aSecondGlyph> is found, the kerning value is
// added to aValue, so that multiple subtables can accumulate a total
// kerning value for a given pair.
static void GetKernValueFmt0(const void* aSubtable, uint32_t aSubtableLen,
                             uint16_t aFirstGlyph, uint16_t aSecondGlyph,
                             int32_t& aValue, bool aIsOverride = false,
                             bool aIsMinimum = false) {
  const KernHeaderFmt0* hdr =
      reinterpret_cast<const KernHeaderFmt0*>(aSubtable);

  const KernPair* lo = reinterpret_cast<const KernPair*>(hdr + 1);
  const KernPair* hi = lo + uint16_t(hdr->nPairs);
  const KernPair* limit = hi;

  if (reinterpret_cast<const char*>(aSubtable) + aSubtableLen <
      reinterpret_cast<const char*>(hi)) {
    // subtable is not large enough to contain the claimed number
    // of kern pairs, so just ignore it
    return;
  }

#define KERN_PAIR_KEY(l, r) (uint32_t((uint16_t(l) << 16) + uint16_t(r)))

  uint32_t key = KERN_PAIR_KEY(aFirstGlyph, aSecondGlyph);
  while (lo < hi) {
    const KernPair* mid = lo + (hi - lo) / 2;
    if (KERN_PAIR_KEY(mid->left, mid->right) < key) {
      lo = mid + 1;
    } else {
      hi = mid;
    }
  }

  if (lo < limit && KERN_PAIR_KEY(lo->left, lo->right) == key) {
    if (aIsOverride) {
      aValue = int16_t(lo->value);
    } else if (aIsMinimum) {
      aValue = std::max(aValue, int32_t(lo->value));
    } else {
      aValue += int16_t(lo->value);
    }
  }
}

// Get kerning value from Apple (version 1.0) kern table,
// subtable format 2 (simple N x M array of kerning values)

// See http://developer.apple.com/fonts/TTRefMan/RM06/Chap6kern.html
// for details of version 1.0 format 2 subtable.

struct KernHeaderVersion1Fmt2 {
  KernTableSubtableHeaderVersion1 header;
  AutoSwap_PRUint16 rowWidth;
  AutoSwap_PRUint16 leftOffsetTable;
  AutoSwap_PRUint16 rightOffsetTable;
  AutoSwap_PRUint16 array;
};

struct KernClassTableHdr {
  AutoSwap_PRUint16 firstGlyph;
  AutoSwap_PRUint16 nGlyphs;
  AutoSwap_PRUint16 offsets[1];  // actually an array of nGlyphs entries
};

static int16_t GetKernValueVersion1Fmt2(const void* aSubtable,
                                        uint32_t aSubtableLen,
                                        uint16_t aFirstGlyph,
                                        uint16_t aSecondGlyph) {
  if (aSubtableLen < sizeof(KernHeaderVersion1Fmt2)) {
    return 0;
  }

  const char* base = reinterpret_cast<const char*>(aSubtable);
  const char* subtableEnd = base + aSubtableLen;

  const KernHeaderVersion1Fmt2* h =
      reinterpret_cast<const KernHeaderVersion1Fmt2*>(aSubtable);
  uint32_t offset = h->array;

  const KernClassTableHdr* leftClassTable =
      reinterpret_cast<const KernClassTableHdr*>(base +
                                                 uint16_t(h->leftOffsetTable));
  if (reinterpret_cast<const char*>(leftClassTable) +
          sizeof(KernClassTableHdr) >
      subtableEnd) {
    return 0;
  }
  if (aFirstGlyph >= uint16_t(leftClassTable->firstGlyph)) {
    aFirstGlyph -= uint16_t(leftClassTable->firstGlyph);
    if (aFirstGlyph < uint16_t(leftClassTable->nGlyphs)) {
      if (reinterpret_cast<const char*>(leftClassTable) +
              sizeof(KernClassTableHdr) + aFirstGlyph * sizeof(uint16_t) >=
          subtableEnd) {
        return 0;
      }
      offset = uint16_t(leftClassTable->offsets[aFirstGlyph]);
    }
  }

  const KernClassTableHdr* rightClassTable =
      reinterpret_cast<const KernClassTableHdr*>(base +
                                                 uint16_t(h->rightOffsetTable));
  if (reinterpret_cast<const char*>(rightClassTable) +
          sizeof(KernClassTableHdr) >
      subtableEnd) {
    return 0;
  }
  if (aSecondGlyph >= uint16_t(rightClassTable->firstGlyph)) {
    aSecondGlyph -= uint16_t(rightClassTable->firstGlyph);
    if (aSecondGlyph < uint16_t(rightClassTable->nGlyphs)) {
      if (reinterpret_cast<const char*>(rightClassTable) +
              sizeof(KernClassTableHdr) + aSecondGlyph * sizeof(uint16_t) >=
          subtableEnd) {
        return 0;
      }
      offset += uint16_t(rightClassTable->offsets[aSecondGlyph]);
    }
  }

  const AutoSwap_PRInt16* pval =
      reinterpret_cast<const AutoSwap_PRInt16*>(base + offset);
  if (reinterpret_cast<const char*>(pval + 1) >= subtableEnd) {
    return 0;
  }
  return *pval;
}

// Get kerning value from Apple (version 1.0) kern table,
// subtable format 3 (simple N x M array of kerning values)

// See http://developer.apple.com/fonts/TTRefMan/RM06/Chap6kern.html
// for details of version 1.0 format 3 subtable.

struct KernHeaderVersion1Fmt3 {
  KernTableSubtableHeaderVersion1 header;
  AutoSwap_PRUint16 glyphCount;
  uint8_t kernValueCount;
  uint8_t leftClassCount;
  uint8_t rightClassCount;
  uint8_t flags;
};

static int16_t GetKernValueVersion1Fmt3(const void* aSubtable,
                                        uint32_t aSubtableLen,
                                        uint16_t aFirstGlyph,
                                        uint16_t aSecondGlyph) {
  // check that we can safely read the header fields
  if (aSubtableLen < sizeof(KernHeaderVersion1Fmt3)) {
    return 0;
  }

  const KernHeaderVersion1Fmt3* hdr =
      reinterpret_cast<const KernHeaderVersion1Fmt3*>(aSubtable);
  if (hdr->flags != 0) {
    return 0;
  }

  uint16_t glyphCount = hdr->glyphCount;

  // check that table is large enough for the arrays
  if (sizeof(KernHeaderVersion1Fmt3) + hdr->kernValueCount * sizeof(int16_t) +
          glyphCount + glyphCount + hdr->leftClassCount * hdr->rightClassCount >
      aSubtableLen) {
    return 0;
  }

  if (aFirstGlyph >= glyphCount || aSecondGlyph >= glyphCount) {
    // glyphs are out of range for the class tables
    return 0;
  }

  // get pointers to the four arrays within the subtable
  const AutoSwap_PRInt16* kernValue =
      reinterpret_cast<const AutoSwap_PRInt16*>(hdr + 1);
  const uint8_t* leftClass =
      reinterpret_cast<const uint8_t*>(kernValue + hdr->kernValueCount);
  const uint8_t* rightClass = leftClass + glyphCount;
  const uint8_t* kernIndex = rightClass + glyphCount;

  uint8_t lc = leftClass[aFirstGlyph];
  uint8_t rc = rightClass[aSecondGlyph];
  if (lc >= hdr->leftClassCount || rc >= hdr->rightClassCount) {
    return 0;
  }

  uint8_t ki = kernIndex[leftClass[aFirstGlyph] * hdr->rightClassCount +
                         rightClass[aSecondGlyph]];
  if (ki >= hdr->kernValueCount) {
    return 0;
  }

  return kernValue[ki];
}

#define KERN0_COVERAGE_HORIZONTAL 0x0001
#define KERN0_COVERAGE_MINIMUM 0x0002
#define KERN0_COVERAGE_CROSS_STREAM 0x0004
#define KERN0_COVERAGE_OVERRIDE 0x0008
#define KERN0_COVERAGE_RESERVED 0x00F0

#define KERN1_COVERAGE_VERTICAL 0x8000
#define KERN1_COVERAGE_CROSS_STREAM 0x4000
#define KERN1_COVERAGE_VARIATION 0x2000
#define KERN1_COVERAGE_RESERVED 0x1F00

hb_position_t gfxHarfBuzzShaper::GetHKerning(uint16_t aFirstGlyph,
                                             uint16_t aSecondGlyph) const {
  // We want to ignore any kern pairs involving <space>, because we are
  // handling words in isolation, the only space characters seen here are
  // the ones artificially added by the textRun code.
  uint32_t spaceGlyph = mFont->GetSpaceGlyph();
  if (aFirstGlyph == spaceGlyph || aSecondGlyph == spaceGlyph) {
    return 0;
  }

  if (!mKernTable) {
    mKernTable =
        mFont->GetFontEntry()->GetFontTable(TRUETYPE_TAG('k', 'e', 'r', 'n'));
    if (!mKernTable) {
      mKernTable = hb_blob_get_empty();
    }
  }

  uint32_t len;
  const char* base = hb_blob_get_data(mKernTable, &len);
  if (len < sizeof(KernTableVersion0)) {
    return 0;
  }
  int32_t value = 0;

  // First try to interpret as "version 0" kern table
  // (see http://www.microsoft.com/typography/otspec/kern.htm)
  const KernTableVersion0* kern0 =
      reinterpret_cast<const KernTableVersion0*>(base);
  if (uint16_t(kern0->version) == 0) {
    uint16_t nTables = kern0->nTables;
    uint32_t offs = sizeof(KernTableVersion0);
    for (uint16_t i = 0; i < nTables; ++i) {
      if (offs + sizeof(KernTableSubtableHeaderVersion0) > len) {
        break;
      }
      const KernTableSubtableHeaderVersion0* st0 =
          reinterpret_cast<const KernTableSubtableHeaderVersion0*>(base + offs);
      uint16_t subtableLen = uint16_t(st0->length);
      if (offs + subtableLen > len) {
        break;
      }
      offs += subtableLen;
      uint16_t coverage = st0->coverage;
      if (!(coverage & KERN0_COVERAGE_HORIZONTAL)) {
        // we only care about horizontal kerning (for now)
        continue;
      }
      if (coverage & (KERN0_COVERAGE_CROSS_STREAM | KERN0_COVERAGE_RESERVED)) {
        // we don't support cross-stream kerning, and
        // reserved bits should be zero;
        // ignore the subtable if not
        continue;
      }
      uint8_t format = (coverage >> 8);
      switch (format) {
        case 0:
          GetKernValueFmt0(st0 + 1, subtableLen - sizeof(*st0), aFirstGlyph,
                           aSecondGlyph, value,
                           (coverage & KERN0_COVERAGE_OVERRIDE) != 0,
                           (coverage & KERN0_COVERAGE_MINIMUM) != 0);
          break;
        default:
          // TODO: implement support for other formats,
          // if they're ever used in practice
#if DEBUG
        {
          char buf[1024];
          SprintfLiteral(buf,
                         "unknown kern subtable in %s: "
                         "ver 0 format %d\n",
                         mFont->GetName().get(), format);
          NS_WARNING(buf);
        }
#endif
        break;
      }
    }
  } else {
    // It wasn't a "version 0" table; check if it is Apple version 1.0
    // (see http://developer.apple.com/fonts/TTRefMan/RM06/Chap6kern.html)
    const KernTableVersion1* kern1 =
        reinterpret_cast<const KernTableVersion1*>(base);
    if (uint32_t(kern1->version) == 0x00010000) {
      uint32_t nTables = kern1->nTables;
      uint32_t offs = sizeof(KernTableVersion1);
      for (uint32_t i = 0; i < nTables; ++i) {
        if (offs + sizeof(KernTableSubtableHeaderVersion1) > len) {
          break;
        }
        const KernTableSubtableHeaderVersion1* st1 =
            reinterpret_cast<const KernTableSubtableHeaderVersion1*>(base +
                                                                     offs);
        uint32_t subtableLen = uint32_t(st1->length);
        offs += subtableLen;
        uint16_t coverage = st1->coverage;
        if (coverage & (KERN1_COVERAGE_VERTICAL | KERN1_COVERAGE_CROSS_STREAM |
                        KERN1_COVERAGE_VARIATION | KERN1_COVERAGE_RESERVED)) {
          // we only care about horizontal kerning (for now),
          // we don't support cross-stream kerning,
          // we don't support variations,
          // reserved bits should be zero;
          // ignore the subtable if not
          continue;
        }
        uint8_t format = (coverage & 0xff);
        switch (format) {
          case 0:
            GetKernValueFmt0(st1 + 1, subtableLen - sizeof(*st1), aFirstGlyph,
                             aSecondGlyph, value);
            break;
          case 2:
            value = GetKernValueVersion1Fmt2(st1, subtableLen, aFirstGlyph,
                                             aSecondGlyph);
            break;
          case 3:
            value = GetKernValueVersion1Fmt3(st1, subtableLen, aFirstGlyph,
                                             aSecondGlyph);
            break;
          default:
            // TODO: implement support for other formats.
            // Note that format 1 cannot be supported here,
            // as it requires the full glyph array to run the FSM,
            // not just the current glyph pair.
#if DEBUG
          {
            char buf[1024];
            SprintfLiteral(buf,
                           "unknown kern subtable in %s: "
                           "ver 0 format %d\n",
                           mFont->GetName().get(), format);
            NS_WARNING(buf);
          }
#endif
          break;
        }
      }
    }
  }

  if (value != 0) {
    return FloatToFixed(mFont->FUnitsToDevUnitsFactor() * value);
  }
  return 0;
}

static hb_position_t HBGetHKerning(hb_font_t* font, void* font_data,
                                   hb_codepoint_t first_glyph,
                                   hb_codepoint_t second_glyph,
                                   void* user_data) {
  const gfxHarfBuzzShaper::FontCallbackData* fcd =
      static_cast<const gfxHarfBuzzShaper::FontCallbackData*>(font_data);
  return fcd->mShaper->GetHKerning(first_glyph, second_glyph);
}

/*
 * HarfBuzz unicode property callbacks
 */

static hb_codepoint_t HBGetMirroring(hb_unicode_funcs_t* ufuncs,
                                     hb_codepoint_t aCh, void* user_data) {
  return GetMirroredChar(aCh);
}

static hb_unicode_general_category_t HBGetGeneralCategory(
    hb_unicode_funcs_t* ufuncs, hb_codepoint_t aCh, void* user_data) {
  return hb_unicode_general_category_t(GetGeneralCategory(aCh));
}

static hb_script_t HBGetScript(hb_unicode_funcs_t* ufuncs, hb_codepoint_t aCh,
                               void* user_data) {
  return hb_script_t(GetScriptTagForCode(GetScriptCode(aCh)));
}

static hb_unicode_combining_class_t HBGetCombiningClass(
    hb_unicode_funcs_t* ufuncs, hb_codepoint_t aCh, void* user_data) {
  return hb_unicode_combining_class_t(GetCombiningClass(aCh));
}

// Hebrew presentation forms with dagesh, for characters 0x05D0..0x05EA;
// note that some letters do not have a dagesh presForm encoded
static const char16_t sDageshForms[0x05EA - 0x05D0 + 1] = {
    0xFB30,  // ALEF
    0xFB31,  // BET
    0xFB32,  // GIMEL
    0xFB33,  // DALET
    0xFB34,  // HE
    0xFB35,  // VAV
    0xFB36,  // ZAYIN
    0,       // HET
    0xFB38,  // TET
    0xFB39,  // YOD
    0xFB3A,  // FINAL KAF
    0xFB3B,  // KAF
    0xFB3C,  // LAMED
    0,       // FINAL MEM
    0xFB3E,  // MEM
    0,       // FINAL NUN
    0xFB40,  // NUN
    0xFB41,  // SAMEKH
    0,       // AYIN
    0xFB43,  // FINAL PE
    0xFB44,  // PE
    0,       // FINAL TSADI
    0xFB46,  // TSADI
    0xFB47,  // QOF
    0xFB48,  // RESH
    0xFB49,  // SHIN
    0xFB4A   // TAV
};

static hb_bool_t HBUnicodeCompose(hb_unicode_funcs_t* ufuncs, hb_codepoint_t a,
                                  hb_codepoint_t b, hb_codepoint_t* ab,
                                  void* user_data) {
  if (sNormalizer) {
    UChar32 ch = unorm2_composePair(sNormalizer, a, b);
    if (ch >= 0) {
      *ab = ch;
      return true;
    }
  }

  return false;
}

static hb_bool_t HBUnicodeDecompose(hb_unicode_funcs_t* ufuncs,
                                    hb_codepoint_t ab, hb_codepoint_t* a,
                                    hb_codepoint_t* b, void* user_data) {
#ifdef MOZ_WIDGET_ANDROID
  // Hack for the SamsungDevanagari font, bug 1012365:
  // support U+0972 by decomposing it.
  if (ab == 0x0972) {
    *a = 0x0905;
    *b = 0x0945;
    return true;
  }
#endif

  if (!sNormalizer) {
    return false;
  }

  // Canonical decompositions are never more than two characters,
  // or a maximum of 4 utf-16 code units.
  const unsigned MAX_DECOMP_LENGTH = 4;

  UErrorCode error = U_ZERO_ERROR;
  UChar decomp[MAX_DECOMP_LENGTH];
  int32_t len = unorm2_getRawDecomposition(sNormalizer, ab, decomp,
                                           MAX_DECOMP_LENGTH, &error);
  if (U_FAILURE(error) || len < 0) {
    return false;
  }

  UText text = UTEXT_INITIALIZER;
  utext_openUChars(&text, decomp, len, &error);
  NS_ASSERTION(U_SUCCESS(error), "UText failure?");

  UChar32 ch = UTEXT_NEXT32(&text);
  if (ch != U_SENTINEL) {
    *a = ch;
  }
  ch = UTEXT_NEXT32(&text);
  if (ch != U_SENTINEL) {
    *b = ch;
  }
  utext_close(&text);

  return *b != 0 || *a != ab;
}

static void AddOpenTypeFeature(const uint32_t& aTag, uint32_t& aValue,
                               void* aUserArg) {
  nsTArray<hb_feature_t>* features =
      static_cast<nsTArray<hb_feature_t>*>(aUserArg);

  hb_feature_t feat = {0, 0, 0, UINT_MAX};
  feat.tag = aTag;
  feat.value = aValue;
  features->AppendElement(feat);
}

/*
 * gfxFontShaper override to initialize the text run using HarfBuzz
 */

static hb_font_funcs_t* sHBFontFuncs = nullptr;
static hb_font_funcs_t* sNominalGlyphFunc = nullptr;
static hb_unicode_funcs_t* sHBUnicodeFuncs = nullptr;
static const hb_script_t sMathScript =
    hb_ot_tag_to_script(HB_TAG('m', 'a', 't', 'h'));

bool gfxHarfBuzzShaper::Initialize() {
  if (mInitialized) {
    return mHBFont != nullptr;
  }
  mInitialized = true;
  mCallbackData.mShaper = this;

  if (!sHBFontFuncs) {
    // static function callback pointers, initialized by the first
    // harfbuzz shaper used
    sHBFontFuncs = hb_font_funcs_create();
    hb_font_funcs_set_nominal_glyph_func(sHBFontFuncs, HBGetNominalGlyph,
                                         nullptr, nullptr);
    hb_font_funcs_set_variation_glyph_func(sHBFontFuncs, HBGetVariationGlyph,
                                           nullptr, nullptr);
    hb_font_funcs_set_glyph_h_advance_func(sHBFontFuncs, HBGetGlyphHAdvance,
                                           nullptr, nullptr);
    hb_font_funcs_set_glyph_v_advance_func(sHBFontFuncs, HBGetGlyphVAdvance,
                                           nullptr, nullptr);
    hb_font_funcs_set_glyph_v_origin_func(sHBFontFuncs, HBGetGlyphVOrigin,
                                          nullptr, nullptr);
    hb_font_funcs_set_glyph_extents_func(sHBFontFuncs, HBGetGlyphExtents,
                                         nullptr, nullptr);
    hb_font_funcs_set_glyph_contour_point_func(sHBFontFuncs, HBGetContourPoint,
                                               nullptr, nullptr);
    hb_font_funcs_set_glyph_h_kerning_func(sHBFontFuncs, HBGetHKerning, nullptr,
                                           nullptr);
    hb_font_funcs_make_immutable(sHBFontFuncs);

    sNominalGlyphFunc = hb_font_funcs_create();
    hb_font_funcs_set_nominal_glyph_func(sNominalGlyphFunc, HBGetNominalGlyph,
                                         nullptr, nullptr);
    hb_font_funcs_make_immutable(sNominalGlyphFunc);

    sHBUnicodeFuncs = hb_unicode_funcs_create(hb_unicode_funcs_get_empty());
    hb_unicode_funcs_set_mirroring_func(sHBUnicodeFuncs, HBGetMirroring,
                                        nullptr, nullptr);
    hb_unicode_funcs_set_script_func(sHBUnicodeFuncs, HBGetScript, nullptr,
                                     nullptr);
    hb_unicode_funcs_set_general_category_func(
        sHBUnicodeFuncs, HBGetGeneralCategory, nullptr, nullptr);
    hb_unicode_funcs_set_combining_class_func(
        sHBUnicodeFuncs, HBGetCombiningClass, nullptr, nullptr);
    hb_unicode_funcs_set_compose_func(sHBUnicodeFuncs, HBUnicodeCompose,
                                      nullptr, nullptr);
    hb_unicode_funcs_set_decompose_func(sHBUnicodeFuncs, HBUnicodeDecompose,
                                        nullptr, nullptr);
    hb_unicode_funcs_make_immutable(sHBUnicodeFuncs);

    UErrorCode error = U_ZERO_ERROR;
    sNormalizer = unorm2_getNFCInstance(&error);
    MOZ_ASSERT(U_SUCCESS(error), "failed to get ICU normalizer");
  }

  gfxFontEntry* entry = mFont->GetFontEntry();
  if (!mUseFontGetGlyph) {
    // get the cmap table and find offset to our subtable
    mCmapTable = entry->GetFontTable(TRUETYPE_TAG('c', 'm', 'a', 'p'));
    if (!mCmapTable) {
      NS_WARNING("failed to load cmap, glyphs will be missing");
      return false;
    }
    uint32_t len;
    const uint8_t* data = (const uint8_t*)hb_blob_get_data(mCmapTable, &len);
    mCmapFormat = gfxFontUtils::FindPreferredSubtable(
        data, len, &mSubtableOffset, &mUVSTableOffset);
    if (mCmapFormat <= 0) {
      return false;
    }
  }

  if (!mUseFontGlyphWidths) {
    // If font doesn't implement GetGlyphWidth, we will be reading
    // the metrics table directly, so make sure we can load it.
    if (!LoadHmtxTable()) {
      return false;
    }
  }

  mBuffer = hb_buffer_create();
  hb_buffer_set_unicode_funcs(mBuffer, sHBUnicodeFuncs);
  hb_buffer_set_cluster_level(mBuffer,
                              HB_BUFFER_CLUSTER_LEVEL_MONOTONE_CHARACTERS);

  auto* funcs =
      mFont->GetFontEntry()->HasFontTable(TRUETYPE_TAG('C', 'F', 'F', ' '))
          ? sNominalGlyphFunc
          : sHBFontFuncs;
  mHBFont = CreateHBFont(mFont, funcs, &mCallbackData);

  return true;
}

hb_font_t* gfxHarfBuzzShaper::CreateHBFont(gfxFont* aFont,
                                           hb_font_funcs_t* aFontFuncs,
                                           FontCallbackData* aCallbackData) {
  hb_face_t* hbFace = aFont->GetFontEntry()->GetHBFace();
  hb_font_t* result = hb_font_create(hbFace);
  hb_face_destroy(hbFace);

  if (aFontFuncs && aCallbackData) {
    if (aFontFuncs == sNominalGlyphFunc) {
      hb_font_t* subfont = hb_font_create_sub_font(result);
      hb_font_destroy(result);
      result = subfont;
    }
    hb_font_set_funcs(result, aFontFuncs, aCallbackData, nullptr);
  }
  hb_font_set_ppem(result, aFont->GetAdjustedSize(), aFont->GetAdjustedSize());
  uint32_t scale = FloatToFixed(aFont->GetAdjustedSize());  // 16.16 fixed-point
  hb_font_set_scale(result, scale, scale);

  AutoTArray<gfxFontVariation, 8> vars;
  aFont->GetFontEntry()->GetVariationsForStyle(vars, *aFont->GetStyle());
  if (vars.Length() > 0) {
    // Fortunately, the hb_variation_t struct is compatible with our
    // gfxFontVariation, so we can simply cast here.
    static_assert(
        sizeof(gfxFontVariation) == sizeof(hb_variation_t) &&
            offsetof(gfxFontVariation, mTag) == offsetof(hb_variation_t, tag) &&
            offsetof(gfxFontVariation, mValue) ==
                offsetof(hb_variation_t, value),
        "Gecko vs HarfBuzz struct mismatch!");
    auto hbVars = reinterpret_cast<const hb_variation_t*>(vars.Elements());
    hb_font_set_variations(result, hbVars, vars.Length());
  }

  return result;
}

bool gfxHarfBuzzShaper::LoadHmtxTable() {
  // Read mNumLongHMetrics from metrics-head table without caching its
  // blob, and preload/cache the metrics table.
  gfxFontEntry* entry = mFont->GetFontEntry();
  gfxFontEntry::AutoTable hheaTable(entry, TRUETYPE_TAG('h', 'h', 'e', 'a'));
  if (hheaTable) {
    uint32_t len;
    const MetricsHeader* hhea = reinterpret_cast<const MetricsHeader*>(
        hb_blob_get_data(hheaTable, &len));
    if (len >= sizeof(MetricsHeader)) {
      mNumLongHMetrics = hhea->numOfLongMetrics;
      if (mNumLongHMetrics > 0 && int16_t(hhea->metricDataFormat) == 0) {
        // no point reading metrics if number of entries is zero!
        // in that case, we won't be able to use this font
        // (this method will return FALSE below if mHmtxTable
        // is null)
        mHmtxTable = entry->GetFontTable(TRUETYPE_TAG('h', 'm', 't', 'x'));
        if (mHmtxTable && hb_blob_get_length(mHmtxTable) <
                              mNumLongHMetrics * sizeof(LongMetric)) {
          // metrics table is not large enough for the claimed
          // number of entries: invalid, do not use.
          hb_blob_destroy(mHmtxTable);
          mHmtxTable = nullptr;
        }
      }
    }
  }
  if (!mHmtxTable) {
    return false;
  }
  return true;
}

void gfxHarfBuzzShaper::InitializeVertical() {
  // We only do this once. If we don't have a mHmtxTable after that,
  // we'll be making up fallback metrics.
  if (mVerticalInitialized) {
    return;
  }
  mVerticalInitialized = true;

  if (!mHmtxTable) {
    if (!LoadHmtxTable()) {
      return;
    }
  }

  // Load vertical metrics if present in the font; if not, we'll synthesize
  // vertical glyph advances based on (horizontal) ascent/descent metrics.
  gfxFontEntry* entry = mFont->GetFontEntry();
  gfxFontEntry::AutoTable vheaTable(entry, TRUETYPE_TAG('v', 'h', 'e', 'a'));
  if (vheaTable) {
    uint32_t len;
    const MetricsHeader* vhea = reinterpret_cast<const MetricsHeader*>(
        hb_blob_get_data(vheaTable, &len));
    if (len >= sizeof(MetricsHeader)) {
      mNumLongVMetrics = vhea->numOfLongMetrics;
      gfxFontEntry::AutoTable maxpTable(entry,
                                        TRUETYPE_TAG('m', 'a', 'x', 'p'));
      int numGlyphs = -1;  // invalid if we fail to read 'maxp'
      if (maxpTable &&
          hb_blob_get_length(maxpTable) >= sizeof(MaxpTableHeader)) {
        const MaxpTableHeader* maxp = reinterpret_cast<const MaxpTableHeader*>(
            hb_blob_get_data(maxpTable, nullptr));
        numGlyphs = uint16_t(maxp->numGlyphs);
      }
      if (mNumLongVMetrics > 0 && mNumLongVMetrics <= numGlyphs &&
          int16_t(vhea->metricDataFormat) == 0) {
        mVmtxTable = entry->GetFontTable(TRUETYPE_TAG('v', 'm', 't', 'x'));
        if (mVmtxTable &&
            hb_blob_get_length(mVmtxTable) <
                mNumLongVMetrics * sizeof(LongMetric) +
                    (numGlyphs - mNumLongVMetrics) * sizeof(int16_t)) {
          // metrics table is not large enough for the claimed
          // number of entries: invalid, do not use.
          hb_blob_destroy(mVmtxTable);
          mVmtxTable = nullptr;
        }
      }
    }
  }

  // For CFF fonts only, load a VORG table if present.
  if (entry->HasFontTable(TRUETYPE_TAG('C', 'F', 'F', ' '))) {
    mVORGTable = entry->GetFontTable(TRUETYPE_TAG('V', 'O', 'R', 'G'));
    if (mVORGTable) {
      uint32_t len;
      const VORG* vorg =
          reinterpret_cast<const VORG*>(hb_blob_get_data(mVORGTable, &len));
      if (len < sizeof(VORG) || uint16_t(vorg->majorVersion) != 1 ||
          uint16_t(vorg->minorVersion) != 0 ||
          len < sizeof(VORG) +
                    uint16_t(vorg->numVertOriginYMetrics) * sizeof(VORGrec)) {
        // VORG table is an unknown version, or not large enough
        // to be valid -- discard it.
        NS_WARNING("discarding invalid VORG table");
        hb_blob_destroy(mVORGTable);
        mVORGTable = nullptr;
      }
    }
  }
}

bool gfxHarfBuzzShaper::ShapeText(DrawTarget* aDrawTarget,
                                  const char16_t* aText, uint32_t aOffset,
                                  uint32_t aLength, Script aScript,
                                  nsAtom* aLanguage, bool aVertical,
                                  RoundingFlags aRounding,
                                  gfxShapedText* aShapedText) {
  mUseVerticalPresentationForms = false;

  if (!Initialize()) {
    return false;
  }

  if (aVertical) {
    InitializeVertical();
    if (!mFont->GetFontEntry()->SupportsOpenTypeFeature(
            aScript, HB_TAG('v', 'e', 'r', 't'))) {
      mUseVerticalPresentationForms = true;
    }
  }

  const gfxFontStyle* style = mFont->GetStyle();

  // determine whether petite-caps falls back to small-caps
  bool addSmallCaps = false;
  if (style->variantCaps != NS_FONT_VARIANT_CAPS_NORMAL) {
    switch (style->variantCaps) {
      case NS_FONT_VARIANT_CAPS_ALLPETITE:
      case NS_FONT_VARIANT_CAPS_PETITECAPS:
        bool synLower, synUpper;
        mFont->SupportsVariantCaps(aScript, style->variantCaps, addSmallCaps,
                                   synLower, synUpper);
        break;
      default:
        break;
    }
  }

  gfxFontEntry* entry = mFont->GetFontEntry();

  // insert any merged features into hb_feature array
  AutoTArray<hb_feature_t, 20> features;
  MergeFontFeatures(style, entry->mFeatureSettings,
                    aShapedText->DisableLigatures(), entry->FamilyName(),
                    addSmallCaps, AddOpenTypeFeature, &features);

  bool isRightToLeft = aShapedText->IsRightToLeft();

  hb_buffer_set_direction(
      mBuffer, aVertical
                   ? HB_DIRECTION_TTB
                   : (isRightToLeft ? HB_DIRECTION_RTL : HB_DIRECTION_LTR));
  hb_script_t scriptTag;
  if (aShapedText->GetFlags() & gfx::ShapedTextFlags::TEXT_USE_MATH_SCRIPT) {
    scriptTag = sMathScript;
  } else {
    scriptTag = GetHBScriptUsedForShaping(aScript);
  }
  hb_buffer_set_script(mBuffer, scriptTag);

  hb_language_t language;
  if (style->languageOverride) {
    language = hb_ot_tag_to_language(style->languageOverride);
  } else if (entry->mLanguageOverride) {
    language = hb_ot_tag_to_language(entry->mLanguageOverride);
  } else if (aLanguage) {
    nsCString langString;
    aLanguage->ToUTF8String(langString);
    language = hb_language_from_string(langString.get(), langString.Length());
  } else {
    language = hb_ot_tag_to_language(HB_OT_TAG_DEFAULT_LANGUAGE);
  }
  hb_buffer_set_language(mBuffer, language);

  uint32_t length = aLength;
  hb_buffer_add_utf16(mBuffer, reinterpret_cast<const uint16_t*>(aText), length,
                      0, length);

  hb_shape(mHBFont, mBuffer, features.Elements(), features.Length());

  if (isRightToLeft) {
    hb_buffer_reverse(mBuffer);
  }

  nsresult rv = SetGlyphsFromRun(aShapedText, aOffset, aLength, aText,
                                 aVertical, aRounding);

  NS_WARNING_ASSERTION(NS_SUCCEEDED(rv),
                       "failed to store glyphs into gfxShapedWord");
  hb_buffer_clear_contents(mBuffer);

  return NS_SUCCEEDED(rv);
}

#define SMALL_GLYPH_RUN \
  128  // some testing indicates that 90%+ of text runs
       // will fit without requiring separate allocation
       // for charToGlyphArray

nsresult gfxHarfBuzzShaper::SetGlyphsFromRun(gfxShapedText* aShapedText,
                                             uint32_t aOffset, uint32_t aLength,
                                             const char16_t* aText,
                                             bool aVertical,
                                             RoundingFlags aRounding) {
  typedef gfxShapedText::CompressedGlyph CompressedGlyph;

  uint32_t numGlyphs;
  const hb_glyph_info_t* ginfo = hb_buffer_get_glyph_infos(mBuffer, &numGlyphs);
  if (numGlyphs == 0) {
    return NS_OK;
  }

  AutoTArray<gfxTextRun::DetailedGlyph, 1> detailedGlyphs;

  uint32_t wordLength = aLength;
  static const int32_t NO_GLYPH = -1;
  AutoTArray<int32_t, SMALL_GLYPH_RUN> charToGlyphArray;
  if (!charToGlyphArray.SetLength(wordLength, fallible)) {
    return NS_ERROR_OUT_OF_MEMORY;
  }

  int32_t* charToGlyph = charToGlyphArray.Elements();
  for (uint32_t offset = 0; offset < wordLength; ++offset) {
    charToGlyph[offset] = NO_GLYPH;
  }

  for (uint32_t i = 0; i < numGlyphs; ++i) {
    uint32_t loc = ginfo[i].cluster;
    if (loc < wordLength) {
      charToGlyph[loc] = i;
    }
  }

  int32_t glyphStart = 0;  // looking for a clump that starts at this glyph
  int32_t charStart = 0;   // and this char index within the range of the run

  bool roundI, roundB;
  if (aVertical) {
    roundI = bool(aRounding & RoundingFlags::kRoundY);
    roundB = bool(aRounding & RoundingFlags::kRoundX);
  } else {
    roundI = bool(aRounding & RoundingFlags::kRoundX);
    roundB = bool(aRounding & RoundingFlags::kRoundY);
  }

  int32_t appUnitsPerDevUnit = aShapedText->GetAppUnitsPerDevUnit();
  CompressedGlyph* charGlyphs = aShapedText->GetCharacterGlyphs() + aOffset;

  // factor to convert 16.16 fixed-point pixels to app units
  // (only used if not rounding)
  double hb2appUnits = FixedToFloat(aShapedText->GetAppUnitsPerDevUnit());

  // Residual from rounding of previous advance, for use in rounding the
  // subsequent offset or advance appropriately.  16.16 fixed-point
  //
  // When rounding, the goal is to make the distance between glyphs and
  // their base glyph equal to the integral number of pixels closest to that
  // suggested by that shaper.
  // i.e. posInfo[n].x_advance - posInfo[n].x_offset + posInfo[n+1].x_offset
  //
  // The value of the residual is the part of the desired distance that has
  // not been included in integer offsets.
  hb_position_t residual = 0;

  // keep track of y-position to set glyph offsets if needed
  nscoord bPos = 0;

  const hb_glyph_position_t* posInfo =
      hb_buffer_get_glyph_positions(mBuffer, nullptr);

  while (glyphStart < int32_t(numGlyphs)) {
    int32_t charEnd = ginfo[glyphStart].cluster;
    int32_t glyphEnd = glyphStart;
    int32_t charLimit = wordLength;
    while (charEnd < charLimit) {
      // This is normally executed once for each iteration of the outer loop,
      // but in unusual cases where the character/glyph association is complex,
      // the initial character range might correspond to a non-contiguous
      // glyph range with "holes" in it. If so, we will repeat this loop to
      // extend the character range until we have a contiguous glyph sequence.
      charEnd += 1;
      while (charEnd != charLimit && charToGlyph[charEnd] == NO_GLYPH) {
        charEnd += 1;
      }

      // find the maximum glyph index covered by the clump so far
      for (int32_t i = charStart; i < charEnd; ++i) {
        if (charToGlyph[i] != NO_GLYPH) {
          glyphEnd = std::max(glyphEnd, charToGlyph[i] + 1);
          // update extent of glyph range
        }
      }

      if (glyphEnd == glyphStart + 1) {
        // for the common case of a single-glyph clump,
        // we can skip the following checks
        break;
      }

      if (glyphEnd == glyphStart) {
        // no glyphs, try to extend the clump
        continue;
      }

      // check whether all glyphs in the range are associated with the
      // characters in our clump; if not, we have a discontinuous range, and
      // should extend it unless we've reached the end of the text
      bool allGlyphsAreWithinCluster = true;
      for (int32_t i = glyphStart; i < glyphEnd; ++i) {
        int32_t glyphCharIndex = ginfo[i].cluster;
        if (glyphCharIndex < charStart || glyphCharIndex >= charEnd) {
          allGlyphsAreWithinCluster = false;
          break;
        }
      }
      if (allGlyphsAreWithinCluster) {
        break;
      }
    }

    NS_ASSERTION(glyphStart < glyphEnd,
                 "character/glyph clump contains no glyphs!");
    NS_ASSERTION(charStart != charEnd,
                 "character/glyph clump contains no characters!");

    // Now charStart..charEnd is a ligature clump, corresponding to
    // glyphStart..glyphEnd; Set baseCharIndex to the char we'll actually attach
    // the glyphs to (1st of ligature), and endCharIndex to the limit (position
    // beyond the last char), adjusting for the offset of the stringRange
    // relative to the textRun.
    int32_t baseCharIndex, endCharIndex;
    while (charEnd < int32_t(wordLength) && charToGlyph[charEnd] == NO_GLYPH)
      charEnd++;
    baseCharIndex = charStart;
    endCharIndex = charEnd;

    // Then we check if the clump falls outside our actual string range;
    // if so, just go to the next.
    if (baseCharIndex >= int32_t(wordLength)) {
      glyphStart = glyphEnd;
      charStart = charEnd;
      continue;
    }
    // Ensure we won't try to go beyond the valid length of the textRun's text
    endCharIndex = std::min<int32_t>(endCharIndex, wordLength);

    // Now we're ready to set the glyph info in the textRun
    int32_t glyphsInClump = glyphEnd - glyphStart;

    // Check for default-ignorable char that didn't get filtered, combined,
    // etc by the shaping process, and remove from the run.
    // (This may be done within harfbuzz eventually.)
    if (glyphsInClump == 1 && baseCharIndex + 1 == endCharIndex &&
        aShapedText->FilterIfIgnorable(aOffset + baseCharIndex,
                                       aText[baseCharIndex])) {
      glyphStart = glyphEnd;
      charStart = charEnd;
      continue;
    }

    // HarfBuzz gives us physical x- and y-coordinates, but we will store
    // them as logical inline- and block-direction values in the textrun.

    hb_position_t i_offset, i_advance;  // inline-direction offset/advance
    hb_position_t b_offset, b_advance;  // block-direction offset/advance
    if (aVertical) {
      // our coordinate directions are the opposite of harfbuzz's
      // when doing top-to-bottom shaping
      i_offset = -posInfo[glyphStart].y_offset;
      i_advance = -posInfo[glyphStart].y_advance;
      b_offset = -posInfo[glyphStart].x_offset;
      b_advance = -posInfo[glyphStart].x_advance;
    } else {
      i_offset = posInfo[glyphStart].x_offset;
      i_advance = posInfo[glyphStart].x_advance;
      b_offset = posInfo[glyphStart].y_offset;
      b_advance = posInfo[glyphStart].y_advance;
    }

    nscoord iOffset, advance;
    if (roundI) {
      iOffset = appUnitsPerDevUnit * FixedToIntRound(i_offset + residual);
      // Desired distance from the base glyph to the next reference point.
      hb_position_t width = i_advance - i_offset;
      int intWidth = FixedToIntRound(width);
      residual = width - FloatToFixed(intWidth);
      advance = appUnitsPerDevUnit * intWidth + iOffset;
    } else {
      iOffset = floor(hb2appUnits * i_offset + 0.5);
      advance = floor(hb2appUnits * i_advance + 0.5);
    }
    // Check if it's a simple one-to-one mapping
    if (glyphsInClump == 1 &&
        CompressedGlyph::IsSimpleGlyphID(ginfo[glyphStart].codepoint) &&
        CompressedGlyph::IsSimpleAdvance(advance) &&
        charGlyphs[baseCharIndex].IsClusterStart() && iOffset == 0 &&
        b_offset == 0 && b_advance == 0 && bPos == 0) {
      charGlyphs[baseCharIndex].SetSimpleGlyph(advance,
                                               ginfo[glyphStart].codepoint);
    } else {
      // Collect all glyphs in a list to be assigned to the first char;
      // there must be at least one in the clump, and we already measured
      // its advance, hence the placement of the loop-exit test and the
      // measurement of the next glyph.
      while (1) {
        gfxTextRun::DetailedGlyph* details = detailedGlyphs.AppendElement();
        details->mGlyphID = ginfo[glyphStart].codepoint;

        details->mAdvance = advance;

        if (aVertical) {
          details->mOffset.x =
              bPos - (roundB ? appUnitsPerDevUnit * FixedToIntRound(b_offset)
                             : floor(hb2appUnits * b_offset + 0.5));
          details->mOffset.y = iOffset;
        } else {
          details->mOffset.x = iOffset;
          details->mOffset.y =
              bPos - (roundB ? appUnitsPerDevUnit * FixedToIntRound(b_offset)
                             : floor(hb2appUnits * b_offset + 0.5));
        }

        if (b_advance != 0) {
          bPos -= roundB ? appUnitsPerDevUnit * FixedToIntRound(b_advance)
                         : floor(hb2appUnits * b_advance + 0.5);
        }
        if (++glyphStart >= glyphEnd) {
          break;
        }

        if (aVertical) {
          i_offset = -posInfo[glyphStart].y_offset;
          i_advance = -posInfo[glyphStart].y_advance;
          b_offset = -posInfo[glyphStart].x_offset;
          b_advance = -posInfo[glyphStart].x_advance;
        } else {
          i_offset = posInfo[glyphStart].x_offset;
          i_advance = posInfo[glyphStart].x_advance;
          b_offset = posInfo[glyphStart].y_offset;
          b_advance = posInfo[glyphStart].y_advance;
        }

        if (roundI) {
          iOffset = appUnitsPerDevUnit * FixedToIntRound(i_offset + residual);
          // Desired distance to the next reference point.  The
          // residual is considered here, and includes the residual
          // from the base glyph offset and subsequent advances, so
          // that the distance from the base glyph is optimized
          // rather than the distance from combining marks.
          i_advance += residual;
          int intAdvance = FixedToIntRound(i_advance);
          residual = i_advance - FloatToFixed(intAdvance);
          advance = appUnitsPerDevUnit * intAdvance;
        } else {
          iOffset = floor(hb2appUnits * i_offset + 0.5);
          advance = floor(hb2appUnits * i_advance + 0.5);
        }
      }

      aShapedText->SetDetailedGlyphs(aOffset + baseCharIndex,
                                     detailedGlyphs.Length(),
                                     detailedGlyphs.Elements());

      detailedGlyphs.Clear();
    }

    // the rest of the chars in the group are ligature continuations,
    // no associated glyphs
    while (++baseCharIndex != endCharIndex &&
           baseCharIndex < int32_t(wordLength)) {
      CompressedGlyph& g = charGlyphs[baseCharIndex];
      NS_ASSERTION(!g.IsSimpleGlyph(), "overwriting a simple glyph");
      g.SetComplex(g.IsClusterStart(), false);
    }

    glyphStart = glyphEnd;
    charStart = charEnd;
  }

  return NS_OK;
}