DXR is a code search and navigation tool aimed at making sense of large projects. It supports full-text and regex searches as well as structural queries.

Header

Mercurial (d38398e5144e)

VCS Links

LanguageSpecificCasingBehavior

Macros

Line Code
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711
/* -*- Mode: C++; tab-width: 2; 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 "nsTextRunTransformations.h"

#include "mozilla/MemoryReporting.h"
#include "mozilla/Move.h"

#include "nsGkAtoms.h"
#include "nsStyleConsts.h"
#include "nsUnicharUtils.h"
#include "nsUnicodeProperties.h"
#include "nsSpecialCasingData.h"
#include "mozilla/gfx/2D.h"
#include "nsTextFrameUtils.h"
#include "nsIPersistentProperties2.h"
#include "GreekCasing.h"
#include "IrishCasing.h"

using namespace mozilla;
using namespace mozilla::gfx;

// Unicode characters needing special casing treatment in tr/az languages
#define LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE  0x0130
#define LATIN_SMALL_LETTER_DOTLESS_I           0x0131

// Greek sigma needs custom handling for the lowercase transform; for details
// see comments under "case NS_STYLE_TEXT_TRANSFORM_LOWERCASE" within
// nsCaseTransformTextRunFactory::RebuildTextRun(), and bug 740120.
#define GREEK_CAPITAL_LETTER_SIGMA             0x03A3
#define GREEK_SMALL_LETTER_FINAL_SIGMA         0x03C2
#define GREEK_SMALL_LETTER_SIGMA               0x03C3

already_AddRefed<nsTransformedTextRun>
nsTransformedTextRun::Create(const gfxTextRunFactory::Parameters* aParams,
                             nsTransformingTextRunFactory* aFactory,
                             gfxFontGroup* aFontGroup,
                             const char16_t* aString, uint32_t aLength,
                             const uint32_t aFlags,
                             nsTArray<RefPtr<nsTransformedCharStyle>>&& aStyles,
                             bool aOwnsFactory)
{
  NS_ASSERTION(!(aFlags & gfxTextRunFactory::TEXT_IS_8BIT),
               "didn't expect text to be marked as 8-bit here");

  void *storage = AllocateStorageForTextRun(sizeof(nsTransformedTextRun), aLength);
  if (!storage) {
    return nullptr;
  }

  RefPtr<nsTransformedTextRun> result =
    new (storage) nsTransformedTextRun(aParams, aFactory, aFontGroup,
                                       aString, aLength, aFlags,
                                       Move(aStyles), aOwnsFactory);
  return result.forget();
}

void
nsTransformedTextRun::SetCapitalization(uint32_t aStart, uint32_t aLength,
                                        bool* aCapitalization)
{
  if (mCapitalize.IsEmpty()) {
    if (!mCapitalize.AppendElements(GetLength()))
      return;
    memset(mCapitalize.Elements(), 0, GetLength()*sizeof(bool));
  }
  memcpy(mCapitalize.Elements() + aStart, aCapitalization, aLength*sizeof(bool));
  mNeedsRebuild = true;
}

bool
nsTransformedTextRun::SetPotentialLineBreaks(Range aRange,
                                             const uint8_t* aBreakBefore)
{
  bool changed = gfxTextRun::SetPotentialLineBreaks(aRange, aBreakBefore);
  if (changed) {
    mNeedsRebuild = true;
  }
  return changed;
}

size_t
nsTransformedTextRun::SizeOfExcludingThis(mozilla::MallocSizeOf aMallocSizeOf)
{
  size_t total = gfxTextRun::SizeOfExcludingThis(aMallocSizeOf);
  total += mStyles.ShallowSizeOfExcludingThis(aMallocSizeOf);
  total += mCapitalize.ShallowSizeOfExcludingThis(aMallocSizeOf);
  if (mOwnsFactory) {
    total += aMallocSizeOf(mFactory);
  }
  return total;
}

size_t
nsTransformedTextRun::SizeOfIncludingThis(mozilla::MallocSizeOf aMallocSizeOf)
{
  return aMallocSizeOf(this) + SizeOfExcludingThis(aMallocSizeOf);
}

already_AddRefed<nsTransformedTextRun>
nsTransformingTextRunFactory::MakeTextRun(const char16_t* aString, uint32_t aLength,
                                          const gfxTextRunFactory::Parameters* aParams,
                                          gfxFontGroup* aFontGroup, uint32_t aFlags,
                                          nsTArray<RefPtr<nsTransformedCharStyle>>&& aStyles,
                                          bool aOwnsFactory)
{
  return nsTransformedTextRun::Create(aParams, this, aFontGroup,
                                      aString, aLength, aFlags, Move(aStyles),
                                      aOwnsFactory);
}

already_AddRefed<nsTransformedTextRun>
nsTransformingTextRunFactory::MakeTextRun(const uint8_t* aString, uint32_t aLength,
                                          const gfxTextRunFactory::Parameters* aParams,
                                          gfxFontGroup* aFontGroup, uint32_t aFlags,
                                          nsTArray<RefPtr<nsTransformedCharStyle>>&& aStyles,
                                          bool aOwnsFactory)
{
  // We'll only have a Unicode code path to minimize the amount of code needed
  // for these rarely used features
  NS_ConvertASCIItoUTF16 unicodeString(reinterpret_cast<const char*>(aString), aLength);
  return MakeTextRun(unicodeString.get(), aLength, aParams, aFontGroup,
                     aFlags & ~(gfxFontGroup::TEXT_IS_PERSISTENT | gfxFontGroup::TEXT_IS_8BIT),
                     Move(aStyles), aOwnsFactory);
}

void
MergeCharactersInTextRun(gfxTextRun* aDest, gfxTextRun* aSrc,
                         const bool* aCharsToMerge, const bool* aDeletedChars)
{
  aDest->ResetGlyphRuns();

  gfxTextRun::GlyphRunIterator iter(aSrc, gfxTextRun::Range(aSrc));
  uint32_t offset = 0;
  AutoTArray<gfxTextRun::DetailedGlyph,2> glyphs;
  while (iter.NextRun()) {
    const gfxTextRun::GlyphRun* run = iter.GetGlyphRun();
    nsresult rv = aDest->AddGlyphRun(run->mFont, run->mMatchType,
                                     offset, false, run->mOrientation);
    if (NS_FAILED(rv))
      return;

    bool anyMissing = false;
    uint32_t mergeRunStart = iter.GetStringStart();
    const gfxTextRun::CompressedGlyph *srcGlyphs = aSrc->GetCharacterGlyphs();
    gfxTextRun::CompressedGlyph mergedGlyph = srcGlyphs[mergeRunStart];
    uint32_t stringEnd = iter.GetStringEnd();
    for (uint32_t k = iter.GetStringStart(); k < stringEnd; ++k) {
      const gfxTextRun::CompressedGlyph g = srcGlyphs[k];
      if (g.IsSimpleGlyph()) {
        if (!anyMissing) {
          gfxTextRun::DetailedGlyph details;
          details.mGlyphID = g.GetSimpleGlyph();
          details.mAdvance = g.GetSimpleAdvance();
          details.mXOffset = 0;
          details.mYOffset = 0;
          glyphs.AppendElement(details);
        }
      } else {
        if (g.IsMissing()) {
          anyMissing = true;
          glyphs.Clear();
        }
        if (g.GetGlyphCount() > 0) {
          glyphs.AppendElements(aSrc->GetDetailedGlyphs(k), g.GetGlyphCount());
        }
      }

      if (k + 1 < iter.GetStringEnd() && aCharsToMerge[k + 1]) {
        // next char is supposed to merge with current, so loop without
        // writing current merged glyph to the destination
        continue;
      }

      // If the start of the merge run is actually a character that should
      // have been merged with the previous character (this can happen
      // if there's a font change in the middle of a case-mapped character,
      // that decomposed into a sequence of base+diacritics, for example),
      // just discard the entire merge run. See comment at start of this
      // function.
      NS_WARNING_ASSERTION(
        !aCharsToMerge[mergeRunStart],
        "unable to merge across a glyph run boundary, glyph(s) discarded");
      if (!aCharsToMerge[mergeRunStart]) {
        if (anyMissing) {
          mergedGlyph.SetMissing(glyphs.Length());
        } else {
          mergedGlyph.SetComplex(mergedGlyph.IsClusterStart(),
                                 mergedGlyph.IsLigatureGroupStart(),
                                 glyphs.Length());
        }
        aDest->SetGlyphs(offset, mergedGlyph, glyphs.Elements());
        ++offset;

        while (offset < aDest->GetLength() && aDeletedChars[offset]) {
          aDest->SetGlyphs(offset++, gfxTextRun::CompressedGlyph(), nullptr);
        }
      }

      glyphs.Clear();
      anyMissing = false;
      mergeRunStart = k + 1;
      if (mergeRunStart < stringEnd) {
        mergedGlyph = srcGlyphs[mergeRunStart];
      }
    }
    NS_ASSERTION(glyphs.Length() == 0,
                 "Leftover glyphs, don't request merging of the last character with its next!");  
  }
  NS_ASSERTION(offset == aDest->GetLength(), "Bad offset calculations");
}

gfxTextRunFactory::Parameters
GetParametersForInner(nsTransformedTextRun* aTextRun, uint32_t* aFlags,
                      DrawTarget* aRefDrawTarget)
{
  gfxTextRunFactory::Parameters params =
    { aRefDrawTarget, nullptr, nullptr,
      nullptr, 0, aTextRun->GetAppUnitsPerDevUnit()
    };
  *aFlags = aTextRun->GetFlags() & ~gfxFontGroup::TEXT_IS_PERSISTENT;
  return params;
}

// Some languages have special casing conventions that differ from the
// default Unicode mappings.
// The enum values here are named for well-known exemplar languages that
// exhibit the behavior in question; multiple lang tags may map to the
// same setting here, if the behavior is shared by other languages.
enum LanguageSpecificCasingBehavior {
  eLSCB_None,    // default non-lang-specific behavior
  eLSCB_Dutch,   // treat "ij" digraph as a unit for capitalization
  eLSCB_Greek,   // strip accent when uppercasing Greek vowels
  eLSCB_Irish,   // keep prefix letters as lowercase when uppercasing Irish
  eLSCB_Turkish  // preserve dotted/dotless-i distinction in uppercase
};

static LanguageSpecificCasingBehavior
GetCasingFor(const nsIAtom* aLang)
{
  if (!aLang) {
      return eLSCB_None;
  }
  if (aLang == nsGkAtoms::tr ||
      aLang == nsGkAtoms::az ||
      aLang == nsGkAtoms::ba ||
      aLang == nsGkAtoms::crh ||
      aLang == nsGkAtoms::tt) {
    return eLSCB_Turkish;
  }
  if (aLang == nsGkAtoms::nl) {
    return eLSCB_Dutch;
  }
  if (aLang == nsGkAtoms::el) {
    return eLSCB_Greek;
  }
  if (aLang == nsGkAtoms::ga) {
    return eLSCB_Irish;
  }

  // Is there a region subtag we should ignore?
  nsAtomString langStr(const_cast<nsIAtom*>(aLang));
  int index = langStr.FindChar('-');
  if (index > 0) {
    langStr.Truncate(index);
    nsCOMPtr<nsIAtom> truncatedLang = NS_Atomize(langStr);
    return GetCasingFor(truncatedLang);
  }

  return eLSCB_None;
}

bool
nsCaseTransformTextRunFactory::TransformString(
    const nsAString& aString,
    nsString& aConvertedString,
    bool aAllUppercase,
    const nsIAtom* aLanguage,
    nsTArray<bool>& aCharsToMergeArray,
    nsTArray<bool>& aDeletedCharsArray,
    const nsTransformedTextRun* aTextRun,
    uint32_t aOffsetInTextRun,
    nsTArray<uint8_t>* aCanBreakBeforeArray,
    nsTArray<RefPtr<nsTransformedCharStyle>>* aStyleArray)
{
  bool auxiliaryOutputArrays = aCanBreakBeforeArray && aStyleArray;
  MOZ_ASSERT(!auxiliaryOutputArrays || aTextRun,
      "text run must be provided to use aux output arrays");

  uint32_t length = aString.Length();
  const char16_t* str = aString.BeginReading();

  bool mergeNeeded = false;

  bool capitalizeDutchIJ = false;
  bool prevIsLetter = false;
  bool ntPrefix = false; // true immediately after a word-initial 'n' or 't'
                         // when doing Irish lowercasing
  uint32_t sigmaIndex = uint32_t(-1);
  nsIUGenCategory::nsUGenCategory cat;

  uint8_t style = aAllUppercase ? NS_STYLE_TEXT_TRANSFORM_UPPERCASE : 0;
  bool forceNonFullWidth = false;
  const nsIAtom* lang = aLanguage;

  LanguageSpecificCasingBehavior languageSpecificCasing = GetCasingFor(lang);
  mozilla::GreekCasing::State greekState;
  mozilla::IrishCasing::State irishState;
  uint32_t irishMark = uint32_t(-1); // location of possible prefix letter(s)
                                     // in the output string
  uint32_t irishMarkSrc = uint32_t(-1); // corresponding location in source
                                        // string (may differ from output due to
                                        // expansions like eszet -> 'SS')
  uint32_t greekMark = uint32_t(-1); // location of uppercase ETA that may need
                                     // tonos added (if it is disjunctive eta)
  const char16_t kGreekUpperEta = 0x0397;

  for (uint32_t i = 0; i < length; ++i, ++aOffsetInTextRun) {
    uint32_t ch = str[i];

    RefPtr<nsTransformedCharStyle> charStyle;
    if (aTextRun) {
      charStyle = aTextRun->mStyles[aOffsetInTextRun];
      style = aAllUppercase ? NS_STYLE_TEXT_TRANSFORM_UPPERCASE :
        charStyle->mTextTransform;
      forceNonFullWidth = charStyle->mForceNonFullWidth;

      nsIAtom* newLang = charStyle->mExplicitLanguage
                         ? charStyle->mLanguage.get() : nullptr;
      if (lang != newLang) {
        lang = newLang;
        languageSpecificCasing = GetCasingFor(lang);
        greekState.Reset();
        irishState.Reset();
        irishMark = uint32_t(-1);
        irishMarkSrc = uint32_t(-1);
        greekMark = uint32_t(-1);
      }
    }

    int extraChars = 0;
    const mozilla::unicode::MultiCharMapping *mcm;
    bool inhibitBreakBefore = false; // have we just deleted preceding hyphen?

    if (NS_IS_HIGH_SURROGATE(ch) && i < length - 1 &&
        NS_IS_LOW_SURROGATE(str[i + 1])) {
      ch = SURROGATE_TO_UCS4(ch, str[i + 1]);
    }

    switch (style) {
    case NS_STYLE_TEXT_TRANSFORM_LOWERCASE:
      if (languageSpecificCasing == eLSCB_Turkish) {
        if (ch == 'I') {
          ch = LATIN_SMALL_LETTER_DOTLESS_I;
          prevIsLetter = true;
          sigmaIndex = uint32_t(-1);
          break;
        }
        if (ch == LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE) {
          ch = 'i';
          prevIsLetter = true;
          sigmaIndex = uint32_t(-1);
          break;
        }
      }

      cat = mozilla::unicode::GetGenCategory(ch);

      if (languageSpecificCasing == eLSCB_Irish &&
          cat == nsIUGenCategory::kLetter) {
        // See bug 1018805 for Irish lowercasing requirements
        if (!prevIsLetter && (ch == 'n' || ch == 't')) {
          ntPrefix = true;
        } else {
          if (ntPrefix && mozilla::IrishCasing::IsUpperVowel(ch)) {
            aConvertedString.Append('-');
            ++extraChars;
          }
          ntPrefix = false;
        }
      } else {
        ntPrefix = false;
      }

      // Special lowercasing behavior for Greek Sigma: note that this is listed
      // as context-sensitive in Unicode's SpecialCasing.txt, but is *not* a
      // language-specific mapping; it applies regardless of the language of
      // the element.
      //
      // The lowercase mapping for CAPITAL SIGMA should be to SMALL SIGMA (i.e.
      // the non-final form) whenever there is a following letter, or when the
      // CAPITAL SIGMA occurs in isolation (neither preceded nor followed by a
      // LETTER); and to FINAL SIGMA when it is preceded by another letter but
      // not followed by one.
      //
      // To implement the context-sensitive nature of this mapping, we keep
      // track of whether the previous character was a letter. If not, CAPITAL
      // SIGMA will map directly to SMALL SIGMA. If the previous character
      // was a letter, CAPITAL SIGMA maps to FINAL SIGMA and we record the
      // position in the converted string; if we then encounter another letter,
      // that FINAL SIGMA is replaced with a standard SMALL SIGMA.

      // If sigmaIndex is not -1, it marks where we have provisionally mapped
      // a CAPITAL SIGMA to FINAL SIGMA; if we now find another letter, we
      // need to change it to SMALL SIGMA.
      if (sigmaIndex != uint32_t(-1)) {
        if (cat == nsIUGenCategory::kLetter) {
          aConvertedString.SetCharAt(GREEK_SMALL_LETTER_SIGMA, sigmaIndex);
        }
      }

      if (ch == GREEK_CAPITAL_LETTER_SIGMA) {
        // If preceding char was a letter, map to FINAL instead of SMALL,
        // and note where it occurred by setting sigmaIndex; we'll change it
        // to standard SMALL SIGMA later if another letter follows
        if (prevIsLetter) {
          ch = GREEK_SMALL_LETTER_FINAL_SIGMA;
          sigmaIndex = aConvertedString.Length();
        } else {
          // CAPITAL SIGMA not preceded by a letter is unconditionally mapped
          // to SMALL SIGMA
          ch = GREEK_SMALL_LETTER_SIGMA;
          sigmaIndex = uint32_t(-1);
        }
        prevIsLetter = true;
        break;
      }

      // ignore diacritics for the purpose of contextual sigma mapping;
      // otherwise, reset prevIsLetter appropriately and clear the
      // sigmaIndex marker
      if (cat != nsIUGenCategory::kMark) {
        prevIsLetter = (cat == nsIUGenCategory::kLetter);
        sigmaIndex = uint32_t(-1);
      }

      mcm = mozilla::unicode::SpecialLower(ch);
      if (mcm) {
        int j = 0;
        while (j < 2 && mcm->mMappedChars[j + 1]) {
          aConvertedString.Append(mcm->mMappedChars[j]);
          ++extraChars;
          ++j;
        }
        ch = mcm->mMappedChars[j];
        break;
      }

      ch = ToLowerCase(ch);
      break;

    case NS_STYLE_TEXT_TRANSFORM_UPPERCASE:
      if (languageSpecificCasing == eLSCB_Turkish && ch == 'i') {
        ch = LATIN_CAPITAL_LETTER_I_WITH_DOT_ABOVE;
        break;
      }

      if (languageSpecificCasing == eLSCB_Greek) {
        bool markEta;
        bool updateEta;
        ch = mozilla::GreekCasing::UpperCase(ch, greekState,
                                             markEta, updateEta);
        if (markEta) {
          greekMark = aConvertedString.Length();
        } else if (updateEta) {
          // Remove the TONOS from an uppercase ETA-TONOS that turned out
          // not to be disjunctive-eta.
          MOZ_ASSERT(aConvertedString.Length() > 0 &&
                     greekMark < aConvertedString.Length(),
                     "bad greekMark!");
          aConvertedString.SetCharAt(kGreekUpperEta, greekMark);
          greekMark = uint32_t(-1);
        }
        break;
      }

      if (languageSpecificCasing == eLSCB_Irish) {
        bool mark;
        uint8_t action;
        ch = mozilla::IrishCasing::UpperCase(ch, irishState, mark, action);
        if (mark) {
          irishMark = aConvertedString.Length();
          irishMarkSrc = i;
          break;
        } else if (action) {
          nsString& str = aConvertedString; // shorthand
          switch (action) {
          case 1:
            // lowercase a single prefix letter
            NS_ASSERTION(str.Length() > 0 && irishMark < str.Length(),
                         "bad irishMark!");
            str.SetCharAt(ToLowerCase(str[irishMark]), irishMark);
            irishMark = uint32_t(-1);
            irishMarkSrc = uint32_t(-1);
            break;
          case 2:
            // lowercase two prefix letters (immediately before current pos)
            NS_ASSERTION(str.Length() >= 2 && irishMark == str.Length() - 2,
                         "bad irishMark!");