C++ TextBreakIterator类代码示例

TextBreakIterator* acquireLineBreakIterator(const LChar* string, int length, const AtomicString& locale, const UChar* priorContext, unsigned priorContextLength)
{
    TextBreakIterator* iterator = LineBreakIteratorPool::sharedPool().take(locale);
    if (!iterator)
        return 0;

    UTextWithBuffer textLocal;
    textLocal.text = emptyText;
    textLocal.text.extraSize = sizeof(textLocal.buffer);
    textLocal.text.pExtra = textLocal.buffer;

    UErrorCode openStatus = U_ZERO_ERROR;
    UText* text = textOpenLatin1(&textLocal, string, length, priorContext, priorContextLength, &openStatus);
    if (U_FAILURE(openStatus)) {
        WTF_LOG_ERROR("textOpenLatin1 failed with status %d", openStatus);
        return 0;
    }

    UErrorCode setTextStatus = U_ZERO_ERROR;
    iterator->setText(text, setTextStatus);
    if (U_FAILURE(setTextStatus)) {
        WTF_LOG_ERROR("ubrk_setUText failed with status %d", setTextStatus);
        return 0;
    }

    utext_close(text);

    return iterator;
}

int findNextWordFromIndex(const UChar* chars, int len, int position, bool forward)
{
    TextBreakIterator* it = wordBreakIterator(chars, len);

    if (forward) {
        position = it->following(position);
        while (position != TextBreakDone) {
            // We stop searching when the character preceeding the break
            // is alphanumeric.
            if (position < len && isAlphanumeric(chars[position - 1]))
                return position;

            position = it->following(position);
        }

        return len;
    } else {
        position = it->preceding(position);
        while (position != TextBreakDone) {
            // We stop searching when the character following the break
            // is alphanumeric.
            if (position > 0 && isAlphanumeric(chars[position]))
                return position;

            position = it->preceding(position);
        }

        return 0;
    }
}

static void findMisspellings(TextCheckerClient& client, const UChar* text, int start, int length, Vector<TextCheckingResult>& results)
{
    TextBreakIterator* iterator = wordBreakIterator(text + start, length);
    if (!iterator)
        return;
    int wordStart = iterator->current();
    while (0 <= wordStart) {
        int wordEnd = iterator->next();
        if (wordEnd < 0)
            break;
        int wordLength = wordEnd - wordStart;
        int misspellingLocation = -1;
        int misspellingLength = 0;
        client.checkSpellingOfString(String(text + start + wordStart, wordLength), &misspellingLocation, &misspellingLength);
        if (0 < misspellingLength) {
            ASSERT(0 <= misspellingLocation && misspellingLocation <= wordLength);
            ASSERT(0 < misspellingLength && misspellingLocation + misspellingLength <= wordLength);
            TextCheckingResult misspelling;
            misspelling.decoration = TextDecorationTypeSpelling;
            misspelling.location = start + wordStart + misspellingLocation;
            misspelling.length = misspellingLength;
            results.append(misspelling);
        }

        wordStart = wordEnd;
    }
}

Dart_Handle Paragraph::getWordBoundary(unsigned offset) {
  String text;
  int start = 0, end = 0;

  for (RenderObject* object = m_renderView.get(); object;
       object = object->nextInPreOrder()) {
    if (!object->isText())
      continue;
    RenderText* renderText = toRenderText(object);
    text.append(renderText->text());
  }

  TextBreakIterator* it = wordBreakIterator(text, 0, text.length());
  if (it) {
    end = it->following(offset);
    if (end < 0)
      end = it->last();
    start = it->previous();
  }

  Dart_Handle result = Dart_NewList(2);
  Dart_ListSetAt(result, 0, ToDart(start));
  Dart_ListSetAt(result, 1, ToDart(end));
  return result;
}

void findWordBoundary(const UChar* chars, int len, int position, int* start, int* end)
{
    TextBreakIterator* it = wordBreakIterator(chars, len);
    *end = it->following(position);
    if (*end < 0)
        *end = it->last();
    *start = it->previous();
}

static inline unsigned countGraphemesInCluster(const UChar* str,
    unsigned strLength, uint16_t startIndex, uint16_t endIndex)
{
    if (startIndex > endIndex) {
        uint16_t tempIndex = startIndex;
        startIndex = endIndex;
        endIndex = tempIndex;
    }
    uint16_t length = endIndex - startIndex;
    ASSERT(static_cast<unsigned>(startIndex + length) <= strLength);
    TextBreakIterator* cursorPosIterator = cursorMovementIterator(&str[startIndex], length);

    int cursorPos = cursorPosIterator->current();
    int numGraphemes = -1;
    while (0 <= cursorPos) {
        cursorPos = cursorPosIterator->next();
        numGraphemes++;
    }
    return std::max(0, numGraphemes);
}

void AbstractInlineTextBox::wordBoundaries(Vector<WordBoundaries>& words) const
{
    if (!m_inlineTextBox)
        return;

    String text = this->text();
    int len = text.length();
    TextBreakIterator* iterator = wordBreakIterator(text, 0, len);

    // FIXME: When http://crbug.com/411764 is fixed, replace this with an ASSERT.
    if (!iterator)
        return;

    int pos = iterator->first();
    while (pos >= 0 && pos < len) {
        int next = iterator->next();
        if (isWordTextBreak(iterator))
            words.append(WordBoundaries(pos, next));
        pos = next;
    }
}

int findWordEndBoundary(const UChar* chars, int len, int position)
{
    TextBreakIterator* it = wordBreakIterator(chars, len);
    int end = it->following(position);
    return end < 0 ? it->last() : end;
}

// Helper function to determine whether text is a single word.
static bool isASingleWord(const String& text)
{
    TextBreakIterator* it = wordBreakIterator(text, 0, text.length());
    return it && it->next() == static_cast<int>(text.length());
}