C++ U16_IS_TRAIL函数代码示例

/*
 * Compare two strings as presented by UCharIterators.
 * Use code unit or code point order.
 * When the function returns, it is undefined where the iterators
 * have stopped.
 */
U_CAPI int32_t U_EXPORT2
u_strCompareIter(UCharIterator *iter1, UCharIterator *iter2, UBool codePointOrder) {
    UChar32 c1, c2;

    /* argument checking */
    if(iter1==NULL || iter2==NULL) {
        return 0; /* bad arguments */
    }
    if(iter1==iter2) {
        return 0; /* identical iterators */
    }

    /* reset iterators to start? */
    iter1->move(iter1, 0, UITER_START);
    iter2->move(iter2, 0, UITER_START);

    /* compare identical prefixes - they do not need to be fixed up */
    for(;;) {
        c1=iter1->next(iter1);
        c2=iter2->next(iter2);
        if(c1!=c2) {
            break;
        }
        if(c1==-1) {
            return 0;
        }
    }

    /* if both values are in or above the surrogate range, fix them up */
    if(c1>=0xd800 && c2>=0xd800 && codePointOrder) {
        /* subtract 0x2800 from BMP code points to make them smaller than supplementary ones */
        if(
            (c1<=0xdbff && U16_IS_TRAIL(iter1->current(iter1))) ||
            (U16_IS_TRAIL(c1) && (iter1->previous(iter1), U16_IS_LEAD(iter1->previous(iter1))))
        ) {
            /* part of a surrogate pair, leave >=d800 */
        } else {
            /* BMP code point - may be surrogate code point - make <d800 */
            c1-=0x2800;
        }

        if(
            (c2<=0xdbff && U16_IS_TRAIL(iter2->current(iter2))) ||
            (U16_IS_TRAIL(c2) && (iter2->previous(iter2), U16_IS_LEAD(iter2->previous(iter2))))
        ) {
            /* part of a surrogate pair, leave >=d800 */
        } else {
            /* BMP code point - may be surrogate code point - make <d800 */
            c2-=0x2800;
        }
    }

    /* now c1 and c2 are in the requested (code unit or code point) order */
    return (int32_t)c1-(int32_t)c2;
}

/*
 * Test if a substring match inside a string is at code point boundaries.
 * All pointers refer to the same buffer.
 * The limit pointer may be NULL, all others must be real pointers.
 */
static U_INLINE UBool
isMatchAtCPBoundary(const UChar* start, const UChar* match, const UChar* matchLimit, const UChar* limit) {
    if (U16_IS_TRAIL(*match) && start != match && U16_IS_LEAD(*(match - 1))) {
        /* the leading edge of the match is in the middle of a surrogate pair */
        return FALSE;
    }
    if (U16_IS_LEAD(*(matchLimit - 1)) && match != limit && U16_IS_TRAIL(*matchLimit)) {
        /* the trailing edge of the match is in the middle of a surrogate pair */
        return FALSE;
    }
    return TRUE;
}

unsigned Font::expansionOpportunityCount(const UChar* characters, size_t length, TextDirection direction, bool& isAfterExpansion)
{
    static bool expandAroundIdeographs = canExpandAroundIdeographsInComplexText();
    unsigned count = 0;
    if (direction == LTR) {
        for (size_t i = 0; i < length; ++i) {
            UChar32 character = characters[i];
            if (treatAsSpace(character)) {
                count++;
                isAfterExpansion = true;
                continue;
            }
            if (U16_IS_LEAD(character) && i + 1 < length && U16_IS_TRAIL(characters[i + 1])) {
                character = U16_GET_SUPPLEMENTARY(character, characters[i + 1]);
                i++;
            }
            if (expandAroundIdeographs && isCJKIdeographOrSymbol(character)) {
                if (!isAfterExpansion)
                    count++;
                count++;
                isAfterExpansion = true;
                continue;
            }
            isAfterExpansion = false;
        }
    } else {
        for (size_t i = length; i > 0; --i) {
            UChar32 character = characters[i - 1];
            if (treatAsSpace(character)) {
                count++;
                isAfterExpansion = true;
                continue;
            }
            if (U16_IS_TRAIL(character) && i > 1 && U16_IS_LEAD(characters[i - 2])) {
                character = U16_GET_SUPPLEMENTARY(characters[i - 2], character);
                i--;
            }
            if (expandAroundIdeographs && isCJKIdeographOrSymbol(character)) {
                if (!isAfterExpansion)
                    count++;
                count++;
                isAfterExpansion = true;
                continue;
            }
            isAfterExpansion = false;
        }
    }
    return count;
}

unsigned Character::expansionOpportunityCount(const UChar* characters, size_t length, TextDirection direction, bool& isAfterExpansion, const TextJustify textJustify)
{
    unsigned count = 0;
    if (direction == LTR) {
        for (size_t i = 0; i < length; ++i) {
            UChar32 character = characters[i];
            if (treatAsSpace(character)) {
                count++;
                isAfterExpansion = true;
                continue;
            }
            if (U16_IS_LEAD(character) && i + 1 < length && U16_IS_TRAIL(characters[i + 1])) {
                character = U16_GET_SUPPLEMENTARY(character, characters[i + 1]);
                i++;
            }
            if (textJustify == TextJustify::TextJustifyAuto && isCJKIdeographOrSymbol(character)) {
                if (!isAfterExpansion)
                    count++;
                count++;
                isAfterExpansion = true;
                continue;
            }
            isAfterExpansion = false;
        }
    } else {
        for (size_t i = length; i > 0; --i) {
            UChar32 character = characters[i - 1];
            if (treatAsSpace(character)) {
                count++;
                isAfterExpansion = true;
                continue;
            }
            if (U16_IS_TRAIL(character) && i > 1 && U16_IS_LEAD(characters[i - 2])) {
                character = U16_GET_SUPPLEMENTARY(characters[i - 2], character);
                i--;
            }
            if (textJustify == TextJustify::TextJustifyAuto && isCJKIdeographOrSymbol(character)) {
                if (!isAfterExpansion)
                    count++;
                count++;
                isAfterExpansion = true;
                continue;
            }
            isAfterExpansion = false;
        }
    }
    return count;
}

static void TestCodeUnitValues()
{
    static uint16_t codeunit[]={0x0000,0xe065,0x20ac,0xd7ff,0xd800,0xd841,0xd905,0xdbff,0xdc00,0xdc02,0xddee,0xdfff,0};
    
    int16_t i;
    for(i=0; i<sizeof(codeunit)/sizeof(codeunit[0]); i++){
        UChar c=codeunit[i];
        log_verbose("Testing code unit value of %x\n", c);
        if(i<4){
            if(!UTF16_IS_SINGLE(c) || UTF16_IS_LEAD(c) || UTF16_IS_TRAIL(c) || !U16_IS_SINGLE(c) || U16_IS_LEAD(c) || U16_IS_TRAIL(c)){
                log_err("ERROR: %x is a single character\n", c);
            }
        }
        if(i >= 4 && i< 8){
            if(!UTF16_IS_LEAD(c) || UTF16_IS_SINGLE(c) || UTF16_IS_TRAIL(c) || !U16_IS_LEAD(c) || U16_IS_SINGLE(c) || U16_IS_TRAIL(c)){
                log_err("ERROR: %x is a first surrogate\n", c);
            }
        }
        if(i >= 8 && i< 12){
            if(!UTF16_IS_TRAIL(c) || UTF16_IS_SINGLE(c) || UTF16_IS_LEAD(c) || !U16_IS_TRAIL(c) || U16_IS_SINGLE(c) || U16_IS_LEAD(c)){
                log_err("ERROR: %x is a second surrogate\n", c);
            }
        }
    }
}

void
MessagePattern::setParseError(UParseError *parseError, int32_t index) {
    if(parseError==NULL) {
        return;
    }
    parseError->offset=index;

    // Set preContext to some of msg before index.
    // Avoid splitting a surrogate pair.
    int32_t length=index;
    if(length>=U_PARSE_CONTEXT_LEN) {
        length=U_PARSE_CONTEXT_LEN-1;
        if(length>0 && U16_IS_TRAIL(msg[index-length])) {
            --length;
        }
    }
    msg.extract(index-length, length, parseError->preContext);
    parseError->preContext[length]=0;

    // Set postContext to some of msg starting at index.
    length=msg.length()-index;
    if(length>=U_PARSE_CONTEXT_LEN) {
        length=U_PARSE_CONTEXT_LEN-1;
        if(length>0 && U16_IS_LEAD(msg[index+length-1])) {
            --length;
        }
    }
    msg.extract(index, length, parseError->postContext);
    parseError->postContext[length]=0;
}

UChar
FCDUIterCollationIterator::handleGetTrailSurrogate() {
    if(state <= ITER_IN_FCD_SEGMENT) {
        UChar32 trail = iter.next(&iter);
        if(U16_IS_TRAIL(trail)) {
            if(state == ITER_IN_FCD_SEGMENT) { ++pos; }
        } else if(trail >= 0) {
            iter.previous(&iter);
        }
        return (UChar)trail;
    } else {
        U_ASSERT(pos < normalized.length());
        UChar trail;
        if(U16_IS_TRAIL(trail = normalized[pos])) { ++pos; }
        return trail;
    }
}

UChar32 StringImpl::characterStartingAt(unsigned i)
{
    if (U16_IS_SINGLE(m_data[i]))
        return m_data[i];
    if (i + 1 < m_length && U16_IS_LEAD(m_data[i]) && U16_IS_TRAIL(m_data[i + 1]))
        return U16_GET_SUPPLEMENTARY(m_data[i], m_data[i + 1]);
    return 0;
}

UChar
FCDUTF8CollationIterator::handleGetTrailSurrogate() {
    if(state != IN_NORMALIZED) { return 0; }
    U_ASSERT(pos < normalized.length());
    UChar trail;
    if(U16_IS_TRAIL(trail = normalized[pos])) { ++pos; }
    return trail;
}

float ShapeResultSpacing::computeSpacing(const TextRun& run,
        size_t index,
        float& offset) {
    UChar32 character = run[index];
    bool treatAsSpace =
        (Character::treatAsSpace(character) ||
         (m_normalizeSpace &&
          Character::isNormalizedCanvasSpaceCharacter(character))) &&
        (character != '\t' || !m_allowTabs);
    if (treatAsSpace && character != noBreakSpaceCharacter)
        character = spaceCharacter;

    float spacing = 0;
    if (m_letterSpacing && !Character::treatAsZeroWidthSpace(character))
        spacing += m_letterSpacing;

    if (treatAsSpace &&
            (index || !isFirstRun(run) || character == noBreakSpaceCharacter))
        spacing += m_wordSpacing;

    if (!hasExpansion())
        return spacing;

    if (treatAsSpace)
        return spacing + nextExpansion();

    if (run.is8Bit() || m_textJustify != TextJustify::TextJustifyAuto)
        return spacing;

    // isCJKIdeographOrSymbol() has expansion opportunities both before and
    // after each character.
    // http://www.w3.org/TR/jlreq/#line_adjustment
    if (U16_IS_LEAD(character) && index + 1 < run.length() &&
            U16_IS_TRAIL(run[index + 1]))
        character = U16_GET_SUPPLEMENTARY(character, run[index + 1]);
    if (!Character::isCJKIdeographOrSymbol(character)) {
        m_isAfterExpansion = false;
        return spacing;
    }

    if (!m_isAfterExpansion) {
        // Take the expansion opportunity before this ideograph.
        float expandBefore = nextExpansion();
        if (expandBefore) {
            offset += expandBefore;
            spacing += expandBefore;
        }
        if (!hasExpansion())
            return spacing;
    }

    return spacing + nextExpansion();
}

U_CAPI int32_t U_EXPORT2
u_countChar32(const UChar *s, int32_t length) {
    int32_t count;

    if(s==NULL || length<-1) {
        return 0;
    }

    count=0;
    if(length>=0) {
        while(length>0) {
            ++count;
            if(U16_IS_LEAD(*s) && length>=2 && U16_IS_TRAIL(*(s+1))) {
                s+=2;
                length-=2;
            } else {
                ++s;
                --length;
            }
        }
    } else /* length==-1 */ {
        UChar c;

        for(;;) {
            if((c=*s++)==0) {
                break;
            }
            ++count;

            /*
             * sufficient to look ahead one because of UTF-16;
             * safe to look ahead one because at worst that would be the terminating NUL
             */
            if(U16_IS_LEAD(c) && U16_IS_TRAIL(*s)) {
                ++s;
            }
        }
    }
    return count;
}

bool UTF16TextIterator::isValidSurrogatePair(UChar32& character)
{
    // If we have a surrogate pair, make sure it starts with the high part.
    if (!U16_IS_SURROGATE_LEAD(character))
        return false;

    // Do we have a surrogate pair? If so, determine the full Unicode (32 bit)
    // code point before glyph lookup.
    // Make sure we have another character and it's a low surrogate.
    if (m_characters + 1 >= m_charactersEnd)
        return false;

    UChar low = m_characters[1];
    if (!U16_IS_TRAIL(low))
        return false;
    return true;
}

FcPattern* createFontConfigPatternForCharacters(const UChar* characters, int length)
{
    FcPattern* pattern = FcPatternCreate();

    FcCharSet* fontConfigCharSet = FcCharSetCreate();
    for (int i = 0; i < length; ++i) {
        if (U16_IS_SURROGATE(characters[i]) && U16_IS_SURROGATE_LEAD(characters[i])
                && i != length - 1 && U16_IS_TRAIL(characters[i + 1])) {
            FcCharSetAddChar(fontConfigCharSet, U16_GET_SUPPLEMENTARY(characters[i], characters[i+1]));
            i++;
        } else
            FcCharSetAddChar(fontConfigCharSet, characters[i]);
    }
    FcPatternAddCharSet(pattern, FC_CHARSET, fontConfigCharSet);
    FcCharSetDestroy(fontConfigCharSet);

    FcPatternAddBool(pattern, FC_SCALABLE, FcTrue);
    FcConfigSubstitute(0, pattern, FcMatchPattern);
    FcDefaultSubstitute(pattern);
    return pattern;
}

bool readUTFChar(const UChar* str, int* begin, int length, unsigned* codePoint)
{
    if (U16_IS_SURROGATE(str[*begin])) {
        if (!U16_IS_SURROGATE_LEAD(str[*begin]) || *begin + 1 >= length || !U16_IS_TRAIL(str[*begin + 1])) {
            // Invalid surrogate pair.
            *codePoint = kUnicodeReplacementCharacter;
            return false;
        }

        // Valid surrogate pair.
        *codePoint = U16_GET_SUPPLEMENTARY(str[*begin], str[*begin + 1]);
        (*begin)++;
    } else {
        // Not a surrogate, just one 16-bit word.
        *codePoint = str[*begin];
    }

    if (U_IS_UNICODE_CHAR(*codePoint))
        return true;

    // Invalid code point.
    *codePoint = kUnicodeReplacementCharacter;
    return false;
}

// FIXME: This function may not work if the emphasis mark uses a complex script, but none of the
// standard emphasis marks do so.
bool Font::getEmphasisMarkGlyphData(const AtomicString& mark, GlyphData& glyphData) const
{
    if (mark.isEmpty())
        return false;

    UChar32 character = mark[0];

    if (U16_IS_SURROGATE(character)) {
        if (!U16_IS_SURROGATE_LEAD(character))
            return false;

        if (mark.length() < 2)
            return false;

        UChar low = mark[1];
        if (!U16_IS_TRAIL(low))
            return false;

        character = U16_GET_SUPPLEMENTARY(character, low);
    }

    glyphData = glyphDataForCharacter(character, false, EmphasisMarkVariant);
    return true;
}