C++ std::u16string类代码示例

void FontAtlas::findNewCharacters(const std::u16string& u16Text, std::unordered_map<unsigned short, unsigned short>& charCodeMap)
{
    std::u16string newChars;
    FT_Encoding charEncoding = _fontFreeType->getEncoding();

    //find new characters
    if (_letterDefinitions.empty())
    {
        // fixed #16169: new android project crash in android 5.0.2 device (Nexus 7) when use 3.12.
        // While using clang compiler with gnustl_static on android, the copy assignment operator of `std::u16string`
        // will affect the memory validity, it means after `newChars` is destroyed, the memory of `u16Text` holds
        // will be a dead region. `u16text` represents the variable in `Label::_utf16Text`, when somewhere
        // allocates memory by `malloc, realloc, new, new[]`, the generated memory address may be the same
        // as `Label::_utf16Text` holds. If doing a `memset` or other memory operations, the orignal `Label::_utf16Text`
        // will be in an unknown state. Meanwhile, a bunch lots of logic which depends on `Label::_utf16Text`
        // will be broken.

        // newChars = u16Text;

        // Using `append` method is a workaround for this issue. So please be carefuly while using the assignment operator
        // of `std::u16string`.
        newChars.append(u16Text);
    }
    else
    {
        auto length = u16Text.length();
        newChars.reserve(length);
        for (size_t i = 0; i < length; ++i)
        {
            auto outIterator = _letterDefinitions.find(u16Text[i]);
            if (outIterator == _letterDefinitions.end())
            {
                newChars.push_back(u16Text[i]);
            }
        }
    }

    if (!newChars.empty())
    {
        switch (charEncoding)
        {
        case FT_ENCODING_UNICODE:
        {
            for (auto u16Code : newChars)
            {
                charCodeMap[u16Code] = u16Code;
            }
            break;
        }
        case FT_ENCODING_GB2312:
        {
            conversionU16TOGB2312(newChars, charCodeMap);
            break;
        }
        default:
            OUTPUT_LOG("FontAtlas::findNewCharacters: Unsupported encoding:%d", charEncoding);
            break;
        }
    }
}

bool mapToPixelLength(std::u16string& value)
{
    if (stripLeadingWhitespace(value).empty())
        return false;
    const char16_t* input = value.c_str();
    const char16_t* end = input + value.length();
    int u;
    end = parseInt(input, end, u);
    if (!end || u < 0)
        return false;
    if (0 < u)
        value.replace(end - input, std::u16string::npos, u"px");
    else
        value.erase(end - input);
    return true;
}

bool UTF8ToUTF16(const std::string& utf8, std::u16string& outUtf16)
{
    if (utf8.empty())
    {
        outUtf16.clear();
        return true;
    }

    bool ret = false;
    
    const size_t utf16Bytes = (utf8.length()+1) * sizeof(char16_t);
    char16_t* utf16 = (char16_t*)malloc(utf16Bytes);
    memset(utf16, 0, utf16Bytes);

    char* utf16ptr = reinterpret_cast<char*>(utf16);
    const UTF8* error = NULL;

    if (llvm::ConvertUTF8toWide(2, utf8, utf16ptr, error))
    {
        outUtf16 = utf16;
        ret = true;
    }

    free(utf16);

    return ret;
}

void MediaListImp::setMediaText(const std::u16string& mediaText)
{
    clear();
    if (!mediaText.empty()) {
        CSSParser parser;
        *this = *parser.parseMediaList(mediaText);
    }
}

void KMX_Environment::Load(std::u16string const & key, std::u16string const & value) {
  assert(!key.empty());

  if (!u16icmp(key.c_str(), KM_KBP_KMX_ENV_PLATFORM)) {
    _platform = value;
  }
  else if (!u16icmp(key.c_str(), KM_KBP_KMX_ENV_BASELAYOUT)) {
    _baseLayout = value;
  }
  else if (!u16icmp(key.c_str(), KM_KBP_KMX_ENV_BASELAYOUTALT)) {
    _baseLayoutAlt = value;
  }
  else if (!u16icmp(key.c_str(), KM_KBP_KMX_ENV_SIMULATEALTGR)) {
    _simulateAltGr = value == u"1";
  }
  else if (!u16icmp(key.c_str(), KM_KBP_KMX_ENV_CAPSLOCK)) {
    _capsLock = value == u"1";
  }
  else if (!u16icmp(key.c_str(), KM_KBP_KMX_ENV_BASELAYOUTGIVESCTRLRALTFORRALT)) {
    _baseLayoutGivesCtrlRAltForRAlt = value == u"1";
  }
  else {
    // Unsupported key
    assert(false);
  }
}

    TextFlow::Word TextFlow::calculateWord(std::u16string content, float scale) const
    {
        // Empty word
        Word word;
        word.spVertices = std::shared_ptr<std::vector<glm::vec3> >(new std::vector<glm::vec3>);
        word.spTextureCoordinates = std::shared_ptr<std::vector<glm::vec2> >(new std::vector<glm::vec2>);

        // Fill word with data
        float xPixelPen = 0;
        for (uint i = 0; i < content.size(); i++)
        {
            Glyph const * pGlyph = mpFont->getGlyph(mFontSize, content[i]);
            if (pGlyph == NULL)
            {
                throwWarning(
                    OperationNotifier::Operation::RUNTIME,
                    "TextFlow has character in content not covered by character set");
                continue;
            }

            float yPixelPen = 0 - (scale * (float)(pGlyph->size.y - pGlyph->bearing.y));

            // Vertices for this quad
            glm::vec3 vertexA = glm::vec3(xPixelPen, yPixelPen, 0);
            glm::vec3 vertexB = glm::vec3(xPixelPen + (scale * pGlyph->size.x), yPixelPen, 0);
            glm::vec3 vertexC = glm::vec3(xPixelPen + (scale * pGlyph->size.x), yPixelPen + (scale * pGlyph->size.y), 0);
            glm::vec3 vertexD = glm::vec3(xPixelPen, yPixelPen + (scale * pGlyph->size.y), 0);

            // Texture coordinates for this quad
            glm::vec2 textureCoordinateA = glm::vec2(pGlyph->atlasPosition.x, pGlyph->atlasPosition.y);
            glm::vec2 textureCoordinateB = glm::vec2(pGlyph->atlasPosition.z, pGlyph->atlasPosition.y);
            glm::vec2 textureCoordinateC = glm::vec2(pGlyph->atlasPosition.z, pGlyph->atlasPosition.w);
            glm::vec2 textureCoordinateD = glm::vec2(pGlyph->atlasPosition.x, pGlyph->atlasPosition.w);

            xPixelPen += scale * pGlyph->advance.x;

            // Fill into data blocks
            word.spVertices->push_back(vertexA);
            word.spVertices->push_back(vertexB);
            word.spVertices->push_back(vertexC);
            word.spVertices->push_back(vertexC);
            word.spVertices->push_back(vertexD);
            word.spVertices->push_back(vertexA);

            word.spTextureCoordinates->push_back(textureCoordinateA);
            word.spTextureCoordinates->push_back(textureCoordinateB);
            word.spTextureCoordinates->push_back(textureCoordinateC);
            word.spTextureCoordinates->push_back(textureCoordinateC);
            word.spTextureCoordinates->push_back(textureCoordinateD);
            word.spTextureCoordinates->push_back(textureCoordinateA);
        }

        // Set width of whole word
        word.pixelWidth = xPixelPen;

        return word;
    }

bool UTF16ToUTF8(const std::u16string& utf16, std::string& outUtf8)
{
    if (utf16.empty())
    {
        outUtf8.clear();
        return true;
    }

    return llvm::convertUTF16ToUTF8String(utf16, outUtf8);
}

Value::
Value(const std::u16string&  s):
type(ConstCharType().make_array(s.size()+1)),
base_address(0),
size(sizeof(char16_t)*(s.size()+1))
{
  data = new uint8_t[size];

  auto  dst = reinterpret_cast<char16_t*>(data);
  auto  src = s.cbegin();

    while(*src)
    {
      *dst++ = *src++;
    }


  dst = 0;
}

void TextLabels::renderString( const std::u16string &str, vec2 *cursor, float stretch )
{
	std::u16string::const_iterator itr;
	for( itr = str.begin(); itr != str.end(); ++itr ) {
		// retrieve character code
		uint16_t id = (uint16_t)*itr;

		if( mFont->contains( id ) ) {
			// get metrics for this character to speed up measurements
			Font::Metrics m = mFont->getMetrics( id );

			// skip whitespace characters
			if( !isWhitespaceUtf16( id ) ) {
				size_t index = mVertices.size();

				Rectf bounds = mFont->getBounds( m, mFontSize );
				mVertices.push_back( vec3( *cursor + bounds.getUpperLeft(), 0 ) );
				mVertices.push_back( vec3( *cursor + bounds.getUpperRight(), 0 ) );
				mVertices.push_back( vec3( *cursor + bounds.getLowerRight(), 0 ) );
				mVertices.push_back( vec3( *cursor + bounds.getLowerLeft(), 0 ) );

				bounds = mFont->getTexCoords( m );
				mTexcoords.push_back( bounds.getUpperLeft() );
				mTexcoords.push_back( bounds.getUpperRight() );
				mTexcoords.push_back( bounds.getLowerRight() );
				mTexcoords.push_back( bounds.getLowerLeft() );

				mIndices.push_back( index + 0 ); mIndices.push_back( index + 3 ); mIndices.push_back( index + 1 );
				mIndices.push_back( index + 1 ); mIndices.push_back( index + 3 ); mIndices.push_back( index + 2 );

				mOffsets.insert( mOffsets.end(), 4, mOffset );
			}

			if( id == 32 )
				cursor->x += stretch * mFont->getAdvance( m, mFontSize );
			else
				cursor->x += mFont->getAdvance( m, mFontSize );
		}
	}

	//
	mBoundsInvalid = true;
}

void UserDictionaryDecoder::DecodeHeading(IBitStream *bstr, unsigned len, std::u16string &res) {
    res.clear();
    unsigned symIdx;
    for (size_t i = 0; i < len; i++) {
        _ltHeadings.Decode(*bstr, symIdx);
        unsigned sym = _headingSymbols.at(symIdx);
        assert(sym <= 0xffff);
        res += (char16_t)sym;
    }
}

bool RegexStore::patternMatch(std::u16string& in_string) {
	//No pattern matches when we don't have a valid pattern
	if (not is_compiled) {
    return false;
	}

  //Check for a match
  regmatch_t pmatch;
	std::string tmp_str(in_string.begin(), in_string.end());
  int match = regexec(&exp, tmp_str.c_str(), 1, &pmatch, 0);
	//Make sure that each character was matched and that the entire input string
	//was consumed by the pattern
  if (0 == match and 0 == pmatch.rm_so and in_string.size() == pmatch.rm_eo) {
		return true;
  }
  else {
		return false;
  }
}

	SimpleString::SimpleString(const std::u16string _str)
	{
		length = _str.size();
		s_value = new char16_t[length + 1];
		
		for (unsigned int i = 0; i < _str.size(); ++i)
			s_value[i] = _str[i];

		s_value[length] = u'\0';

		// calculate hash;
		_hash = 5381;
		int c = 0;

		char16_t * sptr = s_value;

		while ((c = *sptr++))
			_hash = ((_hash << 5) + _hash) + c;
	}

std::vector<char16_t> getChar16VectorFromUTF16String(const std::u16string& utf16)
{
    std::vector<char16_t> ret;
    size_t len = utf16.length();
    ret.reserve(len);
    for (size_t i = 0; i < len; ++i)
    {
        ret.push_back(utf16[i]);
    }
    return ret;
}

bool convertUTF16ToUTF8String(const std::u16string& utf16, std::string &Out) {
  assert(Out.empty());

  // Avoid OOB by returning early on empty input.
  if (utf16.empty())
    return true;

  const UTF16 *Src = reinterpret_cast<const UTF16 *>(utf16.data());
  const UTF16 *SrcEnd = reinterpret_cast<const UTF16 *>(utf16.data() + utf16.length());

  // Byteswap if necessary.
  std::vector<UTF16> ByteSwapped;
  if (Src[0] == UNI_UTF16_BYTE_ORDER_MARK_SWAPPED) {
    ByteSwapped.insert(ByteSwapped.end(), Src, SrcEnd);
    for (size_t I = 0, E = ByteSwapped.size(); I != E; ++I)
      ByteSwapped[I] = SwapByteOrder_16(ByteSwapped[I]);
    Src = &ByteSwapped[0];
    SrcEnd = &ByteSwapped[ByteSwapped.size() - 1] + 1;
  }

  // Skip the BOM for conversion.
  if (Src[0] == UNI_UTF16_BYTE_ORDER_MARK_NATIVE)
    Src++;

  // Just allocate enough space up front.  We'll shrink it later.
  Out.resize(utf16.length() * UNI_MAX_UTF8_BYTES_PER_CODE_POINT + 1);
  UTF8 *Dst = reinterpret_cast<UTF8 *>(&Out[0]);
  UTF8 *DstEnd = Dst + Out.size();

  ConversionResult CR =
      ConvertUTF16toUTF8(&Src, SrcEnd, &Dst, DstEnd, strictConversion);
  assert(CR != targetExhausted);

  if (CR != conversionOK) {
    Out.clear();
    return false;
  }

  Out.resize(reinterpret_cast<char *>(Dst) - &Out[0]);
  return true;
}

std::string to_utf8(const std::u16string& str)
{
    icu_handle::get();
    static_assert(sizeof(char16_t) == sizeof(UChar),
                  "Invalid UChar definition in ICU");
    // looks dangerous, but isn't: UChar is guaranteed to be a 16-bit
    // integer type, so all we're doing here is going between signed vs.
    // unsigned
    auto buf = reinterpret_cast<const UChar*>(str.c_str());
    icu::UnicodeString u16str{buf};
    return icu_to_u8str(u16str);
}