/src/logging-log4cxx/src/main/cpp/charsetencoder.cpp

Source
/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
#include <log4cxx/logstring.h>
#include <log4cxx/helpers/charsetencoder.h>
#include <log4cxx/private/bytebuffer_priv.h>
#include <log4cxx/helpers/exception.h>
#include <apr_xlate.h>
#include <log4cxx/helpers/stringhelper.h>
#include <log4cxx/helpers/transcoder.h>
#include <algorithm>

#if !defined(LOG4CXX)
  #define LOG4CXX 1
#endif

#include <log4cxx/private/log4cxx_private.h>
#include <apr_portable.h>
#include <mutex>

#ifdef LOG4CXX_HAS_WCSTOMBS
  #include <stdlib.h>
#endif

#if 15 < LOG4CXX_ABI_VERSION
#define LOG4CXX_16_VIRTUAL_SPECIFIER override
#else
#define LOG4CXX_16_VIRTUAL_SPECIFIER
#endif

using namespace LOG4CXX_NS;
using namespace LOG4CXX_NS::helpers;

IMPLEMENT_LOG4CXX_OBJECT(CharsetEncoder)

namespace LOG4CXX_NS
{

namespace helpers
{

#if APR_HAS_XLATE
/**
* A character encoder implemented using apr_xlate.
*/
class APRCharsetEncoder : public CharsetEncoder
{
  public:
    APRCharsetEncoder(const LogString& topage) : pool()
    {
#if LOG4CXX_LOGCHAR_IS_WCHAR
      const char* frompage = "WCHAR_T";
#endif
#if LOG4CXX_LOGCHAR_IS_UTF8
      const char* frompage = "UTF-8";
#endif
#if LOG4CXX_LOGCHAR_IS_UNICHAR
      const char* frompage = "UTF-16";
#endif
      std::string tpage(Transcoder::encodeCharsetName(topage));
      apr_status_t stat = apr_xlate_open(&convset,
          tpage.c_str(),
          frompage,
          pool.getAPRPool());

      if (stat != APR_SUCCESS)
      {
        throw IllegalArgumentException(topage);
      }
    }

    virtual ~APRCharsetEncoder()
    {
    }

    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      apr_status_t stat;
      size_t outbytes_left = out.remaining();
      size_t initial_outbytes_left = outbytes_left;
      size_t position = out.position();

      if (iter == in.end())
      {
        std::lock_guard<std::mutex> lock(mutex);
        stat = apr_xlate_conv_buffer(convset, NULL, NULL,
            out.data() + position, &outbytes_left);
      }
      else
      {
        LogString::size_type inOffset = (iter - in.begin());
        apr_size_t inbytes_left =
          (in.size() - inOffset) * sizeof(LogString::value_type);
        apr_size_t initial_inbytes_left = inbytes_left;
        {
          std::lock_guard<std::mutex> lock(mutex);
          stat = apr_xlate_conv_buffer(convset,
              (const char*) (in.data() + inOffset),
              &inbytes_left,
              out.data() + position,
              &outbytes_left);
        }
        iter += ((initial_inbytes_left - inbytes_left) / sizeof(LogString::value_type));
      }

      out.increment_position((initial_outbytes_left - outbytes_left));
      return stat;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (codePoint <= 0x10FFFF && 4 <= out.remaining())
        out.increment_position(putUTF8CodePoint(codePoint, out.current()));
      else
        result = APR_BADARG;
      return result;
    }

  private:
    APRCharsetEncoder(const APRCharsetEncoder&);
    APRCharsetEncoder& operator=(const APRCharsetEncoder&);
    Pool pool;
    std::mutex mutex;
    apr_xlate_t* convset;
};
#endif

#if LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS
/**
 *  A character encoder implemented using wcstombs.
*/
class WcstombsCharsetEncoder : public CharsetEncoder
{
  public:
    WcstombsCharsetEncoder()
    {
    }

    /**
     *   Converts a wchar_t to the default external multibyte encoding.
     */
    log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      log4cxx_status_t stat = APR_SUCCESS;

      if (iter != in.end())
      {
        size_t outbytes_left = out.remaining();
        size_t position = out.position();
        std::wstring::size_type inOffset = (iter - in.begin());
        enum { BUFSIZE = 256 };
        wchar_t buf[BUFSIZE];
        size_t chunkSize = BUFSIZE - 1;

        if (chunkSize * MB_LEN_MAX > outbytes_left)
        {
          chunkSize = outbytes_left / MB_LEN_MAX;
        }

        if (chunkSize > in.length() - inOffset)
        {
          chunkSize = in.length() - inOffset;
        }

        memset(buf, 0, BUFSIZE * sizeof(wchar_t));
        memcpy(buf,
          in.data() + inOffset,
          chunkSize * sizeof(wchar_t));
        size_t converted = wcstombs(out.data() + position, buf, outbytes_left);

        if (converted == (size_t) -1)
        {
          stat = APR_BADARG;

          //
          //   if unconvertable character was encountered
          //       repeatedly halve source to get fragment that
          //       can be converted
          for (chunkSize /= 2;
            chunkSize > 0;
            chunkSize /= 2)
          {
            buf[chunkSize] = 0;
            converted = wcstombs(out.data() + position, buf, outbytes_left);

            if (converted != (size_t) -1)
            {
              iter += chunkSize;
              out.increment_position(converted);
              break;
            }
          }
        }
        else
        {
          iter += chunkSize;
          out.increment_position(converted);
        }
      }

      return stat;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (codePoint <= 0x10FFFF && MB_LEN_MAX <= out.remaining())
      {
        auto ch = static_cast<wchar_t>(codePoint);
        auto converted = wcstombs(out.current(), &ch, 1);
        if (static_cast<std::size_t>(-1) == converted)
          result = APR_BADARG;
        else
          out.increment_position(converted);
      }
      else
        result = APR_BADARG;
      return result;
    }

  private:
    WcstombsCharsetEncoder(const WcstombsCharsetEncoder&);
    WcstombsCharsetEncoder& operator=(const WcstombsCharsetEncoder&);
};
#endif


/**
*   Encodes a LogString to US-ASCII.
*/
class USASCIICharsetEncoder : public CharsetEncoder
{
  public:
    USASCIICharsetEncoder()
    {
    }

    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      log4cxx_status_t stat = APR_SUCCESS;

      if (iter != in.end())
      {
        while (out.remaining() > 0 && iter != in.end())
        {
          LogString::const_iterator prev(iter);
          unsigned int sv = Transcoder::decode(in, iter);

          if (sv <= 0x7F)
          {
            out.put((char) sv);
          }
          else
          {
            iter = prev;
            stat = APR_BADARG;
            break;
          }
        }
      }

      return stat;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (out.remaining() < 1)
        result = APR_BADARG;
      else if (codePoint <= 0x7F)
        out.put(static_cast<char>(codePoint));
      else if (Transcoder::LOSSCHAR == codePoint)
        out.put('?');
      else
        result = APR_BADARG;
      return result;
    }

  private:
    USASCIICharsetEncoder(const USASCIICharsetEncoder&);
    USASCIICharsetEncoder& operator=(const USASCIICharsetEncoder&);
};

/**
*   Converts a LogString to ISO-8859-1.
*/
class ISOLatinCharsetEncoder : public CharsetEncoder
{
  public:
    ISOLatinCharsetEncoder()
    {
    }

    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      log4cxx_status_t stat = APR_SUCCESS;

      while (out.remaining() > 0 && iter != in.end())
      {
        LogString::const_iterator prev(iter);
        unsigned int sv = Transcoder::decode(in, iter);
        if (sv <= 0xFF)
          out.put(static_cast<char>(sv));
        else
        {
          iter = prev;
          stat = APR_BADARG;
          break;
        }
      }
      return stat;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (out.remaining() < 1)
        result = APR_BADARG;
      else if (codePoint <= 0xFF)
        out.put(static_cast<char>(codePoint));
      else if (Transcoder::LOSSCHAR == codePoint)
        out.put('?');
      else
        result = APR_BADARG;
      return result;
    }

  private:
    ISOLatinCharsetEncoder(const ISOLatinCharsetEncoder&);
    ISOLatinCharsetEncoder& operator=(const ISOLatinCharsetEncoder&);
};

/**
*   Encodes a LogString to a byte array when the encodings are identical.
*/
class TrivialCharsetEncoder : public CharsetEncoder
{
  public:
    TrivialCharsetEncoder()
    {
    }


    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      if (iter != in.end())
      {
        size_t requested = in.length() - (iter - in.begin());

        if (requested > out.remaining() / sizeof(logchar))
        {
          requested = out.remaining() / sizeof(logchar);
        }

        memcpy(out.current(),
          (const char*) in.data() + (iter - in.begin()),
          requested * sizeof(logchar));
        iter += requested;
        out.increment_position(requested * sizeof(logchar));
      }

      return APR_SUCCESS;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (out.remaining() < 1)
        result = APR_BADARG;
      else if (codePoint <= 0xFF)
        out.put(static_cast<char>(codePoint));
      else if (Transcoder::LOSSCHAR == codePoint)
        out.put('?');
      else
        result = APR_BADARG;
      return result;
    }

  private:
    TrivialCharsetEncoder(const TrivialCharsetEncoder&);
    TrivialCharsetEncoder& operator=(const TrivialCharsetEncoder&);
};

/**
 *  Converts a LogString to UTF-8.
 */
class UTF8CharsetEncoder : public CharsetEncoder
{
  public:
    UTF8CharsetEncoder()
    {
    }

    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      while (iter != in.end() && 4 <= out.remaining())
      {
        auto sv = Transcoder::getCodePoint(in, iter);
        out.increment_position(putUTF8CodePoint(sv, out.current()));
      }

      return APR_SUCCESS;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (codePoint <= 0x10FFFF && 4 <= out.remaining())
        out.increment_position(putUTF8CodePoint(codePoint, out.current()));
      else
        result = APR_BADARG;
      return result;
    }

  private:
    UTF8CharsetEncoder(const UTF8CharsetEncoder&);
    UTF8CharsetEncoder& operator=(const UTF8CharsetEncoder&);
};

/**
 *   Encodes a LogString to UTF16-BE.
 */
class UTF16BECharsetEncoder : public CharsetEncoder
{
  public:
    UTF16BECharsetEncoder()
    {
    }

    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      while (iter != in.end() && 4 <= out.remaining())
      {
        auto sv = Transcoder::getCodePoint(in, iter);
        out.increment_position(putUTF16BECodePoint(sv, out.current()));
      }

      return APR_SUCCESS;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (codePoint <= 0x10FFFF && 4 <= out.remaining())
        Transcoder::encodeUTF16BE(codePoint, out);
      else
        result = APR_BADARG;
      return result;
    }

  private:
    UTF16BECharsetEncoder(const UTF16BECharsetEncoder&);
    UTF16BECharsetEncoder& operator=(const UTF16BECharsetEncoder&);
};

/**
 *   Encodes a LogString to UTF16-LE.
 */
class UTF16LECharsetEncoder : public CharsetEncoder
{
  public:
    UTF16LECharsetEncoder()
    {
    }


    virtual log4cxx_status_t encode(const LogString& in,
      LogString::const_iterator& iter,
      ByteBuffer& out_param) override
    {
      auto& out = out_param.impl();
      while (iter != in.end() && 4 <= out.remaining())
      {
        auto sv = Transcoder::getCodePoint(in, iter);
        out.increment_position(putUTF16LECodePoint(sv, out.current()));
      }

      return APR_SUCCESS;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (codePoint <= 0x10FFFF && 4 <= out.remaining())
        out.increment_position(putUTF16LECodePoint(codePoint, out.current()));
      else
        result = APR_BADARG;
      return result;
    }
  private:
    UTF16LECharsetEncoder(const UTF16LECharsetEncoder&);
    UTF16LECharsetEncoder& operator=(const UTF16LECharsetEncoder&);
};

/**
 *    Charset encoder that uses current locale settings.
 */
class LocaleCharsetEncoder : public CharsetEncoder
{
  public:
    LocaleCharsetEncoder() : state()
    {
    }
    log4cxx_status_t encode
      ( const LogString&           in
      , LogString::const_iterator& nextCodePoint
      , ByteBuffer&                out_param
      ) override
    {
      auto& out = out_param.impl();
      log4cxx_status_t result = APR_SUCCESS;
#if !LOG4CXX_CHARSET_EBCDIC
      char* current = out.current();
      size_t availableByteCount = out.remaining();
      size_t byteCount = 0;
      if (std::mbsinit(&this->state)) // ByteBuffer not partially encoded?
      {
        // Copy single byte characters
        for (;
          nextCodePoint != in.end() && byteCount < availableByteCount && static_cast<unsigned int>(*nextCodePoint) < 0x80;
          ++nextCodePoint, ++byteCount, ++current)
        {
          *current = static_cast<char>(*nextCodePoint);
        }
      }
#endif
      // Encode characters that may require multiple bytes
      while (nextCodePoint != in.end() && byteCount < availableByteCount && MB_CUR_MAX <= (availableByteCount - byteCount))
      {
        auto ch = Transcoder::getCodePoint(in, nextCodePoint);
        auto n = std::wcrtomb(current, ch, &this->state);
        if (static_cast<std::size_t>(-1) == n) // not a valid wide character?
        {
          result = APR_BADARG;
          break;
        }
        byteCount += n;
        current += n;
      }
      out.increment_position(byteCount);
      return result;
    }

    /**
     * Add onto \c out an encoded equivalent of \c codePoint.
     */
    log4cxx_status_t encode(unsigned int codePoint, ByteBuffer& out) LOG4CXX_16_VIRTUAL_SPECIFIER
    {
      apr_status_t result = APR_SUCCESS;
      if (MB_CUR_MAX <= out.remaining())
      {
        auto n = std::wcrtomb(out.current(), codePoint, &this->state);
        if (static_cast<std::size_t>(-1) == n) // not a valid wide character?
          result = APR_BADARG;
        else
          out.increment_position(n);
      }
      else
        result = APR_BADARG;
      return result;
    }

  private:
    std::mbstate_t state;
};


} // namespace helpers

}  //namespace log4cxx



CharsetEncoder::CharsetEncoder()
{
}

CharsetEncoder::~CharsetEncoder()
{
}

CharsetEncoderPtr CharsetEncoder::getDefaultEncoder()
{
  static WideLife<CharsetEncoderPtr> encoder(createDefaultEncoder());

  //
  //  if invoked after static variable destruction
  //     (if logging is called in the destructor of a static object)
  //     then create a new decoder.
  //
  if (encoder.value() == 0)
  {
    return CharsetEncoderPtr( createDefaultEncoder() );
  }

  return encoder;
}

CharsetEncoder* CharsetEncoder::createDefaultEncoder()
{
#if LOG4CXX_CHARSET_UTF8
#if LOG4CXX_LOGCHAR_IS_UTF8
  return new TrivialCharsetEncoder();
#else
  return new UTF8CharsetEncoder();
#endif
#elif LOG4CXX_CHARSET_ISO88591
  return new ISOLatinCharsetEncoder();
#elif LOG4CXX_CHARSET_USASCII
  return new USASCIICharsetEncoder();
#elif LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS
  return new WcstombsCharsetEncoder();
#else
  return new LocaleCharsetEncoder();
#endif
}


CharsetEncoderPtr CharsetEncoder::getUTF8Encoder()
{
  return std::make_shared<UTF8CharsetEncoder>();
}



CharsetEncoderPtr CharsetEncoder::getEncoder(const LogString& charset)
{
  if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-8"), LOG4CXX_STR("utf-8"))
    || StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP65001"), LOG4CXX_STR("cp65001")))
  {
    return std::make_shared<UTF8CharsetEncoder>();
  }
  else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("C"), LOG4CXX_STR("c")) ||
    charset == LOG4CXX_STR("646") ||
    StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("US-ASCII"), LOG4CXX_STR("us-ascii")) ||
    StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ISO646-US"), LOG4CXX_STR("iso646-US")) ||
    StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ANSI_X3.4-1968"), LOG4CXX_STR("ansi_x3.4-1968")) ||
    StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP20127"), LOG4CXX_STR("cp20127")))
  {
    return std::make_shared<USASCIICharsetEncoder>();
  }
  else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ISO-8859-1"), LOG4CXX_STR("iso-8859-1")) ||
    StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("ISO-LATIN-1"), LOG4CXX_STR("iso-latin-1")) ||
    StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP1252"), LOG4CXX_STR("cp1252")))
  {
    return std::make_shared<ISOLatinCharsetEncoder>();
  }
  else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-16BE"), LOG4CXX_STR("utf-16be"))
    || StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-16"), LOG4CXX_STR("utf-16"))
    || StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("CP1200"), LOG4CXX_STR("cp1200")))
  {
    return std::make_shared<UTF16BECharsetEncoder>();
  }
  else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("UTF-16LE"), LOG4CXX_STR("utf-16le")))
  {
    return std::make_shared<UTF16LECharsetEncoder>();
  }
  else if (StringHelper::equalsIgnoreCase(charset, LOG4CXX_STR("LOCALE"), LOG4CXX_STR("locale")))
  {
    return std::make_shared<LocaleCharsetEncoder>();
  }

#if APR_HAS_XLATE
  return std::make_shared<APRCharsetEncoder>(charset);
#else
  throw IllegalArgumentException(charset);
#endif
}


void CharsetEncoder::reset()
{
}

#if LOG4CXX_ABI_VERSION <= 15
void CharsetEncoder::flush(ByteBuffer& /* out */ )
{
}
#endif

void CharsetEncoder::encode(CharsetEncoderPtr& enc,
  const LogString& src,
  LogString::const_iterator& iter,
  ByteBuffer& dst)
{
  log4cxx_status_t stat = enc->encode(src, iter, dst);

  if (stat != APR_SUCCESS && iter != src.end())
  {
#if LOG4CXX_LOGCHAR_IS_WCHAR || LOG4CXX_LOGCHAR_IS_UNICHAR
    iter++;
#elif LOG4CXX_LOGCHAR_IS_UTF8

    //  advance past this character and all continuation characters
    do
    {
      ++iter;
    }
    while (iter != src.end() &&
         (*iter & 0xC0) == 0x80);

#else
#error logchar is unrecognized
#endif
#if 15 < LOG4CXX_ABI_VERSION
    enc->encode(Transcoder::LOSSCHAR, dst);
#else // LOG4CXX_ABI_VERSION <= 15
    if (auto p = dynamic_cast<TrivialCharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
    else if (auto p = dynamic_cast<UTF8CharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
    else if (auto p = dynamic_cast<LocaleCharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
    else if (auto p = dynamic_cast<USASCIICharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
    else if (auto p = dynamic_cast<ISOLatinCharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
    else if (auto p = dynamic_cast<UTF16BECharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
    else if (auto p = dynamic_cast<UTF16LECharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
#if LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS
    else if (auto p = dynamic_cast<WcstombsCharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
#endif //  LOG4CXX_LOGCHAR_IS_WCHAR && LOG4CXX_HAS_WCSTOMBS
#if APR_HAS_XLATE
    else if (auto p = dynamic_cast<APRCharsetEncoder*>(enc.get()))
      p->encode(Transcoder::LOSSCHAR, dst);
#endif //  APR_HAS_XLATE
#endif // LOG4CXX_ABI_VERSION <= 15
  }
}

bool CharsetEncoder::isTriviallyCopyable(const LogString& src, const CharsetEncoderPtr& enc)
{
  bool result;
#if !LOG4CXX_CHARSET_EBCDIC
  if (dynamic_cast<LocaleCharsetEncoder*>(enc.get()))
  {
    result = src.end() == std::find_if(src.begin(), src.end()
      , [](const logchar& ch) -> bool { return 0x80 <= (unsigned int)ch; });
  }
  else
#endif
    result = !!dynamic_cast<TrivialCharsetEncoder*>(enc.get());
  return result;
}

size_t CharsetEncoder::putUTF8CodePoint(unsigned int ch, char* dst)
{
  if (ch < 0x80)
  {
    dst[0] = (char) ch;
    return 1;
  }
  else if (ch < 0x800)
  {
    dst[0] = (char) (0xC0 + (ch >> 6));
    dst[1] = (char) (0x80 + (ch & 0x3F));
    return 2;
  }
  else if (ch < 0x10000)
  {
    dst[0] = (char) (0xE0 + (ch >> 12));
    dst[1] = (char) (0x80 + ((ch >> 6) & 0x3F));
    dst[2] = (char) (0x80 + (ch & 0x3F));
    return 3;
  }
  else if (ch <= 0x10FFFF)
  {
    dst[0] = (char) (0xF0 + (ch >> 18));
    dst[1] = (char) (0x80 + ((ch >> 12) & 0x3F));
    dst[2] = (char) (0x80 + ((ch >> 6) & 0x3F));
    dst[3] = (char) (0x80 + (ch & 0x3F));
    return 4;
  }
  else
  {
    //
    //  output UTF-8 encoding of 0xFFFF
    //
    dst[0] = (char) 0xEF;
    dst[1] = (char) 0xBF;
    dst[2] = (char) 0xBF;
    return 3;
  }
}

size_t CharsetEncoder::putUTF16BECodePoint(unsigned int ch, char* dst)
{
  if (ch <= 0xFFFF)
  {
    dst[0] = (char) (ch >> 8);
    dst[1] = (char) (ch & 0xFF);
    return 2;
  }

  if (ch <= 0x10FFFF)
  {
    unsigned char w = (unsigned char) ((ch >> 16) - 1);
    dst[0] = (char) (0xD8 + (w >> 2));
    dst[1] = (char) (((w & 0x03) << 6) + ((ch >> 10) & 0x3F));
    dst[2] = (char) (0xDC + ((ch >> 8) & 0x03));
    dst[3] = (char) (ch & 0xFF);
    return 4;
  }

  dst[0] = dst[1] = (char) 0xFF;
  return 2;
}

size_t CharsetEncoder::putUTF16LECodePoint(unsigned int ch, char* dst)
{
  if (ch <= 0xFFFF)
  {
    dst[1] = (char) (ch >> 8);
    dst[0] = (char) (ch & 0xFF);
    return 2;
  }

  if (ch <= 0x10FFFF)
  {
    unsigned char w = (unsigned char) ((ch >> 16) - 1);
    dst[1] = (char) (0xD8 + (w >> 2));
    dst[0] = (char) (((w & 0x03) << 6) + ((ch >> 10) & 0x3F));
    dst[3] = (char) (0xDC + ((ch >> 8) & 0x03));
    dst[2] = (char) (ch & 0xFF);
    return 4;
  }

  dst[0] = dst[1] = (char) 0xFF;
  return 2;
}

Coverage Report

Created: 2026-09-14 06:53