stringencode.cc

本人收集整理的一份c/c++跨平台网络库

        if (bufpos + esclen >= buflen) {
          break;
        }
        memcpy(buffer + bufpos, escseq, esclen);
        bufpos += esclen;
      } else {
        buffer[bufpos++] = ch;
      }
    } else {
      // Largest value is 0x1FFFFF => �  (10 characters)
      char escseq[11];
      unsigned long val;
      if (size_t vallen = utf8_decode(&source[srcpos], srclen - srcpos, &val)) {
        srcpos += vallen;
      } else {
        // Not a valid utf8 sequence, just use the raw character.
        val = static_cast<unsigned char>(source[srcpos++]);
      }
      size_t esclen = sprintfn(escseq, ARRAY_SIZE(escseq), "&#%lu;", val);
      if (bufpos + esclen >= buflen) {
        break;
      }
      memcpy(buffer + bufpos, escseq, esclen);
      bufpos += esclen;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t html_decode(char * buffer, size_t buflen,
                   const char * source, size_t srclen) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  return xml_decode(buffer, buflen, source, srclen);
}

size_t xml_encode(char * buffer, size_t buflen,
                  const char * source, size_t srclen) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    unsigned char ch = source[srcpos++];
    if ((ch < 128) && (ASCII_CLASS[ch] & XML_UNSAFE)) {
      const char * escseq = 0;
      size_t esclen = 0;
      switch (ch) {
        case '<':  escseq = "&lt;";   esclen = 4; break;
        case '>':  escseq = "&gt;";   esclen = 4; break;
        case '\'': escseq = "&apos;"; esclen = 6; break;
        case '\"': escseq = "&quot;"; esclen = 6; break;
        case '&':  escseq = "&amp;";  esclen = 5; break;
        default: ASSERT(false);
      }
      if (bufpos + esclen >= buflen) {
        break;
      }
      memcpy(buffer + bufpos, escseq, esclen);
      bufpos += esclen;
    } else {
      buffer[bufpos++] = ch;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t xml_decode(char * buffer, size_t buflen,
                  const char * source, size_t srclen) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos < srclen) && (bufpos + 1 < buflen)) {
    unsigned char ch = source[srcpos++];
    if (ch != '&') {
      buffer[bufpos++] = ch;
    } else if ((srcpos + 2 < srclen)
               && (memcmp(source + srcpos, "lt;", 3) == 0)) {
      buffer[bufpos++] = '<';
      srcpos += 3;
    } else if ((srcpos + 2 < srclen)
               && (memcmp(source + srcpos, "gt;", 3) == 0)) {
      buffer[bufpos++] = '>';
      srcpos += 3;
    } else if ((srcpos + 4 < srclen)
               && (memcmp(source + srcpos, "apos;", 5) == 0)) {
      buffer[bufpos++] = '\'';
      srcpos += 5;
    } else if ((srcpos + 4 < srclen)
               && (memcmp(source + srcpos, "quot;", 5) == 0)) {
      buffer[bufpos++] = '\"';
      srcpos += 5;
    } else if ((srcpos + 3 < srclen)
               && (memcmp(source + srcpos, "amp;", 4) == 0)) {
      buffer[bufpos++] = '&';
      srcpos += 4;
    } else if ((srcpos < srclen) && (source[srcpos] == '#')) {
      int int_base = 10;
      if ((srcpos + 1 < srclen) && (source[srcpos+1] == 'x')) {
        int_base = 16;
        srcpos += 1;
      }
      char * ptr;
      // TODO: Fix hack (ptr may go past end of data)
      unsigned long val = strtoul(source + srcpos + 1, &ptr, int_base);
      if ((static_cast<size_t>(ptr - source) < srclen) && (*ptr == ';')) {
        srcpos = ptr - source + 1;
      } else {
        // Not a valid escape sequence.
        break;
      }
      if (size_t esclen = utf8_encode(buffer + bufpos, buflen - bufpos, val)) {
        bufpos += esclen;
      } else {
        // Not enough room to encode the character, or illegal character
        break;
      }
    } else {
      // Unrecognized escape sequence.
      break;
    }
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t hex_encode(char * buffer, size_t buflen,
                  const char * csource, size_t srclen) {
  ASSERT(NULL != buffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  const unsigned char * bsource =
    reinterpret_cast<const unsigned char *>(csource);

  size_t srcpos = 0, bufpos = 0;
  srclen = _min(srclen, (buflen - 1) / 2);
  while (srcpos < srclen) {
    unsigned char ch = bsource[srcpos++];
    buffer[bufpos  ] = hex_encode((ch >> 4) & 0xF);
    buffer[bufpos+1] = hex_encode((ch     ) & 0xF);
    bufpos += 2;
  }
  buffer[bufpos] = '\0';
  return bufpos;
}

size_t hex_decode(char * cbuffer, size_t buflen,
                  const char * source, size_t srclen) {
  ASSERT(NULL != cbuffer);  // TODO: estimate output size
  if (buflen <= 0)
    return 0;

  unsigned char * bbuffer = reinterpret_cast<unsigned char *>(cbuffer);

  size_t srcpos = 0, bufpos = 0;
  while ((srcpos + 1 < srclen) && (bufpos + 1 < buflen)) {
    unsigned char v1 = (hex_decode(source[srcpos]) << 4);
    unsigned char v2 = hex_decode(source[srcpos+1]);
    bbuffer[bufpos++] =  v1 | v2;
    srcpos += 2;
  }
  bbuffer[bufpos] = '\0';
  return bufpos;
}

void transform(std::string& value, size_t maxlen, const std::string& source,
               Transform t) {
  char * buffer = static_cast<char *>(_alloca(maxlen + 1));
  value.assign(buffer, t(buffer, maxlen + 1, source.data(), source.length()));
}

std::string s_transform(const std::string& source, Transform t) {
  // Ask transformation function to approximate the destination size (returns upper bound)
  size_t maxlen = t(NULL, 0, source.data(), source.length());
  char * buffer = static_cast<char *>(_alloca(maxlen));
  size_t len = t(buffer, maxlen, source.data(), source.length());
  std::string result(buffer, len);
  return result;
}

char make_char_safe_for_filename(char c) {
  if (c < 32)
    return '_';

  switch (c) {
    case '<':
    case '>':
    case ':':
    case '"':
    case '/':
    case '\\':
    case '|':
    case '*':
    case '?':
      return '_';

    default:
      return c;
  }
}

/*
void sprintf(std::string& value, size_t maxlen, const char * format, ...) {
  char * buffer = static_cast<char *>(alloca(maxlen + 1));
  va_list args;
  va_start(args, format);
  value.assign(buffer, vsprintfn(buffer, maxlen + 1, format, args));
  va_end(args);
}
*/

/////////////////////////////////////////////////////////////////////////////
// Unit Tests
/////////////////////////////////////////////////////////////////////////////

#ifdef _DEBUG

static int utf8_unittest() {
  const struct Utf8Test {
    const char* encoded;
    size_t encsize, enclen;
    unsigned long decoded;
  } kTests[] = {
    { "a    ",             5, 1, 'a' },
    { "\x7F    ",          5, 1, 0x7F },
    { "\xC2\x80   ",       5, 2, 0x80 },
    { "\xDF\xBF   ",       5, 2, 0x7FF },
    { "\xE0\xA0\x80  ",    5, 3, 0x800 },
    { "\xEF\xBF\xBF  ",    5, 3, 0xFFFF },
    { "\xF0\x90\x80\x80 ", 5, 4, 0x10000 },
    { "\xF0\x90\x80\x80 ", 3, 0, 0x10000 },
    { "\xF0\xF0\x80\x80 ", 5, 0, 0 },
    { "\xF0\x90\x80  ",    5, 0, 0 },
    { "\x90\x80\x80  ",    5, 0, 0 },
    { NULL, 0, 0 },
  };
  for (size_t i=0; kTests[i].encoded; ++i) {
    unsigned long val = 0;
    ASSERT(kTests[i].enclen == utf8_decode(kTests[i].encoded,
                                           kTests[i].encsize,
                                           &val));
    unsigned long result = (kTests[i].enclen == 0) ? 0 : kTests[i].decoded;
    ASSERT(val == result);

    if (kTests[i].decoded == 0) {
      // Not an interesting encoding test case
      continue;
    }

    char buffer[5];
    memset(buffer, 0x01, ARRAY_SIZE(buffer));
    ASSERT(kTests[i].enclen == utf8_encode(buffer,
                                           kTests[i].encsize,
                                           kTests[i].decoded));
    ASSERT(memcmp(buffer, kTests[i].encoded, kTests[i].enclen) == 0);
    // Make sure remainder of buffer is unchanged
    ASSERT(memory_check(buffer + kTests[i].enclen,
                        0x1,
                        ARRAY_SIZE(buffer) - kTests[i].enclen));
  }
  return 1;
}

int test = utf8_unittest();

#endif  // _DEBUG

/////////////////////////////////////////////////////////////////////////////

}  // namespace utils_base