convert_utf.c.svn-base

来自「SumatraPDF是一款小型开源的pdf阅读工具。虽然玲珑小巧（只有800多K」· SVN-BASE 代码 · 共 534 行 · 第 1/2 页

	    case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
	    case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
	    case 1: *--target =  (UTF8)(ch | firstByteMark[bytesToWrite]);
    }
    target += bytesToWrite;
  }
*sourceStart = source;
*targetStart = target;
return result;
}

/* --------------------------------------------------------------------- */

/*
 * Utility routine to tell whether a sequence of bytes is legal UTF-8.
 * This must be called with the length pre-determined by the first byte.
 * If not calling this from ConvertUTF8to*, then the length can be set by:
 *  length = trailingBytesForUTF8[*source]+1;
 * and the sequence is illegal right away if there aren't that many bytes
 * available.
 * If presented with a length > 4, this returns false.  The Unicode
 * definition of UTF-8 goes up to 4-byte sequences.
 */

static Boolean isLegalUTF8(const UTF8 *source, int length) {
  UTF8 a;
  const UTF8 *srcptr = source+length;
  switch (length) {
    default: return false;
      /* Everything else falls through when "true"... */
    case 4: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
    case 3: if ((a = (*--srcptr)) < 0x80 || a > 0xBF) return false;
    case 2: if ((a = (*--srcptr)) > 0xBF) return false;

      switch (*source) {
        /* no fall-through in this inner switch */
        case 0xE0: if (a < 0xA0) return false; break;
        case 0xED: if (a > 0x9F) return false; break;
        case 0xF0: if (a < 0x90) return false; break;
        case 0xF4: if (a > 0x8F) return false; break;
        default:   if (a < 0x80) return false;
      }

      case 1: if (*source >= 0x80 && *source < 0xC2) return false;
  }
  if (*source > 0xF4) return false;
  return true;
}

/* --------------------------------------------------------------------- */

/*
 * Exported function to return whether a UTF-8 sequence is legal or not.
 * This is not used here; it's just exported.
 */
Boolean isLegalUTF8Sequence(const UTF8 *source, const UTF8 *sourceEnd) {
  int length = trailingBytesForUTF8[*source]+1;
  if (source+length > sourceEnd) {
    return false;
  }
  return isLegalUTF8(source, length);
}

/* --------------------------------------------------------------------- */

ConversionResult ConvertUTF8toUTF16 (const UTF8** sourceStart, const UTF8* sourceEnd,
                                     UTF16** targetStart, UTF16* targetEnd, ConversionFlags flags) {
  ConversionResult result = conversionOK;
  const UTF8* source = *sourceStart;
  UTF16* target = *targetStart;
  while (source < sourceEnd) {
    UTF32 ch = 0;
    unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
    if (source + extraBytesToRead >= sourceEnd) {
	    result = sourceExhausted; break;
    }
    /* Do this check whether lenient or strict */
    if (! isLegalUTF8(source, extraBytesToRead+1)) {
	    result = sourceIllegal;
	    break;
    }
    /*
     * The cases all fall through. See "Note A" below.
     */
    switch (extraBytesToRead) {
	    case 5: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
	    case 4: ch += *source++; ch <<= 6; /* remember, illegal UTF-8 */
	    case 3: ch += *source++; ch <<= 6;
	    case 2: ch += *source++; ch <<= 6;
	    case 1: ch += *source++; ch <<= 6;
	    case 0: ch += *source++;
    }
    ch -= offsetsFromUTF8[extraBytesToRead];

    if (target >= targetEnd) {
	    source -= (extraBytesToRead+1); /* Back up source pointer! */
	    result = targetExhausted; break;
    }
    if (ch <= UNI_MAX_BMP) { /* Target is a character <= 0xFFFF */
	    /* UTF-16 surrogate values are illegal in UTF-32 */
	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
        if (flags == strictConversion) {
          source -= (extraBytesToRead+1); /* return to the illegal value itself */
          result = sourceIllegal;
          break;
        } else {
          *target++ = UNI_REPLACEMENT_CHAR;
        }
	    } else {
        *target++ = (UTF16)ch; /* normal case */
	    }
    } else if (ch > UNI_MAX_UTF16) {
	    if (flags == strictConversion) {
        result = sourceIllegal;
        source -= (extraBytesToRead+1); /* return to the start */
        break; /* Bail out; shouldn't continue */
	    } else {
        *target++ = UNI_REPLACEMENT_CHAR;
	    }
    } else {
	    /* target is a character in range 0xFFFF - 0x10FFFF. */
	    if (target + 1 >= targetEnd) {
        source -= (extraBytesToRead+1); /* Back up source pointer! */
        result = targetExhausted; break;
	    }
	    ch -= halfBase;
	    *target++ = (UTF16)((ch >> halfShift) + UNI_SUR_HIGH_START);
	    *target++ = (UTF16)((ch & halfMask) + UNI_SUR_LOW_START);
    }
  }
*sourceStart = source;
*targetStart = target;
return result;
}

/* --------------------------------------------------------------------- */

ConversionResult ConvertUTF32toUTF8 (const UTF32** sourceStart, const UTF32* sourceEnd,
                                     UTF8** targetStart, UTF8* targetEnd, ConversionFlags flags) {
  ConversionResult result = conversionOK;
  const UTF32* source = *sourceStart;
  UTF8* target = *targetStart;
  while (source < sourceEnd) {
    UTF32 ch;
    unsigned short bytesToWrite = 0;
    const UTF32 byteMask = 0xBF;
    const UTF32 byteMark = 0x80;
    ch = *source++;
    if (flags == strictConversion ) {
	    /* UTF-16 surrogate values are illegal in UTF-32 */
	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
        --source; /* return to the illegal value itself */
        result = sourceIllegal;
        break;
	    }
    }
    /*
     * Figure out how many bytes the result will require. Turn any
     * illegally large UTF32 things (> Plane 17) into replacement chars.
     */
    if (ch < (UTF32)0x80) {	     bytesToWrite = 1;
    } else if (ch < (UTF32)0x800) {     bytesToWrite = 2;
    } else if (ch < (UTF32)0x10000) {   bytesToWrite = 3;
    } else if (ch <= UNI_MAX_LEGAL_UTF32) {  bytesToWrite = 4;
    } else {			    bytesToWrite = 3;
      ch = UNI_REPLACEMENT_CHAR;
      result = sourceIllegal;
    }

    target += bytesToWrite;
    if (target > targetEnd) {
	    --source; /* Back up source pointer! */
	    target -= bytesToWrite; result = targetExhausted; break;
    }
    switch (bytesToWrite) { /* note: everything falls through. */
	    case 4: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
	    case 3: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
	    case 2: *--target = (UTF8)((ch | byteMark) & byteMask); ch >>= 6;
	    case 1: *--target = (UTF8) (ch | firstByteMark[bytesToWrite]);
    }
    target += bytesToWrite;
  }
*sourceStart = source;
*targetStart = target;
return result;
}

/* --------------------------------------------------------------------- */

ConversionResult ConvertUTF8toUTF32 (const UTF8** sourceStart, const UTF8* sourceEnd,
                                     UTF32** targetStart, UTF32* targetEnd, ConversionFlags flags) {
  ConversionResult result = conversionOK;
  const UTF8* source = *sourceStart;
  UTF32* target = *targetStart;
  while (source < sourceEnd) {
    UTF32 ch = 0;
    unsigned short extraBytesToRead = trailingBytesForUTF8[*source];
    if (source + extraBytesToRead >= sourceEnd) {
	    result = sourceExhausted; break;
    }
    /* Do this check whether lenient or strict */
    if (! isLegalUTF8(source, extraBytesToRead+1)) {
	    result = sourceIllegal;
	    break;
    }
    /*
     * The cases all fall through. See "Note A" below.
     */
    switch (extraBytesToRead) {
	    case 5: ch += *source++; ch <<= 6;
	    case 4: ch += *source++; ch <<= 6;
	    case 3: ch += *source++; ch <<= 6;
	    case 2: ch += *source++; ch <<= 6;
	    case 1: ch += *source++; ch <<= 6;
	    case 0: ch += *source++;
    }
    ch -= offsetsFromUTF8[extraBytesToRead];

    if (target >= targetEnd) {
	    source -= (extraBytesToRead+1); /* Back up the source pointer! */
	    result = targetExhausted; break;
    }
    if (ch <= UNI_MAX_LEGAL_UTF32) {
	    /*
	     * UTF-16 surrogate values are illegal in UTF-32, and anything
	     * over Plane 17 (> 0x10FFFF) is illegal.
	     */
	    if (ch >= UNI_SUR_HIGH_START && ch <= UNI_SUR_LOW_END) {
        if (flags == strictConversion) {
          source -= (extraBytesToRead+1); /* return to the illegal value itself */
          result = sourceIllegal;
          break;
        } else {
          *target++ = UNI_REPLACEMENT_CHAR;
        }
	    } else {
        *target++ = ch;
	    }
    } else { /* i.e., ch > UNI_MAX_LEGAL_UTF32 */
	    result = sourceIllegal;
	    *target++ = UNI_REPLACEMENT_CHAR;
    }
  }
  *sourceStart = source;
  *targetStart = target;
  return result;
}

/* ---------------------------------------------------------------------

Note A.
The fall-through switches in UTF-8 reading code save a
temp variable, some decrements & conditionals.  The switches
are equivalent to the following loop:
{
  int tmpBytesToRead = extraBytesToRead+1;
  do {
		ch += *source++;
		--tmpBytesToRead;
		if (tmpBytesToRead) ch <<= 6;
  } while (tmpBytesToRead > 0);
}
In UTF-8 writing code, the switches on "bytesToWrite" are
similarly unrolled loops.

--------------------------------------------------------------------- */