// Internal version of "GetByteCount" which can handle a rolling
// state between multiple calls to this method.
- private static int InternalGetByteCount (char[] chars, int index, int count, ref uint leftOver, bool flush)
+ private static int InternalGetByteCount (char[] chars, int index, int count, uint leftOver, bool flush)
{
// Validate the parameters.
if (chars == null) {
throw new ArgumentOutOfRangeException ("count", _("ArgRange_Array"));
}
- if (index == chars.Length) {
- if (flush && leftOver != 0) {
- // Flush the left-over surrogate pair start.
- leftOver = 0;
- return 3;
- }
- return 0;
- }
-
- unsafe {
- fixed (char* cptr = chars) {
- return InternalGetByteCount (cptr + index, count, ref leftOver, flush);
- }
- }
- }
-
-
- private unsafe static int InternalGetByteCount (char* chars, int count, ref uint leftOver, bool flush)
- {
- int index = 0;
-
// Determine the lengths of all characters.
char ch;
int length = 0;
length += 3;
}
- leftOver = pair;
-
// Return the final length to the caller.
return length;
}
// Get the number of bytes needed to encode a character buffer.
public override int GetByteCount (char[] chars, int index, int count)
{
- uint dummy = 0;
- return InternalGetByteCount (chars, index, count, ref dummy, true);
+ return InternalGetByteCount (chars, index, count, 0, true);
}
// Convenience wrappers for "GetByteCount".
throw new ArgumentOutOfRangeException ("byteIndex", _("ArgRange_Array"));
}
- if (charIndex == chars.Length) {
- if (flush && leftOver != 0) {
- // Flush the left-over surrogate pair start.
- bytes [byteIndex++] = 0xEF;
- bytes [byteIndex++] = 0xBB;
- bytes [byteIndex++] = 0xBF;
- leftOver = 0;
- return 3;
- }
- return 0;
- }
-
- unsafe {
- fixed (char* cptr = chars) {
- fixed (byte *bptr = bytes) {
- return InternalGetBytes (
- cptr + charIndex, charCount,
- bptr + byteIndex, bytes.Length - byteIndex,
- ref leftOver, flush);
- }
- }
- }
- }
-
- private unsafe static int InternalGetBytes (char* chars, int charCount,
- byte* bytes, int byteCount,
- ref uint leftOver, bool flush)
- {
- int charIndex = 0;
- int byteIndex = 0;
-
// Convert the characters into bytes.
char ch;
- int length = byteCount;
+ int length = bytes.Length;
uint pair;
uint left = leftOver;
int posn = byteIndex;
// state between multiple calls to this method.
private static int InternalGetCharCount (
byte[] bytes, int index, int count, uint leftOverBits,
- uint leftOverCount, object provider,
- ref DecoderFallbackBuffer fallbackBuffer, bool flush)
+ uint leftOverCount, DecoderFallbackBuffer fallbackBuffer, bool flush)
#else
private static int InternalGetCharCount (
byte[] bytes, int index, int count, uint leftOverBits,
throw new ArgumentOutOfRangeException ("count", _("ArgRange_Array"));
}
- int length = 0;
-
- if (leftOverCount == 0) {
- int end = index + count;
- for (; index < end; index++, count--) {
- if (bytes [index] < 0x80)
- length++;
- else
- break;
- }
- }
-
// Determine the number of characters that we have.
uint ch;
+ int length = 0;
uint leftBits = leftOverBits;
uint leftSoFar = (leftOverCount & (uint)0x0F);
uint leftSize = ((leftOverCount >> 4) & (uint)0x0F);
} else {
// Invalid UTF-8 start character.
#if NET_2_0
- length += Fallback (provider, ref fallbackBuffer, bytes, index - 1);
+ length += Fallback (fallbackBuffer, bytes, index - 1);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
}
if (overlong) {
#if NET_2_0
- length += Fallback (provider, ref fallbackBuffer, bytes, index - 1);
+ length += Fallback (fallbackBuffer, bytes, index - 1);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Overlong"), leftBits.ToString ());
length += 2;
} else {
#if NET_2_0
- length += Fallback (provider, ref fallbackBuffer, bytes, index - 1);
+ length += Fallback (fallbackBuffer, bytes, index - 1);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
} else {
// Invalid UTF-8 sequence: clear and restart.
#if NET_2_0
- length += Fallback (provider, ref fallbackBuffer, bytes, index - 1);
+ length += Fallback (fallbackBuffer, bytes, index - 1);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
// We had left-over bytes that didn't make up
// a complete UTF-8 character sequence.
#if NET_2_0
- length += Fallback (provider, ref fallbackBuffer, bytes, index - 1);
+ length += Fallback (fallbackBuffer, bytes, index - 1);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
#if NET_2_0
// for GetCharCount()
- static int Fallback (object provider, ref DecoderFallbackBuffer buffer, byte [] bytes, int index)
+ static int Fallback (DecoderFallbackBuffer buffer, byte [] bytes, int index)
{
- if (buffer == null) {
- DecoderFallback fb = provider as DecoderFallback;
- if (fb != null)
- buffer = fb.CreateFallbackBuffer ();
- else
- buffer = ((Decoder) provider).FallbackBuffer;
- }
buffer.Fallback (bytes, index - 1);
return buffer.Remaining;
}
// for GetChars()
- static void Fallback (object provider, ref DecoderFallbackBuffer buffer, byte [] bytes, int byteIndex,
+ static void Fallback (DecoderFallbackBuffer buffer, byte [] bytes, int byteIndex,
char [] chars, ref int charIndex)
{
- if (buffer == null) {
- DecoderFallback fb = provider as DecoderFallback;
- if (fb != null)
- buffer = fb.CreateFallbackBuffer ();
- else
- buffer = ((Decoder) provider).FallbackBuffer;
- }
buffer.Fallback (bytes, byteIndex - 1);
while (buffer.Remaining > 0)
chars [charIndex++] = buffer.GetNextChar ();
public override int GetCharCount (byte[] bytes, int index, int count)
{
#if NET_2_0
- DecoderFallbackBuffer buf = null;
- return InternalGetCharCount (bytes, index, count, 0, 0, DecoderFallback, ref buf, true);
+ return InternalGetCharCount (bytes, index, count, 0, 0, DecoderFallback.CreateFallbackBuffer (), true);
#else
return InternalGetCharCount (bytes, index, count, 0, 0, throwOnInvalid, true);
#endif
private static int InternalGetChars (
byte[] bytes, int byteIndex, int byteCount, char[] chars,
int charIndex, ref uint leftOverBits, ref uint leftOverCount,
- object provider,
- ref DecoderFallbackBuffer fallbackBuffer, bool flush)
+ DecoderFallbackBuffer fallbackBuffer, bool flush)
#else
private static int InternalGetChars (
byte[] bytes, int byteIndex, int byteCount, char[] chars,
if (charIndex == chars.Length)
return 0;
- int posn = charIndex;
-
- if (leftOverCount == 0) {
- int end = byteIndex + byteCount;
- for (; byteIndex < end; posn++, byteIndex++, byteCount--) {
- if (bytes [byteIndex] < 0x80)
- chars [posn] = (char) bytes [byteIndex];
- else
- break;
- }
- }
-
// Convert the bytes into the output buffer.
uint ch;
int length = chars.Length;
+ int posn = charIndex;
uint leftBits = leftOverBits;
uint leftSoFar = (leftOverCount & (uint)0x0F);
uint leftSize = ((leftOverCount >> 4) & (uint)0x0F);
-
- int byteEnd = byteIndex + byteCount;
- if (byteEnd < 0 || byteEnd > bytes.Length)
- throw new SystemException (String.Format ("INTERNAL ERROR: should not happen: {0} {1} {2}", byteIndex, byteCount, byteEnd));
-
- for(; byteIndex < byteEnd; byteIndex++) {
+ while (byteCount > 0) {
// Fetch the next character from the byte buffer.
- ch = (uint)(bytes[byteIndex]);
+ ch = (uint)(bytes[byteIndex++]);
+ --byteCount;
if (leftSize == 0) {
// Process a UTF-8 start character.
if (ch < (uint)0x0080) {
} else {
// Invalid UTF-8 start character.
#if NET_2_0
- Fallback (provider, ref fallbackBuffer, bytes, byteIndex, chars, ref posn);
+ Fallback (fallbackBuffer, bytes, byteIndex, chars, ref posn);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
}
if (overlong) {
#if NET_2_0
- Fallback (provider, ref fallbackBuffer, bytes, byteIndex, chars, ref posn);
+ Fallback (fallbackBuffer, bytes, byteIndex, chars, ref posn);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Overlong"), leftBits.ToString ());
-#endif
- }
- else if ((leftBits & 0xF800) == 0xD800) {
- // UTF-8 doesn't use surrogate characters
-#if NET_2_0
- Fallback (provider, ref fallbackBuffer, bytes, byteIndex, chars, ref posn);
-#else
- if (throwOnInvalid)
- throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
#endif
}
else {
(char)((leftBits & (uint)0x3FF) + (uint)0xDC00);
} else {
#if NET_2_0
- Fallback (provider, ref fallbackBuffer, bytes, byteIndex, chars, ref posn);
+ Fallback (fallbackBuffer, bytes, byteIndex, chars, ref posn);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
} else {
// Invalid UTF-8 sequence: clear and restart.
#if NET_2_0
- Fallback (provider, ref fallbackBuffer, bytes, byteIndex, chars, ref posn);
+ Fallback (fallbackBuffer, bytes, byteIndex, chars, ref posn);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
#endif
leftSize = 0;
--byteIndex;
+ ++byteCount;
}
}
}
// We had left-over bytes that didn't make up
// a complete UTF-8 character sequence.
#if NET_2_0
- Fallback (provider, ref fallbackBuffer, bytes, byteIndex, chars, ref posn);
+ Fallback (fallbackBuffer, bytes, byteIndex, chars, ref posn);
#else
if (throwOnInvalid)
throw new ArgumentException (_("Arg_InvalidUTF8"), "bytes");
uint leftOverBits = 0;
uint leftOverCount = 0;
#if NET_2_0
- DecoderFallbackBuffer buf = null;
return InternalGetChars (bytes, byteIndex, byteCount, chars,
- charIndex, ref leftOverBits, ref leftOverCount, DecoderFallback, ref buf, true);
+ charIndex, ref leftOverBits, ref leftOverCount, DecoderFallback.CreateFallbackBuffer (), true);
#else
return InternalGetChars (bytes, byteIndex, byteCount, chars,
charIndex, ref leftOverBits, ref leftOverCount, throwOnInvalid, true);
public override int GetCharCount (byte[] bytes, int index, int count)
{
#if NET_2_0
- DecoderFallbackBuffer buf = null;
return InternalGetCharCount (bytes, index, count,
- leftOverBits, leftOverCount, this, ref buf, false);
+ leftOverBits, leftOverCount, FallbackBuffer, false);
#else
return InternalGetCharCount (bytes, index, count,
leftOverBits, leftOverCount, throwOnInvalid, false);
int byteCount, char[] chars, int charIndex)
{
#if NET_2_0
- DecoderFallbackBuffer buf = null;
return InternalGetChars (bytes, byteIndex, byteCount,
- chars, charIndex, ref leftOverBits, ref leftOverCount, this, ref buf, false);
+ chars, charIndex, ref leftOverBits, ref leftOverCount, FallbackBuffer, false);
#else
return InternalGetChars (bytes, byteIndex, byteCount,
chars, charIndex, ref leftOverBits, ref leftOverCount, throwOnInvalid, false);
private class UTF8Encoder : Encoder
{
private bool emitIdentifier;
- private uint leftOverForCount;
- private uint leftOverForConv;
+ private uint leftOver;
// Constructor.
public UTF8Encoder (bool emitIdentifier)
{
this.emitIdentifier = emitIdentifier;
- leftOverForCount = 0;
- leftOverForConv = 0;
+ leftOver = 0;
}
// Override inherited methods.
public override int GetByteCount (char[] chars, int index,
int count, bool flush)
{
- return InternalGetByteCount (chars, index, count, ref leftOverForCount, flush);
+ return InternalGetByteCount (chars, index, count, leftOver, flush);
}
public override int GetBytes (char[] chars, int charIndex,
- int charCount, byte[] bytes, int byteIndex, bool flush)
- {
- int result;
- result = InternalGetBytes (chars, charIndex, charCount, bytes, byteIndex, ref leftOverForConv, flush);
- emitIdentifier = false;
- return result;
- }
-
-#if NET_2_0
- public unsafe override int GetByteCount (char* chars, int count, bool flush)
- {
- return InternalGetByteCount (chars, count, ref leftOverForCount, flush);
- }
-
- public unsafe override int GetBytes (char* chars, int charCount,
- byte* bytes, int byteCount, bool flush)
+ int charCount, byte[] bytes, int byteCount, bool flush)
{
int result;
- result = InternalGetBytes (chars, charCount, bytes, byteCount, ref leftOverForConv, flush);
+ result = InternalGetBytes (chars, charIndex, charCount, bytes, byteCount, ref leftOver, flush);
emitIdentifier = false;
return result;
}
-#endif
} // class UTF8Encoder