// Format of the dynamic block header:
// 5 Bits: HLIT, # of Literal/Length codes - 257 (257 - 286)
// 5 Bits: HDIST, # of Distance codes - 1 (1 - 32)
// 4 Bits: HCLEN, # of Code Length codes - 4 (4 - 19)
//
// (HCLEN + 4) x 3 bits: code lengths for the code length
// alphabet given just above, in the order: 16, 17, 18,
// 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15
//
// These code lengths are interpreted as 3-bit integers
// (0-7); as above, a code length of 0 means the
// corresponding symbol (literal/length or distance code
// length) is not used.
//
// HLIT + 257 code lengths for the literal/length alphabet,
// encoded using the code length Huffman code
//
// HDIST + 1 code lengths for the distance alphabet,
// encoded using the code length Huffman code
//
// The code length repeat codes can cross from HLIT + 257 to the
// HDIST + 1 code lengths. In other words, all code lengths form
// a single sequence of HLIT + HDIST + 258 values.
private bool DecodeDynamicBlockHeader()
{
switch (state)
{
case InflaterState.ReadingNumLitCodes:
literalLengthCodeCount = input.GetBits(5);
if (literalLengthCodeCount < 0)
{
return(false);
}
literalLengthCodeCount += 257;
state = InflaterState.ReadingNumDistCodes;
goto case InflaterState.ReadingNumDistCodes;
case InflaterState.ReadingNumDistCodes:
distanceCodeCount = input.GetBits(5);
if (distanceCodeCount < 0)
{
return(false);
}
distanceCodeCount += 1;
state = InflaterState.ReadingNumCodeLengthCodes;
goto case InflaterState.ReadingNumCodeLengthCodes;
case InflaterState.ReadingNumCodeLengthCodes:
codeLengthCodeCount = input.GetBits(4);
if (codeLengthCodeCount < 0)
{
return(false);
}
codeLengthCodeCount += 4;
loopCounter = 0;
state = InflaterState.ReadingCodeLengthCodes;
goto case InflaterState.ReadingCodeLengthCodes;
case InflaterState.ReadingCodeLengthCodes:
while (loopCounter < codeLengthCodeCount)
{
int bits = input.GetBits(3);
if (bits < 0)
{
return(false);
}
codeLengthTreeCodeLength[codeOrder[loopCounter]] = (byte)bits;
++loopCounter;
}
for (int i = codeLengthCodeCount; i < codeOrder.Length; i++)
{
codeLengthTreeCodeLength[codeOrder[i]] = 0;
}
// create huffman tree for code length
codeLengthTree = new HuffmanTree(codeLengthTreeCodeLength);
codeArraySize = literalLengthCodeCount + distanceCodeCount;
loopCounter = 0; // reset loop count
state = InflaterState.ReadingTreeCodesBefore;
goto case InflaterState.ReadingTreeCodesBefore;
case InflaterState.ReadingTreeCodesBefore:
case InflaterState.ReadingTreeCodesAfter:
while (loopCounter < codeArraySize)
{
if (state == InflaterState.ReadingTreeCodesBefore)
{
if ((lengthCode = codeLengthTree.GetNextSymbol(input)) < 0)
{
return(false);
}
}
// The alphabet for code lengths is as follows:
// 0 - 15: Represent code lengths of 0 - 15
// 16: Copy the previous code length 3 - 6 times.
// The next 2 bits indicate repeat length
// (0 = 3, ... , 3 = 6)
// Example: Codes 8, 16 (+2 bits 11),
// 16 (+2 bits 10) will expand to
// 12 code lengths of 8 (1 + 6 + 5)
// 17: Repeat a code length of 0 for 3 - 10 times.
// (3 bits of length)
// 18: Repeat a code length of 0 for 11 - 138 times
// (7 bits of length)
if (lengthCode <= 15)
{
codeList[loopCounter++] = (byte)lengthCode;
}
else
{
if (!input.EnsureBitsAvailable(7))
{ // it doesn't matter if we require more bits here
state = InflaterState.ReadingTreeCodesAfter;
return(false);
}
int repeatCount;
if (lengthCode == 16)
{
if (loopCounter == 0)
{ // can't have "prev code" on first code
throw new InvalidDataException();
}
byte previousCode = codeList[loopCounter - 1];
repeatCount = input.GetBits(2) + 3;
if (loopCounter + repeatCount > codeArraySize)
{
throw new InvalidDataException();
}
for (int j = 0; j < repeatCount; j++)
{
codeList[loopCounter++] = previousCode;
}
}
else if (lengthCode == 17)
{
repeatCount = input.GetBits(3) + 3;
if (loopCounter + repeatCount > codeArraySize)
{
throw new InvalidDataException();
}
for (int j = 0; j < repeatCount; j++)
{
codeList[loopCounter++] = 0;
}
}
else
{ // code == 18
repeatCount = input.GetBits(7) + 11;
if (loopCounter + repeatCount > codeArraySize)
{
throw new InvalidDataException();
}
for (int j = 0; j < repeatCount; j++)
{
codeList[loopCounter++] = 0;
}
}
}
state = InflaterState.ReadingTreeCodesBefore; // we want to read the next code.
}
break;
default:
Debug.Assert(false, "check why we are here!");
throw new InvalidDataException("UnknownState");
}
byte[] literalTreeCodeLength = new byte[HuffmanTree.MaxLiteralTreeElements];
byte[] distanceTreeCodeLength = new byte[HuffmanTree.MaxDistTreeElements];
// Create literal and distance tables
Array.Copy(codeList, literalTreeCodeLength, literalLengthCodeCount);
Array.Copy(codeList, literalLengthCodeCount, distanceTreeCodeLength, 0, distanceCodeCount);
// Make sure there is an end-of-block code, otherwise how could we ever end?
if (literalTreeCodeLength[HuffmanTree.EndOfBlockCode] == 0)
{
throw new InvalidDataException();
}
literalLengthTree = new HuffmanTree(literalTreeCodeLength);
distanceTree = new HuffmanTree(distanceTreeCodeLength);
state = InflaterState.DecodeTop;
return(true);
}
private bool DecodeBlock(out bool end_of_block_code_seen)
{
end_of_block_code_seen = false;
int freeBytes = output.FreeBytes; // it is a little bit faster than frequently accessing the property
while (freeBytes > 258)
{
// 258 means we can safely do decoding since maximum repeat length is 258
int symbol;
switch (state)
{
case InflaterState.DecodeTop:
// decode an element from the literal tree
// TODO: optimize this!!!
symbol = literalLengthTree.GetNextSymbol(input);
if (symbol < 0)
{ // running out of input
return(false);
}
if (symbol < 256)
{ // literal
output.Write((byte)symbol);
--freeBytes;
}
else if (symbol == 256)
{ // end of block
end_of_block_code_seen = true;
Debug.WriteLineIf(CompressionTracingSwitch.Informational, "End of Block", "Compression");
// Reset state
state = InflaterState.ReadingBFinal;
return(true); // ***********
}
else
{ // length/distance pair
symbol -= 257; // length code started at 257
if (symbol < 8)
{
symbol += 3; // match length = 3,4,5,6,7,8,9,10
extraBits = 0;
}
else if (symbol == 28)
{ // extra bits for code 285 is 0
symbol = 258; // code 285 means length 258
extraBits = 0;
}
else
{
if (symbol < 0 || symbol >= extraLengthBits.Length)
{
throw new InvalidDataException("GenericInvalidData");
}
extraBits = extraLengthBits[symbol];
Debug.Assert(extraBits != 0, "We handle other cases seperately!");
}
length = symbol;
goto case InflaterState.HaveInitialLength;
}
break;
case InflaterState.HaveInitialLength:
if (extraBits > 0)
{
state = InflaterState.HaveInitialLength;
int bits = input.GetBits(extraBits);
if (bits < 0)
{
return(false);
}
if (length < 0 || length >= lengthBase.Length)
{
throw new InvalidDataException("GenericInvalidData");
}
length = lengthBase[length] + bits;
}
state = InflaterState.HaveFullLength;
goto case InflaterState.HaveFullLength;
case InflaterState.HaveFullLength:
if (blockType == BlockType.Dynamic)
{
distanceCode = distanceTree.GetNextSymbol(input);
}
else
{ // get distance code directly for static block
distanceCode = input.GetBits(5);
if (distanceCode >= 0)
{
distanceCode = staticDistanceTreeTable[distanceCode];
}
}
if (distanceCode < 0)
{ // running out input
return(false);
}
state = InflaterState.HaveDistCode;
goto case InflaterState.HaveDistCode;
case InflaterState.HaveDistCode:
// To avoid a table lookup we note that for distanceCode >= 2,
// extra_bits = (distanceCode-2) >> 1
int offset;
if (distanceCode > 3)
{
extraBits = (distanceCode - 2) >> 1;
int bits = input.GetBits(extraBits);
if (bits < 0)
{
return(false);
}
offset = distanceBasePosition[distanceCode] + bits;
}
else
{
offset = distanceCode + 1;
}
Debug.Assert(freeBytes >= 258, "following operation is not safe!");
output.WriteLengthDistance(length, offset);
freeBytes -= length;
state = InflaterState.DecodeTop;
break;
default:
Debug.Assert(false, "check why we are here!");
throw new InvalidDataException("UnknownState");
}
}
return(true);
}