// 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) { var bits = _input.GetBits(3); if (bits < 0) { return(false); } _codeLengthTreeCodeLength[_codeOrder[_loopCounter]] = (byte)bits; ++_loopCounter; } for (var 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; switch (_lengthCode) { case 16: { if (_loopCounter == 0) { throw new InvalidDataContractException(); } var previousCode = _codeList[_loopCounter - 1]; repeatCount = _input.GetBits(2) + 3; if (_loopCounter + repeatCount > _codeArraySize) { throw new InvalidDataContractException(); } for (var j = 0; j < repeatCount; j++) { _codeList[_loopCounter++] = previousCode; } } break; case 17: repeatCount = _input.GetBits(3) + 3; if (_loopCounter + repeatCount > _codeArraySize) { throw new InvalidDataContractException(); } for (var j = 0; j < repeatCount; j++) { _codeList[_loopCounter++] = 0; } break; default: repeatCount = _input.GetBits(7) + 11; if (_loopCounter + repeatCount > _codeArraySize) { throw new InvalidDataContractException(); } for (var j = 0; j < repeatCount; j++) { _codeList[_loopCounter++] = 0; } break; } } _state = InflaterState.ReadingTreeCodesBefore; // we want to read the next code. } break; default: throw new InvalidDataContractException("Unknown state."); } var literalTreeCodeLength = new byte[HuffmanTree.MaxLiteralTreeElements]; var 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 InvalidDataContractException(); } _literalLengthTree = new HuffmanTree(literalTreeCodeLength); _distanceTree = new HuffmanTree(distanceTreeCodeLength); _state = InflaterState.DecodeTop; return(true); }
private bool DecodeBlock(out bool endOfBlockCodeSeen) { endOfBlockCodeSeen = false; var 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 switch (_state) { case InflaterState.DecodeTop: // decode an element from the literal tree // var symbol = _literalLengthTree.GetNextSymbol(_input); if (symbol < 0) { return(false); } if (symbol < 256) { // literal _output.Write((byte)symbol); --freeBytes; } else if (symbol == 256) { // end of block endOfBlockCodeSeen = true; // 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 InvalidDataContractException(); } _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; var bits = _input.GetBits(_extraBits); if (bits < 0) { return(false); } if (_length < 0 || _length >= _lengthBase.Length) { throw new InvalidDataContractException(); } _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: throw new InvalidDataContractException("Unknown state."); } } return(true); }