diff options
Diffstat (limited to 'src')
-rw-r--r-- | src/blocks/lzssblock.nim | 31 | ||||
-rw-r--r-- | src/huffman/huffmantree.nim | 10 | ||||
-rw-r--r-- | src/lzss/lzsschain.nim | 17 | ||||
-rw-r--r-- | src/lzss/lzssencoder.nim | 2 | ||||
-rw-r--r-- | src/lzsshuffman/lzsshuffmandecoder.nim | 34 | ||||
-rw-r--r-- | src/lzsshuffman/lzsshuffmanencoder.nim | 34 | ||||
-rw-r--r-- | src/lzsshuffman/lzsshuffmanstats.nim | 32 | ||||
-rw-r--r-- | src/lzsshuffman/lzsshuffmansymbol.nim | 34 |
8 files changed, 171 insertions, 23 deletions
diff --git a/src/blocks/lzssblock.nim b/src/blocks/lzssblock.nim index f68f665..b23cee2 100644 --- a/src/blocks/lzssblock.nim +++ b/src/blocks/lzssblock.nim | |||
@@ -14,19 +14,38 @@ | |||
14 | # You should have received a copy of the GNU Affero General Public License | 14 | # You should have received a copy of the GNU Affero General Public License |
15 | # along with this program. If not, see <https://www.gnu.org/licenses/>. | 15 | # along with this program. If not, see <https://www.gnu.org/licenses/>. |
16 | 16 | ||
17 | import ../bitio/bitreader, ../bitio/bitwriter | 17 | import lists |
18 | import ../bitio/integers, ../bitio/bitreader, ../bitio/bitwriter | ||
19 | import ../lzss/lzsschain, ../lzss/lzssencoder | ||
20 | import ../huffman/huffmantree, ../huffman/huffmantreebuilder, ../huffman/huffmanencoder, ../huffman/huffmandecoder | ||
21 | import ../lzsshuffman/lzsshuffmanstats, ../lzsshuffman/lzsshuffmandecoder, ../lzsshuffman/lzsshuffmanencoder | ||
22 | |||
23 | const maxDataByteLength = 32_000 | ||
18 | 24 | ||
19 | type LzssBlock* = object | 25 | type LzssBlock* = object |
20 | discard | 26 | lzssChain: LzssChain |
21 | 27 | ||
22 | proc readSerialised*(bitReader: BitReader): LzssBlock = | 28 | proc readSerialised*(bitReader: BitReader): LzssBlock = |
23 | discard | 29 | let symbolHuffmanTree = huffmantree.deserialise(bitReader, uint16) |
30 | let positionHuffmanTree = huffmantree.deserialise(bitReader, uint16) | ||
31 | let symbolDecoder = symbolHuffmanTree.decoder() | ||
32 | let positionDecoder = positionHuffmanTree.decoder() | ||
33 | LzssBlock(lzssChain: readChain(bitReader, symbolDecoder, positionDecoder, maxDataByteLength)) | ||
24 | 34 | ||
25 | proc writeSerialisedTo*(lzssBlock: LzssBlock, bitWriter: BitWriter) = | 35 | proc writeSerialisedTo*(lzssBlock: LzssBlock, bitWriter: BitWriter) = |
26 | discard | 36 | let (symbolStats, positionStats) = aggregateStats(lzssBlock.lzssChain) |
37 | let symbolHuffmanTree = buildHuffmanTree(symbolStats) | ||
38 | let positionHuffmanTree = buildHuffmanTree(positionStats) | ||
39 | let symbolEncoder = symbolHuffmanTree.encoder(uint16) | ||
40 | let positionEncoder = positionHuffmanTree.encoder(uint16) | ||
41 | symbolHuffmanTree.serialise(bitWriter) | ||
42 | positionHuffmanTree.serialise(bitWriter) | ||
43 | lzssBlock.lzssChain.writeChain(symbolEncoder, positionEncoder, bitWriter) | ||
27 | 44 | ||
28 | proc readRaw*(bitReader: BitReader): LzssBlock = | 45 | proc readRaw*(bitReader: BitReader): LzssBlock = |
29 | discard | 46 | let byteBuf = bitReader.readSeq(maxDataByteLength, uint8) |
47 | LzssBlock(lzssChain: lzssEncode(byteBuf.data)) | ||
30 | 48 | ||
31 | proc writeRawTo*(lzssBlock: LzssBlock, bitWriter: BitWriter) = | 49 | proc writeRawTo*(lzssBlock: LzssBlock, bitWriter: BitWriter) = |
32 | discard | 50 | let byteSeq = lzssBlock.lzssChain.decode() |
51 | bitWriter.writeSeq(byteSeq.len * wordBitLength, byteSeq) | ||
diff --git a/src/huffman/huffmantree.nim b/src/huffman/huffmantree.nim index 58a840e..f3fce1b 100644 --- a/src/huffman/huffmantree.nim +++ b/src/huffman/huffmantree.nim | |||
@@ -31,6 +31,11 @@ type HuffmanTreeNode*[T: SomeUnsignedInt] = ref object | |||
31 | of leaf: | 31 | of leaf: |
32 | value*: T | 32 | value*: T |
33 | 33 | ||
34 | proc maxValue*[T](node: HuffmanTreeNode[T]): T = | ||
35 | case node.kind: | ||
36 | of branch: node.maxChildValue | ||
37 | of leaf: node.value | ||
38 | |||
34 | proc huffmanBranch*[T](left, right: HuffmanTreeNode[T]): HuffmanTreeNode[T] = | 39 | proc huffmanBranch*[T](left, right: HuffmanTreeNode[T]): HuffmanTreeNode[T] = |
35 | HuffmanTreeNode[T]( | 40 | HuffmanTreeNode[T]( |
36 | kind: branch, left: left, right: right, | 41 | kind: branch, left: left, right: right, |
@@ -45,11 +50,6 @@ proc `==`*[T](a, b: HuffmanTreeNode[T]): bool = | |||
45 | of branch: a.left == b.left and a.right == b.right | 50 | of branch: a.left == b.left and a.right == b.right |
46 | of leaf: a.value == b.value | 51 | of leaf: a.value == b.value |
47 | 52 | ||
48 | proc maxValue*[T](node: HuffmanTreeNode[T]): T = | ||
49 | case node.kind: | ||
50 | of branch: node.maxChildValue | ||
51 | of leaf: node.value | ||
52 | |||
53 | proc deserialise*[T](bitReader: BitReader, valueType: typedesc[T]): HuffmanTreeNode[T] = | 53 | proc deserialise*[T](bitReader: BitReader, valueType: typedesc[T]): HuffmanTreeNode[T] = |
54 | let valueBitLength = bitReader.readBits(valueLengthFieldBitLength, uint8).int | 54 | let valueBitLength = bitReader.readBits(valueLengthFieldBitLength, uint8).int |
55 | proc readNode(): HuffmanTreeNode[T] = | 55 | proc readNode(): HuffmanTreeNode[T] = |
diff --git a/src/lzss/lzsschain.nim b/src/lzss/lzsschain.nim index 8b49914..8ebcb1a 100644 --- a/src/lzss/lzsschain.nim +++ b/src/lzss/lzsschain.nim | |||
@@ -15,7 +15,7 @@ | |||
15 | # along with this program. If not, see <https://www.gnu.org/licenses/>. | 15 | # along with this program. If not, see <https://www.gnu.org/licenses/>. |
16 | 16 | ||
17 | import lists, tables, sugar | 17 | import lists, tables, sugar |
18 | import ../bitio/integers, ../huffman/huffmantree | 18 | import ../bitio/integers |
19 | import listpolyfill, lzssnode | 19 | import listpolyfill, lzssnode |
20 | 20 | ||
21 | const maxChainByteLength = 32_000 * wordBitLength | 21 | const maxChainByteLength = 32_000 * wordBitLength |
@@ -26,6 +26,11 @@ type LzssChain* = | |||
26 | proc lzssChain*(): LzssChain = | 26 | proc lzssChain*(): LzssChain = |
27 | initSinglyLinkedList[LzssNode]() | 27 | initSinglyLinkedList[LzssNode]() |
28 | 28 | ||
29 | proc lzssChain*(chainArray: openArray[LzssNode]): LzssChain = | ||
30 | var chain = lzssChain() | ||
31 | for node in chainArray: chain.append(node) | ||
32 | chain | ||
33 | |||
29 | proc decode*(lzssChain: LzssChain): seq[uint8] = | 34 | proc decode*(lzssChain: LzssChain): seq[uint8] = |
30 | result = newSeqOfCap[uint8](maxChainByteLength) | 35 | result = newSeqOfCap[uint8](maxChainByteLength) |
31 | for node in lzssChain.items: | 36 | for node in lzssChain.items: |
@@ -35,13 +40,3 @@ proc decode*(lzssChain: LzssChain): seq[uint8] = | |||
35 | of reference: | 40 | of reference: |
36 | let absolutePos = result.len - node.relativePos | 41 | let absolutePos = result.len - node.relativePos |
37 | result.add(result.toOpenArray(absolutePos, absolutePos + node.length - 1)) | 42 | result.add(result.toOpenArray(absolutePos, absolutePos + node.length - 1)) |
38 | |||
39 | proc stats*(lzssChain: LzssChain): tuple[characters: CountTableRef[uint8], lengths, positions: CountTableRef[int]] = | ||
40 | result = (newCountTable[uint8](), newCountTable[int](), newCountTable[int]()) | ||
41 | for node in lzssChain.items: | ||
42 | case node.kind: | ||
43 | of character: | ||
44 | result.characters.inc(node.character) | ||
45 | of reference: | ||
46 | result.lengths.inc(node.length) | ||
47 | result.positions.inc(node.relativePos) | ||
diff --git a/src/lzss/lzssencoder.nim b/src/lzss/lzssencoder.nim index 8b750fb..82fbe7b 100644 --- a/src/lzss/lzssencoder.nim +++ b/src/lzss/lzssencoder.nim | |||
@@ -17,7 +17,7 @@ | |||
17 | import lists | 17 | import lists |
18 | import listpolyfill, matchtable, lzssnode, lzsschain | 18 | import listpolyfill, matchtable, lzssnode, lzsschain |
19 | 19 | ||
20 | const matchGroupLength = 3 | 20 | const matchGroupLength* = 3 |
21 | const maxRefByteLength = high(uint8).int + matchGroupLength | 21 | const maxRefByteLength = high(uint8).int + matchGroupLength |
22 | let emptySinglyLinkedList = initSinglyLinkedList[int]() | 22 | let emptySinglyLinkedList = initSinglyLinkedList[int]() |
23 | 23 | ||
diff --git a/src/lzsshuffman/lzsshuffmandecoder.nim b/src/lzsshuffman/lzsshuffmandecoder.nim new file mode 100644 index 0000000..cd71914 --- /dev/null +++ b/src/lzsshuffman/lzsshuffmandecoder.nim | |||
@@ -0,0 +1,34 @@ | |||
1 | # gzip-like LZSS compressor | ||
2 | # Copyright (C) 2018 Pacien TRAN-GIRARD | ||
3 | # | ||
4 | # This program is free software: you can redistribute it and/or modify | ||
5 | # it under the terms of the GNU Affero General Public License as | ||
6 | # published by the Free Software Foundation, either version 3 of the | ||
7 | # License, or (at your option) any later version. | ||
8 | # | ||
9 | # This program is distributed in the hope that it will be useful, | ||
10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | ||
12 | # GNU Affero General Public License for more details. | ||
13 | # | ||
14 | # You should have received a copy of the GNU Affero General Public License | ||
15 | # along with this program. If not, see <https://www.gnu.org/licenses/>. | ||
16 | |||
17 | import lists | ||
18 | import ../bitio/bitreader | ||
19 | import ../lzss/listpolyfill, ../lzss/lzssnode, ../lzss/lzsschain | ||
20 | import ../huffman/huffmantree, ../huffman/huffmandecoder | ||
21 | import lzsshuffmansymbol | ||
22 | |||
23 | proc readChain*(bitReader: BitReader, symbolDecoder, positionDecoder: HuffmanDecoder[uint16], maxDataByteLength: int): LzssChain = | ||
24 | var chain = lzssChain() | ||
25 | var (symbol, byteCursor) = (symbolDecoder.decode(bitReader).Symbol, 0) | ||
26 | while not symbol.isEndMarker(): | ||
27 | if byteCursor > maxDataByteLength: raise newException(IOError, "lzss block too long") | ||
28 | if symbol.isCharacter(): | ||
29 | chain.append(lzssCharacter(symbol.uint8)) | ||
30 | else: | ||
31 | let position = positionDecoder.decode(bitReader) | ||
32 | chain.append(unpackLzssReference(symbol, position)) | ||
33 | (symbol, byteCursor) = (symbolDecoder.decode(bitReader).Symbol, byteCursor + 1) | ||
34 | chain | ||
diff --git a/src/lzsshuffman/lzsshuffmanencoder.nim b/src/lzsshuffman/lzsshuffmanencoder.nim new file mode 100644 index 0000000..ea89f85 --- /dev/null +++ b/src/lzsshuffman/lzsshuffmanencoder.nim | |||
@@ -0,0 +1,34 @@ | |||
1 | # gzip-like LZSS compressor | ||
2 | # Copyright (C) 2018 Pacien TRAN-GIRARD | ||
3 | # | ||
4 | # This program is free software: you can redistribute it and/or modify | ||
5 | # it under the terms of the GNU Affero General Public License as | ||
6 | # published by the Free Software Foundation, either version 3 of the | ||
7 | # License, or (at your option) any later version. | ||
8 | # | ||
9 | # This program is distributed in the hope that it will be useful, | ||
10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | ||
12 | # GNU Affero General Public License for more details. | ||
13 | # | ||
14 | # You should have received a copy of the GNU Affero General Public License | ||
15 | # along with this program. If not, see <https://www.gnu.org/licenses/>. | ||
16 | |||
17 | import lists | ||
18 | import ../bitio/bitwriter | ||
19 | import ../lzss/listpolyfill, ../lzss/lzssnode, ../lzss/lzsschain, ../lzss/lzssencoder | ||
20 | import ../huffman/huffmantree, ../huffman/huffmantreebuilder, ../huffman/huffmanencoder | ||
21 | import lzsshuffmansymbol | ||
22 | |||
23 | proc writeSymbol(bitWriter: BitWriter, encodedSymbol: tuple[bitLength: int, value: uint16]) = | ||
24 | bitWriter.writeBits(encodedSymbol.bitLength, encodedSymbol.value) | ||
25 | |||
26 | proc writeChain*(lzssChain: LzssChain, symbolEncoder, positionEncoder: HuffmanEncoder[uint16, uint16], bitWriter: BitWriter) = | ||
27 | for node in lzssChain.items: | ||
28 | case node.kind: | ||
29 | of character: | ||
30 | bitWriter.writeSymbol(symbolEncoder.encode(node.character)) | ||
31 | of reference: | ||
32 | bitWriter.writeSymbol(symbolEncoder.encode(shiftLzssLength(node.length))) | ||
33 | bitWriter.writeSymbol(positionEncoder.encode(node.relativePos.uint16)) | ||
34 | bitWriter.writeSymbol(symbolEncoder.encode(endSymbol)) | ||
diff --git a/src/lzsshuffman/lzsshuffmanstats.nim b/src/lzsshuffman/lzsshuffmanstats.nim new file mode 100644 index 0000000..037ce5f --- /dev/null +++ b/src/lzsshuffman/lzsshuffmanstats.nim | |||
@@ -0,0 +1,32 @@ | |||
1 | # gzip-like LZSS compressor | ||
2 | # Copyright (C) 2018 Pacien TRAN-GIRARD | ||
3 | # | ||
4 | # This program is free software: you can redistribute it and/or modify | ||
5 | # it under the terms of the GNU Affero General Public License as | ||
6 | # published by the Free Software Foundation, either version 3 of the | ||
7 | # License, or (at your option) any later version. | ||
8 | # | ||
9 | # This program is distributed in the hope that it will be useful, | ||
10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | ||
12 | # GNU Affero General Public License for more details. | ||
13 | # | ||