import { Bzip2Error, BZIP2_ERROR_CODES } from '../errors.js'; import type { IBitReader, IHuffmanGroup } from '../interfaces.js'; // Re-export Bzip2Error for backward compatibility export { Bzip2Error }; /** * Throw a BZIP2 error with proper error code */ function throwError(message: string, code: string = BZIP2_ERROR_CODES.INVALID_BLOCK_DATA): never { throw new Bzip2Error(message, code); } /** * BZIP2 decompression implementation */ export class Bzip2 { // CRC32 lookup table for BZIP2 public readonly crcTable: readonly number[] = [ 0x00000000, 0x04c11db7, 0x09823b6e, 0x0d4326d9, 0x130476dc, 0x17c56b6b, 0x1a864db2, 0x1e475005, 0x2608edb8, 0x22c9f00f, 0x2f8ad6d6, 0x2b4bcb61, 0x350c9b64, 0x31cd86d3, 0x3c8ea00a, 0x384fbdbd, 0x4c11db70, 0x48d0c6c7, 0x4593e01e, 0x4152fda9, 0x5f15adac, 0x5bd4b01b, 0x569796c2, 0x52568b75, 0x6a1936c8, 0x6ed82b7f, 0x639b0da6, 0x675a1011, 0x791d4014, 0x7ddc5da3, 0x709f7b7a, 0x745e66cd, 0x9823b6e0, 0x9ce2ab57, 0x91a18d8e, 0x95609039, 0x8b27c03c, 0x8fe6dd8b, 0x82a5fb52, 0x8664e6e5, 0xbe2b5b58, 0xbaea46ef, 0xb7a96036, 0xb3687d81, 0xad2f2d84, 0xa9ee3033, 0xa4ad16ea, 0xa06c0b5d, 0xd4326d90, 0xd0f37027, 0xddb056fe, 0xd9714b49, 0xc7361b4c, 0xc3f706fb, 0xceb42022, 0xca753d95, 0xf23a8028, 0xf6fb9d9f, 0xfbb8bb46, 0xff79a6f1, 0xe13ef6f4, 0xe5ffeb43, 0xe8bccd9a, 0xec7dd02d, 0x34867077, 0x30476dc0, 0x3d044b19, 0x39c556ae, 0x278206ab, 0x23431b1c, 0x2e003dc5, 0x2ac12072, 0x128e9dcf, 0x164f8078, 0x1b0ca6a1, 0x1fcdbb16, 0x018aeb13, 0x054bf6a4, 0x0808d07d, 0x0cc9cdca, 0x7897ab07, 0x7c56b6b0, 0x71159069, 0x75d48dde, 0x6b93dddb, 0x6f52c06c, 0x6211e6b5, 0x66d0fb02, 0x5e9f46bf, 0x5a5e5b08, 0x571d7dd1, 0x53dc6066, 0x4d9b3063, 0x495a2dd4, 0x44190b0d, 0x40d816ba, 0xaca5c697, 0xa864db20, 0xa527fdf9, 0xa1e6e04e, 0xbfa1b04b, 0xbb60adfc, 0xb6238b25, 0xb2e29692, 0x8aad2b2f, 0x8e6c3698, 0x832f1041, 0x87ee0df6, 0x99a95df3, 0x9d684044, 0x902b669d, 0x94ea7b2a, 0xe0b41de7, 0xe4750050, 0xe9362689, 0xedf73b3e, 0xf3b06b3b, 0xf771768c, 0xfa325055, 0xfef34de2, 0xc6bcf05f, 0xc27dede8, 0xcf3ecb31, 0xcbffd686, 0xd5b88683, 0xd1799b34, 0xdc3abded, 0xd8fba05a, 0x690ce0ee, 0x6dcdfd59, 0x608edb80, 0x644fc637, 0x7a089632, 0x7ec98b85, 0x738aad5c, 0x774bb0eb, 0x4f040d56, 0x4bc510e1, 0x46863638, 0x42472b8f, 0x5c007b8a, 0x58c1663d, 0x558240e4, 0x51435d53, 0x251d3b9e, 0x21dc2629, 0x2c9f00f0, 0x285e1d47, 0x36194d42, 0x32d850f5, 0x3f9b762c, 0x3b5a6b9b, 0x0315d626, 0x07d4cb91, 0x0a97ed48, 0x0e56f0ff, 0x1011a0fa, 0x14d0bd4d, 0x19939b94, 0x1d528623, 0xf12f560e, 0xf5ee4bb9, 0xf8ad6d60, 0xfc6c70d7, 0xe22b20d2, 0xe6ea3d65, 0xeba91bbc, 0xef68060b, 0xd727bbb6, 0xd3e6a601, 0xdea580d8, 0xda649d6f, 0xc423cd6a, 0xc0e2d0dd, 0xcda1f604, 0xc960ebb3, 0xbd3e8d7e, 0xb9ff90c9, 0xb4bcb610, 0xb07daba7, 0xae3afba2, 0xaafbe615, 0xa7b8c0cc, 0xa379dd7b, 0x9b3660c6, 0x9ff77d71, 0x92b45ba8, 0x9675461f, 0x8832161a, 0x8cf30bad, 0x81b02d74, 0x857130c3, 0x5d8a9099, 0x594b8d2e, 0x5408abf7, 0x50c9b640, 0x4e8ee645, 0x4a4ffbf2, 0x470cdd2b, 0x43cdc09c, 0x7b827d21, 0x7f436096, 0x7200464f, 0x76c15bf8, 0x68860bfd, 0x6c47164a, 0x61043093, 0x65c52d24, 0x119b4be9, 0x155a565e, 0x18197087, 0x1cd86d30, 0x029f3d35, 0x065e2082, 0x0b1d065b, 0x0fdc1bec, 0x3793a651, 0x3352bbe6, 0x3e119d3f, 0x3ad08088, 0x2497d08d, 0x2056cd3a, 0x2d15ebe3, 0x29d4f654, 0xc5a92679, 0xc1683bce, 0xcc2b1d17, 0xc8ea00a0, 0xd6ad50a5, 0xd26c4d12, 0xdf2f6bcb, 0xdbee767c, 0xe3a1cbc1, 0xe760d676, 0xea23f0af, 0xeee2ed18, 0xf0a5bd1d, 0xf464a0aa, 0xf9278673, 0xfde69bc4, 0x89b8fd09, 0x8d79e0be, 0x803ac667, 0x84fbdbd0, 0x9abc8bd5, 0x9e7d9662, 0x933eb0bb, 0x97ffad0c, 0xafb010b1, 0xab710d06, 0xa6322bdf, 0xa2f33668, 0xbcb4666d, 0xb8757bda, 0xb5365d03, 0xb1f740b4, ]; // State arrays initialized in header() private byteCount!: Int32Array; private symToByte!: Uint8Array; private mtfSymbol!: Int32Array; private selectors!: Uint8Array; /** * Create a bit reader from a byte array */ array(bytes: Uint8Array | Buffer): (n: number) => number { let bit = 0; let byte = 0; const BITMASK = [0, 0x01, 0x03, 0x07, 0x0f, 0x1f, 0x3f, 0x7f, 0xff]; return function (n: number): number { let result = 0; while (n > 0) { const left = 8 - bit; if (n >= left) { result <<= left; result |= BITMASK[left] & bytes[byte++]; bit = 0; n -= left; } else { result <<= n; result |= (bytes[byte] & (BITMASK[n] << (8 - n - bit))) >> (8 - n - bit); bit += n; n = 0; } } return result; }; } /** * Simple decompression from a buffer */ simple(srcbuffer: Uint8Array | Buffer, stream: (byte: number) => void): void { const bits = this.array(srcbuffer); const size = this.header(bits as IBitReader); let ret: number | null = 0; const bufsize = 100000 * size; const buf = new Int32Array(bufsize); do { ret = this.decompress(bits as IBitReader, stream, buf, bufsize, ret); } while (ret !== null); } /** * Parse BZIP2 header and return block size */ header(bits: IBitReader): number { this.byteCount = new Int32Array(256); this.symToByte = new Uint8Array(256); this.mtfSymbol = new Int32Array(256); this.selectors = new Uint8Array(0x8000); if (bits(8 * 3) !== 4348520) { throwError('No BZIP2 magic number found at start of stream', BZIP2_ERROR_CODES.NO_MAGIC_NUMBER); } const blockSize = (bits(8) as number) - 48; if (blockSize < 1 || blockSize > 9) { throwError('Invalid BZIP2 archive: block size must be 1-9', BZIP2_ERROR_CODES.INVALID_ARCHIVE); } return blockSize; } /** * Decompress a BZIP2 block */ decompress( bits: IBitReader, stream: (byte: number) => void, buf: Int32Array, bufsize: number, streamCRC?: number | null ): number | null { const MAX_HUFCODE_BITS = 20; const MAX_SYMBOLS = 258; const SYMBOL_RUNA = 0; const SYMBOL_RUNB = 1; const GROUP_SIZE = 50; let crc = 0 ^ -1; // Read block header let headerHex = ''; for (let i = 0; i < 6; i++) { headerHex += (bits(8) as number).toString(16); } // Check for end-of-stream marker if (headerHex === '177245385090') { const finalCRC = bits(32) as number | 0; if (finalCRC !== streamCRC) { throwError('CRC32 mismatch: stream checksum verification failed', BZIP2_ERROR_CODES.CRC_MISMATCH); } // Align stream to byte boundary bits(null); return null; } // Verify block signature (pi digits) if (headerHex !== '314159265359') { throwError('Invalid block header: expected pi signature (0x314159265359)', BZIP2_ERROR_CODES.INVALID_BLOCK_DATA); } const crcblock = bits(32) as number | 0; if (bits(1)) { throwError('Unsupported obsolete BZIP2 format version', BZIP2_ERROR_CODES.INVALID_ARCHIVE); } const origPtr = bits(24) as number; if (origPtr > bufsize) { throwError('Initial position larger than buffer size', BZIP2_ERROR_CODES.BUFFER_OVERFLOW); } // Read symbol map let symbolMapBits = bits(16) as number; let symTotal = 0; for (let i = 0; i < 16; i++) { if (symbolMapBits & (1 << (15 - i))) { const subMap = bits(16) as number; for (let j = 0; j < 16; j++) { if (subMap & (1 << (15 - j))) { this.symToByte[symTotal++] = 16 * i + j; } } } } // Read Huffman groups const groupCount = bits(3) as number; if (groupCount < 2 || groupCount > 6) { throwError('Invalid group count: must be between 2 and 6', BZIP2_ERROR_CODES.INVALID_HUFFMAN); } const nSelectors = bits(15) as number; if (nSelectors === 0) { throwError('Invalid selector count: cannot be zero', BZIP2_ERROR_CODES.INVALID_SELECTOR); } // Initialize MTF symbol array for (let i = 0; i < groupCount; i++) { this.mtfSymbol[i] = i; } // Read selectors using MTF decoding for (let i = 0; i < nSelectors; i++) { let j = 0; while (bits(1)) { j++; if (j >= groupCount) { throwError('Invalid MTF index: exceeds group count', BZIP2_ERROR_CODES.INVALID_HUFFMAN); } } const uc = this.mtfSymbol[j]; for (let k = j - 1; k >= 0; k--) { this.mtfSymbol[k + 1] = this.mtfSymbol[k]; } this.mtfSymbol[0] = uc; this.selectors[i] = uc; } // Build Huffman tables const symCount = symTotal + 2; const groups: IHuffmanGroup[] = []; const length = new Uint8Array(MAX_SYMBOLS); const temp = new Uint16Array(MAX_HUFCODE_BITS + 1); for (let j = 0; j < groupCount; j++) { let t = bits(5) as number; for (let i = 0; i < symCount; i++) { while (true) { if (t < 1 || t > MAX_HUFCODE_BITS) { throwError('Invalid Huffman code length: must be between 1 and 20', BZIP2_ERROR_CODES.INVALID_HUFFMAN); } if (!bits(1)) break; if (!bits(1)) t++; else t--; } length[i] = t; } let minLen = length[0]; let maxLen = length[0]; for (let i = 1; i < symCount; i++) { if (length[i] > maxLen) maxLen = length[i]; else if (length[i] < minLen) minLen = length[i]; } const hufGroup: IHuffmanGroup = { permute: new Int32Array(MAX_SYMBOLS), limit: new Int32Array(MAX_HUFCODE_BITS + 1), base: new Int32Array(MAX_HUFCODE_BITS + 1), minLen, maxLen, }; groups[j] = hufGroup; const base = hufGroup.base; const limit = hufGroup.limit; let pp = 0; for (let i = minLen; i <= maxLen; i++) { for (let t = 0; t < symCount; t++) { if (length[t] === i) hufGroup.permute[pp++] = t; } } for (let i = minLen; i <= maxLen; i++) { temp[i] = 0; limit[i] = 0; } for (let i = 0; i < symCount; i++) { temp[length[i]]++; } pp = 0; let tt = 0; for (let i = minLen; i < maxLen; i++) { pp += temp[i]; limit[i] = pp - 1; pp <<= 1; base[i + 1] = pp - (tt += temp[i]); } limit[maxLen] = pp + temp[maxLen] - 1; base[minLen] = 0; } // Initialize for decoding for (let i = 0; i < 256; i++) { this.mtfSymbol[i] = i; this.byteCount[i] = 0; } let runPos = 0; let count = 0; let symCountRemaining = 0; let selector = 0; let hufGroup = groups[0]; let base = hufGroup.base; let limit = hufGroup.limit; // Main decoding loop while (true) { if (!symCountRemaining--) { symCountRemaining = GROUP_SIZE - 1; if (selector >= nSelectors) { throwError('Invalid selector index: exceeds available groups', BZIP2_ERROR_CODES.INVALID_SELECTOR); } hufGroup = groups[this.selectors[selector++]]; base = hufGroup.base; limit = hufGroup.limit; } let i = hufGroup.minLen; let j = bits(i) as number; while (true) { if (i > hufGroup.maxLen) { throwError('Huffman decoding error: bit length exceeds maximum allowed', BZIP2_ERROR_CODES.INVALID_HUFFMAN); } if (j <= limit[i]) break; i++; j = (j << 1) | (bits(1) as number); } j -= base[i]; if (j < 0 || j >= MAX_SYMBOLS) { throwError('Symbol index out of bounds during Huffman decoding', BZIP2_ERROR_CODES.INVALID_HUFFMAN); } const nextSym = hufGroup.permute[j]; if (nextSym === SYMBOL_RUNA || nextSym === SYMBOL_RUNB) { if (!runPos) { runPos = 1; j = 0; } if (nextSym === SYMBOL_RUNA) j += runPos; else j += 2 * runPos; runPos <<= 1; continue; } if (runPos) { runPos = 0; const runLength = j; if (count + runLength > bufsize) { throwError('Run-length overflow: decoded run exceeds buffer capacity', BZIP2_ERROR_CODES.BUFFER_OVERFLOW); } const uc = this.symToByte[this.mtfSymbol[0]]; this.byteCount[uc] += runLength; for (let t = 0; t < runLength; t++) { buf[count++] = uc; } } if (nextSym > symTotal) break; if (count >= bufsize) { throwError('Buffer overflow: decoded data exceeds buffer capacity', BZIP2_ERROR_CODES.BUFFER_OVERFLOW); } const mtfIndex = nextSym - 1; const uc = this.mtfSymbol[mtfIndex]; for (let k = mtfIndex - 1; k >= 0; k--) { this.mtfSymbol[k + 1] = this.mtfSymbol[k]; } this.mtfSymbol[0] = uc; const decodedByte = this.symToByte[uc]; this.byteCount[decodedByte]++; buf[count++] = decodedByte; } if (origPtr < 0 || origPtr >= count) { throwError('Invalid original pointer: position outside decoded block', BZIP2_ERROR_CODES.INVALID_POSITION); } // Inverse BWT transform let j = 0; for (let i = 0; i < 256; i++) { const k = j + this.byteCount[i]; this.byteCount[i] = j; j = k; } for (let i = 0; i < count; i++) { const uc = buf[i] & 0xff; buf[this.byteCount[uc]] |= i << 8; this.byteCount[uc]++; } // Output decoded data let pos = 0; let current = 0; let run = 0; if (count) { pos = buf[origPtr]; current = pos & 0xff; pos >>= 8; run = -1; } let remaining = count; while (remaining) { remaining--; const previous = current; pos = buf[pos]; current = pos & 0xff; pos >>= 8; let copies: number; let outbyte: number; if (run++ === 3) { copies = current; outbyte = previous; current = -1; } else { copies = 1; outbyte = current; } while (copies--) { crc = ((crc << 8) ^ this.crcTable[((crc >> 24) ^ outbyte) & 0xff]) & 0xffffffff; stream(outbyte); } if (current !== previous) run = 0; } crc = (crc ^ -1) >>> 0; if ((crc | 0) !== (crcblock | 0)) { throwError('CRC32 mismatch: block checksum verification failed', BZIP2_ERROR_CODES.CRC_MISMATCH); } const newStreamCRC = (crc ^ (((streamCRC || 0) << 1) | ((streamCRC || 0) >>> 31))) & 0xffffffff; return newStreamCRC; } }