| 1 | n/a | """Interface to the libbzip2 compression library. |
|---|
| 2 | n/a | |
|---|
| 3 | n/a | This module provides a file interface, classes for incremental |
|---|
| 4 | n/a | (de)compression, and functions for one-shot (de)compression. |
|---|
| 5 | n/a | """ |
|---|
| 6 | n/a | |
|---|
| 7 | n/a | __all__ = ["BZ2File", "BZ2Compressor", "BZ2Decompressor", |
|---|
| 8 | n/a | "open", "compress", "decompress"] |
|---|
| 9 | n/a | |
|---|
| 10 | n/a | __author__ = "Nadeem Vawda <nadeem.vawda@gmail.com>" |
|---|
| 11 | n/a | |
|---|
| 12 | n/a | from builtins import open as _builtin_open |
|---|
| 13 | n/a | import io |
|---|
| 14 | n/a | import os |
|---|
| 15 | n/a | import warnings |
|---|
| 16 | n/a | import _compression |
|---|
| 17 | n/a | |
|---|
| 18 | n/a | try: |
|---|
| 19 | n/a | from threading import RLock |
|---|
| 20 | n/a | except ImportError: |
|---|
| 21 | n/a | from dummy_threading import RLock |
|---|
| 22 | n/a | |
|---|
| 23 | n/a | from _bz2 import BZ2Compressor, BZ2Decompressor |
|---|
| 24 | n/a | |
|---|
| 25 | n/a | |
|---|
| 26 | n/a | _MODE_CLOSED = 0 |
|---|
| 27 | n/a | _MODE_READ = 1 |
|---|
| 28 | n/a | # Value 2 no longer used |
|---|
| 29 | n/a | _MODE_WRITE = 3 |
|---|
| 30 | n/a | |
|---|
| 31 | n/a | |
|---|
| 32 | n/a | class BZ2File(_compression.BaseStream): |
|---|
| 33 | n/a | |
|---|
| 34 | n/a | """A file object providing transparent bzip2 (de)compression. |
|---|
| 35 | n/a | |
|---|
| 36 | n/a | A BZ2File can act as a wrapper for an existing file object, or refer |
|---|
| 37 | n/a | directly to a named file on disk. |
|---|
| 38 | n/a | |
|---|
| 39 | n/a | Note that BZ2File provides a *binary* file interface - data read is |
|---|
| 40 | n/a | returned as bytes, and data to be written should be given as bytes. |
|---|
| 41 | n/a | """ |
|---|
| 42 | n/a | |
|---|
| 43 | n/a | def __init__(self, filename, mode="r", buffering=None, compresslevel=9): |
|---|
| 44 | n/a | """Open a bzip2-compressed file. |
|---|
| 45 | n/a | |
|---|
| 46 | n/a | If filename is a str, bytes, or PathLike object, it gives the |
|---|
| 47 | n/a | name of the file to be opened. Otherwise, it should be a file |
|---|
| 48 | n/a | object, which will be used to read or write the compressed data. |
|---|
| 49 | n/a | |
|---|
| 50 | n/a | mode can be 'r' for reading (default), 'w' for (over)writing, |
|---|
| 51 | n/a | 'x' for creating exclusively, or 'a' for appending. These can |
|---|
| 52 | n/a | equivalently be given as 'rb', 'wb', 'xb', and 'ab'. |
|---|
| 53 | n/a | |
|---|
| 54 | n/a | buffering is ignored. Its use is deprecated. |
|---|
| 55 | n/a | |
|---|
| 56 | n/a | If mode is 'w', 'x' or 'a', compresslevel can be a number between 1 |
|---|
| 57 | n/a | and 9 specifying the level of compression: 1 produces the least |
|---|
| 58 | n/a | compression, and 9 (default) produces the most compression. |
|---|
| 59 | n/a | |
|---|
| 60 | n/a | If mode is 'r', the input file may be the concatenation of |
|---|
| 61 | n/a | multiple compressed streams. |
|---|
| 62 | n/a | """ |
|---|
| 63 | n/a | # This lock must be recursive, so that BufferedIOBase's |
|---|
| 64 | n/a | # writelines() does not deadlock. |
|---|
| 65 | n/a | self._lock = RLock() |
|---|
| 66 | n/a | self._fp = None |
|---|
| 67 | n/a | self._closefp = False |
|---|
| 68 | n/a | self._mode = _MODE_CLOSED |
|---|
| 69 | n/a | |
|---|
| 70 | n/a | if buffering is not None: |
|---|
| 71 | n/a | warnings.warn("Use of 'buffering' argument is deprecated", |
|---|
| 72 | n/a | DeprecationWarning) |
|---|
| 73 | n/a | |
|---|
| 74 | n/a | if not (1 <= compresslevel <= 9): |
|---|
| 75 | n/a | raise ValueError("compresslevel must be between 1 and 9") |
|---|
| 76 | n/a | |
|---|
| 77 | n/a | if mode in ("", "r", "rb"): |
|---|
| 78 | n/a | mode = "rb" |
|---|
| 79 | n/a | mode_code = _MODE_READ |
|---|
| 80 | n/a | elif mode in ("w", "wb"): |
|---|
| 81 | n/a | mode = "wb" |
|---|
| 82 | n/a | mode_code = _MODE_WRITE |
|---|
| 83 | n/a | self._compressor = BZ2Compressor(compresslevel) |
|---|
| 84 | n/a | elif mode in ("x", "xb"): |
|---|
| 85 | n/a | mode = "xb" |
|---|
| 86 | n/a | mode_code = _MODE_WRITE |
|---|
| 87 | n/a | self._compressor = BZ2Compressor(compresslevel) |
|---|
| 88 | n/a | elif mode in ("a", "ab"): |
|---|
| 89 | n/a | mode = "ab" |
|---|
| 90 | n/a | mode_code = _MODE_WRITE |
|---|
| 91 | n/a | self._compressor = BZ2Compressor(compresslevel) |
|---|
| 92 | n/a | else: |
|---|
| 93 | n/a | raise ValueError("Invalid mode: %r" % (mode,)) |
|---|
| 94 | n/a | |
|---|
| 95 | n/a | if isinstance(filename, (str, bytes, os.PathLike)): |
|---|
| 96 | n/a | self._fp = _builtin_open(filename, mode) |
|---|
| 97 | n/a | self._closefp = True |
|---|
| 98 | n/a | self._mode = mode_code |
|---|
| 99 | n/a | elif hasattr(filename, "read") or hasattr(filename, "write"): |
|---|
| 100 | n/a | self._fp = filename |
|---|
| 101 | n/a | self._mode = mode_code |
|---|
| 102 | n/a | else: |
|---|
| 103 | n/a | raise TypeError("filename must be a str, bytes, file or PathLike object") |
|---|
| 104 | n/a | |
|---|
| 105 | n/a | if self._mode == _MODE_READ: |
|---|
| 106 | n/a | raw = _compression.DecompressReader(self._fp, |
|---|
| 107 | n/a | BZ2Decompressor, trailing_error=OSError) |
|---|
| 108 | n/a | self._buffer = io.BufferedReader(raw) |
|---|
| 109 | n/a | else: |
|---|
| 110 | n/a | self._pos = 0 |
|---|
| 111 | n/a | |
|---|
| 112 | n/a | def close(self): |
|---|
| 113 | n/a | """Flush and close the file. |
|---|
| 114 | n/a | |
|---|
| 115 | n/a | May be called more than once without error. Once the file is |
|---|
| 116 | n/a | closed, any other operation on it will raise a ValueError. |
|---|
| 117 | n/a | """ |
|---|
| 118 | n/a | with self._lock: |
|---|
| 119 | n/a | if self._mode == _MODE_CLOSED: |
|---|
| 120 | n/a | return |
|---|
| 121 | n/a | try: |
|---|
| 122 | n/a | if self._mode == _MODE_READ: |
|---|
| 123 | n/a | self._buffer.close() |
|---|
| 124 | n/a | elif self._mode == _MODE_WRITE: |
|---|
| 125 | n/a | self._fp.write(self._compressor.flush()) |
|---|
| 126 | n/a | self._compressor = None |
|---|
| 127 | n/a | finally: |
|---|
| 128 | n/a | try: |
|---|
| 129 | n/a | if self._closefp: |
|---|
| 130 | n/a | self._fp.close() |
|---|
| 131 | n/a | finally: |
|---|
| 132 | n/a | self._fp = None |
|---|
| 133 | n/a | self._closefp = False |
|---|
| 134 | n/a | self._mode = _MODE_CLOSED |
|---|
| 135 | n/a | self._buffer = None |
|---|
| 136 | n/a | |
|---|
| 137 | n/a | @property |
|---|
| 138 | n/a | def closed(self): |
|---|
| 139 | n/a | """True if this file is closed.""" |
|---|
| 140 | n/a | return self._mode == _MODE_CLOSED |
|---|
| 141 | n/a | |
|---|
| 142 | n/a | def fileno(self): |
|---|
| 143 | n/a | """Return the file descriptor for the underlying file.""" |
|---|
| 144 | n/a | self._check_not_closed() |
|---|
| 145 | n/a | return self._fp.fileno() |
|---|
| 146 | n/a | |
|---|
| 147 | n/a | def seekable(self): |
|---|
| 148 | n/a | """Return whether the file supports seeking.""" |
|---|
| 149 | n/a | return self.readable() and self._buffer.seekable() |
|---|
| 150 | n/a | |
|---|
| 151 | n/a | def readable(self): |
|---|
| 152 | n/a | """Return whether the file was opened for reading.""" |
|---|
| 153 | n/a | self._check_not_closed() |
|---|
| 154 | n/a | return self._mode == _MODE_READ |
|---|
| 155 | n/a | |
|---|
| 156 | n/a | def writable(self): |
|---|
| 157 | n/a | """Return whether the file was opened for writing.""" |
|---|
| 158 | n/a | self._check_not_closed() |
|---|
| 159 | n/a | return self._mode == _MODE_WRITE |
|---|
| 160 | n/a | |
|---|
| 161 | n/a | def peek(self, n=0): |
|---|
| 162 | n/a | """Return buffered data without advancing the file position. |
|---|
| 163 | n/a | |
|---|
| 164 | n/a | Always returns at least one byte of data, unless at EOF. |
|---|
| 165 | n/a | The exact number of bytes returned is unspecified. |
|---|
| 166 | n/a | """ |
|---|
| 167 | n/a | with self._lock: |
|---|
| 168 | n/a | self._check_can_read() |
|---|
| 169 | n/a | # Relies on the undocumented fact that BufferedReader.peek() |
|---|
| 170 | n/a | # always returns at least one byte (except at EOF), independent |
|---|
| 171 | n/a | # of the value of n |
|---|
| 172 | n/a | return self._buffer.peek(n) |
|---|
| 173 | n/a | |
|---|
| 174 | n/a | def read(self, size=-1): |
|---|
| 175 | n/a | """Read up to size uncompressed bytes from the file. |
|---|
| 176 | n/a | |
|---|
| 177 | n/a | If size is negative or omitted, read until EOF is reached. |
|---|
| 178 | n/a | Returns b'' if the file is already at EOF. |
|---|
| 179 | n/a | """ |
|---|
| 180 | n/a | with self._lock: |
|---|
| 181 | n/a | self._check_can_read() |
|---|
| 182 | n/a | return self._buffer.read(size) |
|---|
| 183 | n/a | |
|---|
| 184 | n/a | def read1(self, size=-1): |
|---|
| 185 | n/a | """Read up to size uncompressed bytes, while trying to avoid |
|---|
| 186 | n/a | making multiple reads from the underlying stream. Reads up to a |
|---|
| 187 | n/a | buffer's worth of data if size is negative. |
|---|
| 188 | n/a | |
|---|
| 189 | n/a | Returns b'' if the file is at EOF. |
|---|
| 190 | n/a | """ |
|---|
| 191 | n/a | with self._lock: |
|---|
| 192 | n/a | self._check_can_read() |
|---|
| 193 | n/a | if size < 0: |
|---|
| 194 | n/a | size = io.DEFAULT_BUFFER_SIZE |
|---|
| 195 | n/a | return self._buffer.read1(size) |
|---|
| 196 | n/a | |
|---|
| 197 | n/a | def readinto(self, b): |
|---|
| 198 | n/a | """Read bytes into b. |
|---|
| 199 | n/a | |
|---|
| 200 | n/a | Returns the number of bytes read (0 for EOF). |
|---|
| 201 | n/a | """ |
|---|
| 202 | n/a | with self._lock: |
|---|
| 203 | n/a | self._check_can_read() |
|---|
| 204 | n/a | return self._buffer.readinto(b) |
|---|
| 205 | n/a | |
|---|
| 206 | n/a | def readline(self, size=-1): |
|---|
| 207 | n/a | """Read a line of uncompressed bytes from the file. |
|---|
| 208 | n/a | |
|---|
| 209 | n/a | The terminating newline (if present) is retained. If size is |
|---|
| 210 | n/a | non-negative, no more than size bytes will be read (in which |
|---|
| 211 | n/a | case the line may be incomplete). Returns b'' if already at EOF. |
|---|
| 212 | n/a | """ |
|---|
| 213 | n/a | if not isinstance(size, int): |
|---|
| 214 | n/a | if not hasattr(size, "__index__"): |
|---|
| 215 | n/a | raise TypeError("Integer argument expected") |
|---|
| 216 | n/a | size = size.__index__() |
|---|
| 217 | n/a | with self._lock: |
|---|
| 218 | n/a | self._check_can_read() |
|---|
| 219 | n/a | return self._buffer.readline(size) |
|---|
| 220 | n/a | |
|---|
| 221 | n/a | def readlines(self, size=-1): |
|---|
| 222 | n/a | """Read a list of lines of uncompressed bytes from the file. |
|---|
| 223 | n/a | |
|---|
| 224 | n/a | size can be specified to control the number of lines read: no |
|---|
| 225 | n/a | further lines will be read once the total size of the lines read |
|---|
| 226 | n/a | so far equals or exceeds size. |
|---|
| 227 | n/a | """ |
|---|
| 228 | n/a | if not isinstance(size, int): |
|---|
| 229 | n/a | if not hasattr(size, "__index__"): |
|---|
| 230 | n/a | raise TypeError("Integer argument expected") |
|---|
| 231 | n/a | size = size.__index__() |
|---|
| 232 | n/a | with self._lock: |
|---|
| 233 | n/a | self._check_can_read() |
|---|
| 234 | n/a | return self._buffer.readlines(size) |
|---|
| 235 | n/a | |
|---|
| 236 | n/a | def write(self, data): |
|---|
| 237 | n/a | """Write a byte string to the file. |
|---|
| 238 | n/a | |
|---|
| 239 | n/a | Returns the number of uncompressed bytes written, which is |
|---|
| 240 | n/a | always len(data). Note that due to buffering, the file on disk |
|---|
| 241 | n/a | may not reflect the data written until close() is called. |
|---|
| 242 | n/a | """ |
|---|
| 243 | n/a | with self._lock: |
|---|
| 244 | n/a | self._check_can_write() |
|---|
| 245 | n/a | compressed = self._compressor.compress(data) |
|---|
| 246 | n/a | self._fp.write(compressed) |
|---|
| 247 | n/a | self._pos += len(data) |
|---|
| 248 | n/a | return len(data) |
|---|
| 249 | n/a | |
|---|
| 250 | n/a | def writelines(self, seq): |
|---|
| 251 | n/a | """Write a sequence of byte strings to the file. |
|---|
| 252 | n/a | |
|---|
| 253 | n/a | Returns the number of uncompressed bytes written. |
|---|
| 254 | n/a | seq can be any iterable yielding byte strings. |
|---|
| 255 | n/a | |
|---|
| 256 | n/a | Line separators are not added between the written byte strings. |
|---|
| 257 | n/a | """ |
|---|
| 258 | n/a | with self._lock: |
|---|
| 259 | n/a | return _compression.BaseStream.writelines(self, seq) |
|---|
| 260 | n/a | |
|---|
| 261 | n/a | def seek(self, offset, whence=io.SEEK_SET): |
|---|
| 262 | n/a | """Change the file position. |
|---|
| 263 | n/a | |
|---|
| 264 | n/a | The new position is specified by offset, relative to the |
|---|
| 265 | n/a | position indicated by whence. Values for whence are: |
|---|
| 266 | n/a | |
|---|
| 267 | n/a | 0: start of stream (default); offset must not be negative |
|---|
| 268 | n/a | 1: current stream position |
|---|
| 269 | n/a | 2: end of stream; offset must not be positive |
|---|
| 270 | n/a | |
|---|
| 271 | n/a | Returns the new file position. |
|---|
| 272 | n/a | |
|---|
| 273 | n/a | Note that seeking is emulated, so depending on the parameters, |
|---|
| 274 | n/a | this operation may be extremely slow. |
|---|
| 275 | n/a | """ |
|---|
| 276 | n/a | with self._lock: |
|---|
| 277 | n/a | self._check_can_seek() |
|---|
| 278 | n/a | return self._buffer.seek(offset, whence) |
|---|
| 279 | n/a | |
|---|
| 280 | n/a | def tell(self): |
|---|
| 281 | n/a | """Return the current file position.""" |
|---|
| 282 | n/a | with self._lock: |
|---|
| 283 | n/a | self._check_not_closed() |
|---|
| 284 | n/a | if self._mode == _MODE_READ: |
|---|
| 285 | n/a | return self._buffer.tell() |
|---|
| 286 | n/a | return self._pos |
|---|
| 287 | n/a | |
|---|
| 288 | n/a | |
|---|
| 289 | n/a | def open(filename, mode="rb", compresslevel=9, |
|---|
| 290 | n/a | encoding=None, errors=None, newline=None): |
|---|
| 291 | n/a | """Open a bzip2-compressed file in binary or text mode. |
|---|
| 292 | n/a | |
|---|
| 293 | n/a | The filename argument can be an actual filename (a str, bytes, or |
|---|
| 294 | n/a | PathLike object), or an existing file object to read from or write |
|---|
| 295 | n/a | to. |
|---|
| 296 | n/a | |
|---|
| 297 | n/a | The mode argument can be "r", "rb", "w", "wb", "x", "xb", "a" or |
|---|
| 298 | n/a | "ab" for binary mode, or "rt", "wt", "xt" or "at" for text mode. |
|---|
| 299 | n/a | The default mode is "rb", and the default compresslevel is 9. |
|---|
| 300 | n/a | |
|---|
| 301 | n/a | For binary mode, this function is equivalent to the BZ2File |
|---|
| 302 | n/a | constructor: BZ2File(filename, mode, compresslevel). In this case, |
|---|
| 303 | n/a | the encoding, errors and newline arguments must not be provided. |
|---|
| 304 | n/a | |
|---|
| 305 | n/a | For text mode, a BZ2File object is created, and wrapped in an |
|---|
| 306 | n/a | io.TextIOWrapper instance with the specified encoding, error |
|---|
| 307 | n/a | handling behavior, and line ending(s). |
|---|
| 308 | n/a | |
|---|
| 309 | n/a | """ |
|---|
| 310 | n/a | if "t" in mode: |
|---|
| 311 | n/a | if "b" in mode: |
|---|
| 312 | n/a | raise ValueError("Invalid mode: %r" % (mode,)) |
|---|
| 313 | n/a | else: |
|---|
| 314 | n/a | if encoding is not None: |
|---|
| 315 | n/a | raise ValueError("Argument 'encoding' not supported in binary mode") |
|---|
| 316 | n/a | if errors is not None: |
|---|
| 317 | n/a | raise ValueError("Argument 'errors' not supported in binary mode") |
|---|
| 318 | n/a | if newline is not None: |
|---|
| 319 | n/a | raise ValueError("Argument 'newline' not supported in binary mode") |
|---|
| 320 | n/a | |
|---|
| 321 | n/a | bz_mode = mode.replace("t", "") |
|---|
| 322 | n/a | binary_file = BZ2File(filename, bz_mode, compresslevel=compresslevel) |
|---|
| 323 | n/a | |
|---|
| 324 | n/a | if "t" in mode: |
|---|
| 325 | n/a | return io.TextIOWrapper(binary_file, encoding, errors, newline) |
|---|
| 326 | n/a | else: |
|---|
| 327 | n/a | return binary_file |
|---|
| 328 | n/a | |
|---|
| 329 | n/a | |
|---|
| 330 | n/a | def compress(data, compresslevel=9): |
|---|
| 331 | n/a | """Compress a block of data. |
|---|
| 332 | n/a | |
|---|
| 333 | n/a | compresslevel, if given, must be a number between 1 and 9. |
|---|
| 334 | n/a | |
|---|
| 335 | n/a | For incremental compression, use a BZ2Compressor object instead. |
|---|
| 336 | n/a | """ |
|---|
| 337 | n/a | comp = BZ2Compressor(compresslevel) |
|---|
| 338 | n/a | return comp.compress(data) + comp.flush() |
|---|
| 339 | n/a | |
|---|
| 340 | n/a | |
|---|
| 341 | n/a | def decompress(data): |
|---|
| 342 | n/a | """Decompress a block of data. |
|---|
| 343 | n/a | |
|---|
| 344 | n/a | For incremental decompression, use a BZ2Decompressor object instead. |
|---|
| 345 | n/a | """ |
|---|
| 346 | n/a | results = [] |
|---|
| 347 | n/a | while data: |
|---|
| 348 | n/a | decomp = BZ2Decompressor() |
|---|
| 349 | n/a | try: |
|---|
| 350 | n/a | res = decomp.decompress(data) |
|---|
| 351 | n/a | except OSError: |
|---|
| 352 | n/a | if results: |
|---|
| 353 | n/a | break # Leftover data is not a valid bzip2 stream; ignore it. |
|---|
| 354 | n/a | else: |
|---|
| 355 | n/a | raise # Error on the first iteration; bail out. |
|---|
| 356 | n/a | results.append(res) |
|---|
| 357 | n/a | if not decomp.eof: |
|---|
| 358 | n/a | raise ValueError("Compressed data ended before the " |
|---|
| 359 | n/a | "end-of-stream marker was reached") |
|---|
| 360 | n/a | data = decomp.unused_data |
|---|
| 361 | n/a | return b"".join(results) |
|---|