| 1 | #!/usr/bin/env python |
| 2 | # -*- coding: utf-8 -*- |
| 3 | # |
| 4 | # dtrx -- Intelligently extract various archive types. |
| 5 | # Copyright © 2006-2009 Brett Smith <brettcsmith@brettcsmith.org> |
| 6 | # Copyright © 2008 Peter Kelemen <Peter.Kelemen@gmail.com> |
| 7 | # |
| 8 | # This program is free software; you can redistribute it and/or modify it |
| 9 | # under the terms of the GNU General Public License as published by the |
| 10 | # Free Software Foundation; either version 3 of the License, or (at your |
| 11 | # option) any later version. |
| 12 | # |
| 13 | # This program is distributed in the hope that it will be useful, but |
| 14 | # WITHOUT ANY WARRANTY; without even the implied warranty of |
| 15 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General |
| 16 | # Public License for more details. |
| 17 | # |
| 18 | # You should have received a copy of the GNU General Public License along |
| 19 | # with this program; if not, see <http://www.gnu.org/licenses/>. |
| 20 | |
| 21 | # Python 2.3 string methods: 'rfind', 'rindex', 'rjust', 'rstrip' |
| 22 | |
| 23 | import errno |
| 24 | import fcntl |
| 25 | import logging |
| 26 | import mimetypes |
| 27 | import optparse |
| 28 | import os |
| 29 | import re |
| 30 | import shutil |
| 31 | import signal |
| 32 | import stat |
| 33 | import string |
| 34 | import struct |
| 35 | import subprocess |
| 36 | import sys |
| 37 | import tempfile |
| 38 | import termios |
| 39 | import textwrap |
| 40 | import traceback |
| 41 | import urllib |
| 42 | |
| 43 | try: |
| 44 | set |
| 45 | except NameError: |
| 46 | from sets import Set as set |
| 47 | |
| 48 | VERSION = "6.6" |
| 49 | VERSION_BANNER = """dtrx version %s |
| 50 | Copyright © 2006-2009 Brett Smith <brettcsmith@brettcsmith.org> |
| 51 | Copyright © 2008 Peter Kelemen <Peter.Kelemen@gmail.com> |
| 52 | |
| 53 | This program is free software; you can redistribute it and/or modify it |
| 54 | under the terms of the GNU General Public License as published by the |
| 55 | Free Software Foundation; either version 3 of the License, or (at your |
| 56 | option) any later version. |
| 57 | |
| 58 | This program is distributed in the hope that it will be useful, but |
| 59 | WITHOUT ANY WARRANTY; without even the implied warranty of |
| 60 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General |
| 61 | Public License for more details.""" % (VERSION,) |
| 62 | |
| 63 | MATCHING_DIRECTORY = 1 |
| 64 | ONE_ENTRY_KNOWN = 2 |
| 65 | BOMB = 3 |
| 66 | EMPTY = 4 |
| 67 | ONE_ENTRY_FILE = 'file' |
| 68 | ONE_ENTRY_DIRECTORY = 'directory' |
| 69 | |
| 70 | ONE_ENTRY_UNKNOWN = [ONE_ENTRY_FILE, ONE_ENTRY_DIRECTORY] |
| 71 | |
| 72 | EXTRACT_HERE = 1 |
| 73 | EXTRACT_WRAP = 2 |
| 74 | EXTRACT_RENAME = 3 |
| 75 | |
| 76 | RECURSE_ALWAYS = 1 |
| 77 | RECURSE_ONCE = 2 |
| 78 | RECURSE_NOT_NOW = 3 |
| 79 | RECURSE_NEVER = 4 |
| 80 | RECURSE_LIST = 5 |
| 81 | |
| 82 | mimetypes.encodings_map.setdefault('.bz2', 'bzip2') |
| 83 | mimetypes.encodings_map.setdefault('.lzma', 'lzma') |
| 84 | mimetypes.encodings_map.setdefault('.xz', 'xz') |
| 85 | mimetypes.types_map.setdefault('.gem', 'application/x-ruby-gem') |
| 86 | |
| 87 | logger = logging.getLogger('dtrx-log') |
| 88 | |
| 89 | class FilenameChecker(object): |
| 90 | free_func = os.open |
| 91 | free_args = (os.O_CREAT | os.O_EXCL,) |
| 92 | free_close = os.close |
| 93 | |
| 94 | def __init__(self, original_name): |
| 95 | self.original_name = original_name |
| 96 | |
| 97 | def is_free(self, filename): |
| 98 | try: |
| 99 | result = self.free_func(filename, *self.free_args) |
| 100 | except OSError, error: |
| 101 | if error.errno == errno.EEXIST: |
| 102 | return False |
| 103 | raise |
| 104 | if self.free_close: |
| 105 | self.free_close(result) |
| 106 | return True |
| 107 | |
| 108 | def create(self): |
| 109 | fd, filename = tempfile.mkstemp(prefix=self.original_name + '.', |
| 110 | dir='.') |
| 111 | os.close(fd) |
| 112 | return filename |
| 113 | |
| 114 | def check(self): |
| 115 | for suffix in [''] + ['.%s' % (x,) for x in range(1, 10)]: |
| 116 | filename = '%s%s' % (self.original_name, suffix) |
| 117 | if self.is_free(filename): |
| 118 | return filename |
| 119 | return self.create() |
| 120 | |
| 121 | |
| 122 | class DirectoryChecker(FilenameChecker): |
| 123 | free_func = os.mkdir |
| 124 | free_args = () |
| 125 | free_close = None |
| 126 | |
| 127 | def create(self): |
| 128 | return tempfile.mkdtemp(prefix=self.original_name + '.', dir='.') |
| 129 | |
| 130 | |
| 131 | class ExtractorError(Exception): |
| 132 | pass |
| 133 | |
| 134 | |
| 135 | class ExtractorUnusable(Exception): |
| 136 | pass |
| 137 | |
| 138 | |
| 139 | EXTRACTION_ERRORS = (ExtractorError, ExtractorUnusable, OSError, IOError) |
| 140 | |
| 141 | class BaseExtractor(object): |
| 142 | decoders = {'bzip2': 'bzcat', 'gzip': 'zcat', 'compress': 'zcat', |
| 143 | 'lzma': 'lzcat', 'xz': 'xzcat'} |
| 144 | name_checker = DirectoryChecker |
| 145 | |
| 146 | def __init__(self, filename, encoding): |
| 147 | if encoding and (not self.decoders.has_key(encoding)): |
| 148 | raise ValueError("unrecognized encoding %s" % (encoding,)) |
| 149 | self.filename = os.path.realpath(filename) |
| 150 | self.encoding = encoding |
| 151 | self.file_count = 0 |
| 152 | self.included_archives = [] |
| 153 | self.target = None |
| 154 | self.content_type = None |
| 155 | self.content_name = None |
| 156 | self.pipes = [] |
| 157 | self.stderr = tempfile.TemporaryFile() |
| 158 | self.exit_codes = [] |
| 159 | try: |
| 160 | self.archive = open(filename, 'r') |
| 161 | except (IOError, OSError), error: |
| 162 | raise ExtractorError("could not open %s: %s" % |
| 163 | (filename, error.strerror)) |
| 164 | if encoding: |
| 165 | self.pipe([self.decoders[encoding]], "decoding") |
| 166 | self.prepare() |
| 167 | |
| 168 | def pipe(self, command, description="extraction"): |
| 169 | self.pipes.append((command, description)) |
| 170 | |
| 171 | def first_bad_exit_code(self): |
| 172 | for index, code in enumerate(self.exit_codes): |
| 173 | if code != 0: |
| 174 | return index |
| 175 | return None |
| 176 | |
| 177 | def add_process(self, processes, command, stdin, stdout): |
| 178 | try: |
| 179 | processes.append(subprocess.Popen(command, stdin=stdin, |
| 180 | stdout=stdout, |
| 181 | stderr=self.stderr)) |
| 182 | except OSError, error: |
| 183 | if error.errno == errno.ENOENT: |
| 184 | raise ExtractorUnusable("could not run %s" % (command[0],)) |
| 185 | raise |
| 186 | |
| 187 | def run_pipes(self, final_stdout=None): |
| 188 | if not self.pipes: |
| 189 | return |
| 190 | elif final_stdout is None: |
| 191 | final_stdout = open('/dev/null', 'w') |
| 192 | num_pipes = len(self.pipes) |
| 193 | last_pipe = num_pipes - 1 |
| 194 | processes = [] |
| 195 | for index, command in enumerate([pipe[0] for pipe in self.pipes]): |
| 196 | if index == 0: |
| 197 | stdin = self.archive |
| 198 | else: |
| 199 | stdin = processes[-1].stdout |
| 200 | if index == last_pipe: |
| 201 | stdout = final_stdout |
| 202 | else: |
| 203 | stdout = subprocess.PIPE |
| 204 | self.add_process(processes, command, stdin, stdout) |
| 205 | self.exit_codes = [pipe.wait() for pipe in processes] |
| 206 | self.archive.close() |
| 207 | for index in range(last_pipe): |
| 208 | processes[index].stdout.close() |
| 209 | self.archive = final_stdout |
| 210 | |
| 211 | def prepare(self): |
| 212 | pass |
| 213 | |
| 214 | def check_included_archives(self): |
| 215 | if (self.content_name is None) or (not self.content_name.endswith('/')): |
| 216 | self.included_root = './' |
| 217 | else: |
| 218 | self.included_root = self.content_name |
| 219 | start_index = len(self.included_root) |
| 220 | for path, dirname, filenames in os.walk(self.included_root): |
| 221 | self.file_count += len(filenames) |
| 222 | path = path[start_index:] |
| 223 | for filename in filenames: |
| 224 | if (ExtractorBuilder.try_by_mimetype(filename) or |
| 225 | ExtractorBuilder.try_by_extension(filename)): |
| 226 | self.included_archives.append(os.path.join(path, filename)) |
| 227 | |
| 228 | def check_contents(self): |
| 229 | if not self.contents: |
| 230 | self.content_type = EMPTY |
| 231 | elif len(self.contents) == 1: |
| 232 | if self.basename() == self.contents[0]: |
| 233 | self.content_type = MATCHING_DIRECTORY |
| 234 | elif os.path.isdir(self.contents[0]): |
| 235 | self.content_type = ONE_ENTRY_DIRECTORY |
| 236 | else: |
| 237 | self.content_type = ONE_ENTRY_FILE |
| 238 | self.content_name = self.contents[0] |
| 239 | if os.path.isdir(self.contents[0]): |
| 240 | self.content_name += '/' |
| 241 | else: |
| 242 | self.content_type = BOMB |
| 243 | self.check_included_archives() |
| 244 | |
| 245 | def basename(self): |
| 246 | pieces = os.path.basename(self.filename).split('.') |
| 247 | extension = '.' + pieces[-1] |
| 248 | if mimetypes.encodings_map.has_key(extension): |
| 249 | pieces.pop() |
| 250 | extension = '.' + pieces[-1] |
| 251 | if (mimetypes.types_map.has_key(extension) or |
| 252 | mimetypes.common_types.has_key(extension) or |
| 253 | mimetypes.suffix_map.has_key(extension)): |
| 254 | pieces.pop() |
| 255 | return '.'.join(pieces) |
| 256 | |
| 257 | def get_stderr(self): |
| 258 | self.stderr.seek(0, 0) |
| 259 | errors = self.stderr.read(-1) |
| 260 | self.stderr.close() |
| 261 | return errors |
| 262 | |
| 263 | def check_success(self, got_output): |
| 264 | error_index = self.first_bad_exit_code() |
| 265 | if (not got_output) and (error_index is not None): |
| 266 | command = ' '.join(self.pipes[error_index][0]) |
| 267 | raise ExtractorError("%s error: '%s' returned status code %s" % |
| 268 | (self.pipes[error_index][1], command, |
| 269 | self.exit_codes[error_index])) |
| 270 | |
| 271 | def extract_archive(self): |
| 272 | self.pipe(self.extract_pipe) |
| 273 | self.run_pipes() |
| 274 | |
| 275 | def extract(self): |
| 276 | try: |
| 277 | self.target = tempfile.mkdtemp(prefix='.dtrx-', dir='.') |
| 278 | except (OSError, IOError), error: |
| 279 | raise ExtractorError("cannot extract here: %s" % (error.strerror,)) |
| 280 | old_path = os.path.realpath(os.curdir) |
| 281 | os.chdir(self.target) |
| 282 | try: |
| 283 | self.archive.seek(0, 0) |
| 284 | self.extract_archive() |
| 285 | self.contents = os.listdir('.') |
| 286 | self.check_contents() |
| 287 | self.check_success(self.content_type != EMPTY) |
| 288 | except EXTRACTION_ERRORS: |
| 289 | self.archive.close() |
| 290 | os.chdir(old_path) |
| 291 | shutil.rmtree(self.target, ignore_errors=True) |
| 292 | raise |
| 293 | self.archive.close() |
| 294 | os.chdir(old_path) |
| 295 | |
| 296 | def get_filenames(self, internal=False): |
| 297 | if not internal: |
| 298 | self.pipe(self.list_pipe, "listing") |
| 299 | processes = [] |
| 300 | stdin = self.archive |
| 301 | for command in [pipe[0] for pipe in self.pipes]: |
| 302 | self.add_process(processes, command, stdin, subprocess.PIPE) |
| 303 | stdin = processes[-1].stdout |
| 304 | get_output_line = processes[-1].stdout.readline |
| 305 | while True: |
| 306 | line = get_output_line() |
| 307 | if not line: |
| 308 | break |
| 309 | yield line.rstrip('\n') |
| 310 | self.exit_codes = [pipe.wait() for pipe in processes] |
| 311 | self.archive.close() |
| 312 | for process in processes: |
| 313 | process.stdout.close() |
| 314 | self.check_success(False) |
| 315 | |
| 316 | |
| 317 | class CompressionExtractor(BaseExtractor): |
| 318 | file_type = 'compressed file' |
| 319 | name_checker = FilenameChecker |
| 320 | |
| 321 | def basename(self): |
| 322 | pieces = os.path.basename(self.filename).split('.') |
| 323 | extension = '.' + pieces[-1] |
| 324 | if mimetypes.encodings_map.has_key(extension): |
| 325 | pieces.pop() |
| 326 | return '.'.join(pieces) |
| 327 | |
| 328 | def get_filenames(self): |
| 329 | # This code used to just immediately yield the basename, under the |
| 330 | # assumption that that would be the filename. However, if that |
| 331 | # happens, dtrx -l will report this as a valid result for files with |
| 332 | # compression extensions, even if those files shouldn't actually be |
| 333 | # handled this way. So, we call out to the file command to do a quick |
| 334 | # check and make sure this actually looks like a compressed file. |
| 335 | if 'compress' not in [match[0] for match in |
| 336 | ExtractorBuilder.try_by_magic(self.filename)]: |
| 337 | raise ExtractorError("doesn't look like a compressed file") |
| 338 | yield self.basename() |
| 339 | |
| 340 | def extract(self): |
| 341 | self.content_type = ONE_ENTRY_KNOWN |
| 342 | self.content_name = self.basename() |
| 343 | self.contents = None |
| 344 | self.included_root = './' |
| 345 | try: |
| 346 | output_fd, self.target = tempfile.mkstemp(prefix='.dtrx-', dir='.') |
| 347 | except (OSError, IOError), error: |
| 348 | raise ExtractorError("cannot extract here: %s" % (error.strerror,)) |
| 349 | self.run_pipes(output_fd) |
| 350 | os.close(output_fd) |
| 351 | try: |
| 352 | self.check_success(os.stat(self.target)[stat.ST_SIZE] > 0) |
| 353 | except EXTRACTION_ERRORS: |
| 354 | os.unlink(self.target) |
| 355 | raise |
| 356 | |
| 357 | class TarExtractor(BaseExtractor): |
| 358 | file_type = 'tar file' |
| 359 | extract_pipe = ['tar', '-x'] |
| 360 | list_pipe = ['tar', '-t'] |
| 361 | |
| 362 | |
| 363 | class CpioExtractor(BaseExtractor): |
| 364 | file_type = 'cpio file' |
| 365 | extract_pipe = ['cpio', '-i', '--make-directories', '--quiet', |
| 366 | '--no-absolute-filenames'] |
| 367 | list_pipe = ['cpio', '-t', '--quiet'] |
| 368 | |
| 369 | |
| 370 | class RPMExtractor(CpioExtractor): |
| 371 | file_type = 'RPM' |
| 372 | |
| 373 | def prepare(self): |
| 374 | self.pipe(['rpm2cpio', '-'], "rpm2cpio") |
| 375 | |
| 376 | def basename(self): |
| 377 | pieces = os.path.basename(self.filename).split('.') |
| 378 | if len(pieces) == 1: |
| 379 | return pieces[0] |
| 380 | elif pieces[-1] != 'rpm': |
| 381 | return BaseExtractor.basename(self) |
| 382 | pieces.pop() |
| 383 | if len(pieces) == 1: |
| 384 | return pieces[0] |
| 385 | elif len(pieces[-1]) < 8: |
| 386 | pieces.pop() |
| 387 | return '.'.join(pieces) |
| 388 | |
| 389 | def check_contents(self): |
| 390 | self.check_included_archives() |
| 391 | self.content_type = BOMB |
| 392 | |
| 393 | |
| 394 | class DebExtractor(TarExtractor): |
| 395 | file_type = 'Debian package' |
| 396 | data_re = re.compile(r'^data\.tar\.[a-z0-9]+$') |
| 397 | |
| 398 | def prepare(self): |
| 399 | self.pipe(['ar', 't', self.filename], "finding package data file") |
| 400 | for filename in self.get_filenames(internal=True): |
| 401 | if self.data_re.match(filename): |
| 402 | data_filename = filename |
| 403 | break |
| 404 | else: |
| 405 | raise ExtractorError(".deb contains no data.tar file") |
| 406 | self.archive.seek(0, 0) |
| 407 | self.pipes.pop() |
| 408 | # self.pipes = start_pipes |
| 409 | encoding = mimetypes.guess_type(data_filename)[1] |
| 410 | if not encoding: |
| 411 | raise ExtractorError("data.tar file has unrecognized encoding") |
| 412 | self.pipe(['ar', 'p', self.filename, data_filename], |
| 413 | "extracting data.tar from .deb") |
| 414 | self.pipe([self.decoders[encoding]], "decoding data.tar") |
| 415 | |
| 416 | def basename(self): |
| 417 | pieces = os.path.basename(self.filename).split('_') |
| 418 | if len(pieces) == 1: |
| 419 | return pieces[0] |
| 420 | last_piece = pieces.pop() |
| 421 | if (len(last_piece) > 10) or (not last_piece.endswith('.deb')): |
| 422 | return BaseExtractor.basename(self) |
| 423 | return '_'.join(pieces) |
| 424 | |
| 425 | def check_contents(self): |
| 426 | self.check_included_archives() |
| 427 | self.content_type = BOMB |
| 428 | |
| 429 | |
| 430 | class DebMetadataExtractor(DebExtractor): |
| 431 | def prepare(self): |
| 432 | self.pipe(['ar', 'p', self.filename, 'control.tar.gz'], |
| 433 | "control.tar.gz extraction") |
| 434 | self.pipe(['zcat'], "control.tar.gz decompression") |
| 435 | |
| 436 | |
| 437 | class GemExtractor(TarExtractor): |
| 438 | file_type = 'Ruby gem' |
| 439 | |
| 440 | def prepare(self): |
| 441 | self.pipe(['tar', '-xO', 'data.tar.gz'], "data.tar.gz extraction") |
| 442 | self.pipe(['zcat'], "data.tar.gz decompression") |
| 443 | |
| 444 | def check_contents(self): |
| 445 | self.check_included_archives() |
| 446 | self.content_type = BOMB |
| 447 | |
| 448 | |
| 449 | class GemMetadataExtractor(CompressionExtractor): |
| 450 | file_type = 'Ruby gem' |
| 451 | |
| 452 | def prepare(self): |
| 453 | self.pipe(['tar', '-xO', 'metadata.gz'], "metadata.gz extraction") |
| 454 | self.pipe(['zcat'], "metadata.gz decompression") |
| 455 | |
| 456 | def basename(self): |
| 457 | return os.path.basename(self.filename) + '-metadata.txt' |
| 458 | |
| 459 | |
| 460 | class NoPipeExtractor(BaseExtractor): |
| 461 | # Some extraction tools won't accept the archive from stdin. With |
| 462 | # these, the piping infrastructure we normally set up generally doesn't |
| 463 | # work, at least at first. We can still use most of it; we just don't |
| 464 | # want to seed self.archive with the archive file, since that sucks up |
| 465 | # memory. So instead we seed it with /dev/null, and specify the |
| 466 | # filename on the command line as necessary. We also open the actual |
| 467 | # file with os.open, to make sure we can actually do it (permissions |
| 468 | # are good, etc.). This class doesn't do anything by itself; it's just |
| 469 | # meant to be a base class for extractors that rely on these dumb |
| 470 | # tools. |
| 471 | def __init__(self, filename, encoding): |
| 472 | os.close(os.open(filename, os.O_RDONLY)) |
| 473 | BaseExtractor.__init__(self, '/dev/null', None) |
| 474 | self.filename = os.path.realpath(filename) |
| 475 | |
| 476 | def extract_archive(self): |
| 477 | self.extract_pipe = self.extract_command + [self.filename] |
| 478 | BaseExtractor.extract_archive(self) |
| 479 | |
| 480 | def get_filenames(self): |
| 481 | self.list_pipe = self.list_command + [self.filename] |
| 482 | return BaseExtractor.get_filenames(self) |
| 483 | |
| 484 | |
| 485 | class ZipExtractor(NoPipeExtractor): |
| 486 | file_type = 'Zip file' |
| 487 | extract_command = ['unzip', '-q'] |
| 488 | list_command = ['zipinfo', '-1'] |
| 489 | |
| 490 | |
| 491 | class SevenExtractor(NoPipeExtractor): |
| 492 | file_type = '7z file' |
| 493 | extract_command = ['7z', 'x'] |
| 494 | list_command = ['7z', 'l'] |
| 495 | border_re = re.compile('^[- ]+$') |
| 496 | |
| 497 | def get_filenames(self): |
| 498 | fn_index = None |
| 499 | for line in NoPipeExtractor.get_filenames(self): |
| 500 | if self.border_re.match(line): |
| 501 | if fn_index is not None: |
| 502 | break |
| 503 | else: |
| 504 | fn_index = string.rindex(line, ' ') + 1 |
| 505 | elif fn_index is not None: |
| 506 | yield line[fn_index:] |
| 507 | self.archive.close() |
| 508 | |
| 509 | |
| 510 | class CABExtractor(NoPipeExtractor): |
| 511 | file_type = 'CAB archive' |
| 512 | extract_command = ['cabextract', '-q'] |
| 513 | list_command = ['cabextract', '-l'] |
| 514 | border_re = re.compile(r'^[-\+]+$') |
| 515 | |
| 516 | def get_filenames(self): |
| 517 | fn_index = None |
| 518 | filenames = NoPipeExtractor.get_filenames(self) |
| 519 | for line in filenames: |
| 520 | if self.border_re.match(line): |
| 521 | break |
| 522 | for line in filenames: |
| 523 | try: |
| 524 | yield line.split(' | ', 2)[2] |
| 525 | except IndexError: |
| 526 | break |
| 527 | self.archive.close() |
| 528 | |
| 529 | |
| 530 | class ShieldExtractor(NoPipeExtractor): |
| 531 | file_type = 'InstallShield archive' |
| 532 | extract_command = ['unshield', 'x'] |
| 533 | list_command = ['unshield', 'l'] |
| 534 | prefix_re = re.compile(r'^\s+\d+\s+') |
| 535 | end_re = re.compile(r'^\s+-+\s+-+\s*$') |
| 536 | |
| 537 | def get_filenames(self): |
| 538 | for line in NoPipeExtractor.get_filenames(self): |
| 539 | if self.end_re.match(line): |
| 540 | break |
| 541 | else: |
| 542 | match = self.prefix_re.match(line) |
| 543 | if match: |
| 544 | yield line[match.end():] |
| 545 | self.archive.close() |
| 546 | |
| 547 | def basename(self): |
| 548 | result = NoPipeExtractor.basename(self) |
| 549 | if result.endswith('.hdr'): |
| 550 | result = result[:-4] |
| 551 | return result |
| 552 | |
| 553 | |
| 554 | class RarExtractor(NoPipeExtractor): |
| 555 | file_type = 'RAR archive' |
| 556 | extract_command = ['unrar', 'x'] |
| 557 | list_command = ['unrar', 'l'] |
| 558 | border_re = re.compile('^-+$') |
| 559 | |
| 560 | def get_filenames(self): |
| 561 | inside = False |
| 562 | for line in NoPipeExtractor.get_filenames(self): |
| 563 | if self.border_re.match(line): |
| 564 | if inside: |
| 565 | break |
| 566 | else: |
| 567 | inside = True |
| 568 | elif inside: |
| 569 | yield line.split(' ')[1] |
| 570 | self.archive.close() |
| 571 | |
| 572 | |
| 573 | class BaseHandler(object): |
| 574 | def __init__(self, extractor, options): |
| 575 | self.extractor = extractor |
| 576 | self.options = options |
| 577 | self.target = None |
| 578 | |
| 579 | def handle(self): |
| 580 | command = 'find' |
| 581 | status = subprocess.call(['find', self.extractor.target, '-type', 'd', |
| 582 | '-exec', 'chmod', 'u+rwx', '{}', ';']) |
| 583 | if status == 0: |
| 584 | command = 'chmod' |
| 585 | status = subprocess.call(['chmod', '-R', 'u+rwX', |
| 586 | self.extractor.target]) |
| 587 | if status != 0: |
| 588 | return "%s returned with exit status %s" % (command, status) |
| 589 | return self.organize() |
| 590 | |
| 591 | def set_target(self, target, checker): |
| 592 | self.target = checker(target).check() |
| 593 | if self.target != target: |
| 594 | logger.warning("extracting %s to %s" % |
| 595 | (self.extractor.filename, self.target)) |
| 596 | |
| 597 | |
| 598 | # The "where to extract" table, with options and archive types. |
| 599 | # This dictates the contents of each can_handle method. |
| 600 | # |
| 601 | # Flat Overwrite None |
| 602 | # File basename basename FilenameChecked |
| 603 | # Match . . tempdir + checked |
| 604 | # Bomb . basename DirectoryChecked |
| 605 | |
| 606 | class FlatHandler(BaseHandler): |
| 607 | def can_handle(contents, options): |
| 608 | return ((options.flat and (contents != ONE_ENTRY_KNOWN)) or |
| 609 | (options.overwrite and (contents == MATCHING_DIRECTORY))) |
| 610 | can_handle = staticmethod(can_handle) |
| 611 | |
| 612 | def organize(self): |
| 613 | self.target = '.' |
| 614 | for curdir, dirs, filenames in os.walk(self.extractor.target, |
| 615 | topdown=False): |
| 616 | path_parts = curdir.split(os.sep) |
| 617 | if path_parts[0] == '.': |
| 618 | del path_parts[1] |
| 619 | else: |
| 620 | del path_parts[0] |
| 621 | newdir = os.path.join(*path_parts) |
| 622 | if not os.path.isdir(newdir): |
| 623 | os.makedirs(newdir) |
| 624 | for filename in filenames: |
| 625 | os.rename(os.path.join(curdir, filename), |
| 626 | os.path.join(newdir, filename)) |
| 627 | os.rmdir(curdir) |
| 628 | |
| 629 | |
| 630 | class OverwriteHandler(BaseHandler): |
| 631 | def can_handle(contents, options): |
| 632 | return ((options.flat and (contents == ONE_ENTRY_KNOWN)) or |
| 633 | (options.overwrite and (contents != MATCHING_DIRECTORY))) |
| 634 | can_handle = staticmethod(can_handle) |
| 635 | |
| 636 | def organize(self): |
| 637 | self.target = self.extractor.basename() |
| 638 | if os.path.isdir(self.target): |
| 639 | shutil.rmtree(self.target) |
| 640 | os.rename(self.extractor.target, self.target) |
| 641 | |
| 642 | |
| 643 | class MatchHandler(BaseHandler): |
| 644 | def can_handle(contents, options): |
| 645 | return ((contents == MATCHING_DIRECTORY) or |
| 646 | ((contents in ONE_ENTRY_UNKNOWN) and |
| 647 | options.one_entry_policy.ok_for_match())) |
| 648 | can_handle = staticmethod(can_handle) |
| 649 | |
| 650 | def organize(self): |
| 651 | source = os.path.join(self.extractor.target, |
| 652 | os.listdir(self.extractor.target)[0]) |
| 653 | if os.path.isdir(source): |
| 654 | checker = DirectoryChecker |
| 655 | else: |
| 656 | checker = FilenameChecker |
| 657 | if self.options.one_entry_policy == EXTRACT_HERE: |
| 658 | destination = self.extractor.content_name.rstrip('/') |
| 659 | else: |
| 660 | destination = self.extractor.basename() |
| 661 | self.set_target(destination, checker) |
| 662 | if os.path.isdir(self.extractor.target): |
| 663 | os.rename(source, self.target) |
| 664 | os.rmdir(self.extractor.target) |
| 665 | else: |
| 666 | os.rename(self.extractor.target, self.target) |
| 667 | self.extractor.included_root = './' |
| 668 | |
| 669 | |
| 670 | class EmptyHandler(object): |
| 671 | target = '' |
| 672 | |
| 673 | def can_handle(contents, options): |
| 674 | return contents == EMPTY |
| 675 | can_handle = staticmethod(can_handle) |
| 676 | |
| 677 | def __init__(self, extractor, options): pass |
| 678 | def handle(self): pass |
| 679 | |
| 680 | |
| 681 | class BombHandler(BaseHandler): |
| 682 | def can_handle(contents, options): |
| 683 | return True |
| 684 | can_handle = staticmethod(can_handle) |
| 685 | |
| 686 | def organize(self): |
| 687 | basename = self.extractor.basename() |
| 688 | self.set_target(basename, self.extractor.name_checker) |
| 689 | os.rename(self.extractor.target, self.target) |
| 690 | |
| 691 | |
| 692 | class BasePolicy(object): |
| 693 | try: |
| 694 | size = fcntl.ioctl(sys.stdout.fileno(), termios.TIOCGWINSZ, |
| 695 | struct.pack("HHHH", 0, 0, 0, 0)) |
| 696 | width = struct.unpack("HHHH", size)[1] |
| 697 | except IOError: |
| 698 | width = 80 |
| 699 | width = width - 1 |
| 700 | choice_wrapper = textwrap.TextWrapper(width=width, initial_indent=' * ', |
| 701 | subsequent_indent=' ', |
| 702 | break_long_words=False) |
| 703 | |
| 704 | def __init__(self, options): |
| 705 | self.current_policy = None |
| 706 | if options.batch: |
| 707 | self.permanent_policy = self.answers[''] |
| 708 | else: |
| 709 | self.permanent_policy = None |
| 710 | |
| 711 | def ask_question(self, question): |
| 712 | question = question + ["You can:"] |
| 713 | for choice in self.choices: |
| 714 | question.extend(self.choice_wrapper.wrap(choice)) |
| 715 | while True: |
| 716 | print "\n".join(question) |
| 717 | try: |
| 718 | answer = raw_input(self.prompt) |
| 719 | except EOFError: |
| 720 | return self.answers[''] |
| 721 | try: |
| 722 | return self.answers[answer.lower()] |
| 723 | except KeyError: |
| 724 | print |
| 725 | |
| 726 | def wrap(self, question, *args): |
| 727 | words = question.split() |
| 728 | for arg in args: |
| 729 | words[words.index('%s')] = arg |
| 730 | result = [words.pop(0)] |
| 731 | for word in words: |
| 732 | extend = '%s %s' % (result[-1], word) |
| 733 | if len(extend) > self.width: |
| 734 | result.append(word) |
| 735 | else: |
| 736 | result[-1] = extend |
| 737 | return result |
| 738 | |
| 739 | def __cmp__(self, other): |
| 740 | return cmp(self.current_policy, other) |
| 741 | |
| 742 | |
| 743 | class OneEntryPolicy(BasePolicy): |
| 744 | answers = {'h': EXTRACT_HERE, 'i': EXTRACT_WRAP, 'r': EXTRACT_RENAME, |
| 745 | '': EXTRACT_WRAP} |
| 746 | choice_template = ["extract the %s _I_nside a new directory named %s", |
| 747 | "extract the %s and _R_ename it %s", |
| 748 | "extract the %s _H_ere"] |
| 749 | prompt = "What do you want to do? (I/r/h) " |
| 750 | |
| 751 | def __init__(self, options): |
| 752 | BasePolicy.__init__(self, options) |
| 753 | if options.flat: |
| 754 | default = 'h' |
| 755 | elif options.one_entry_default is not None: |
| 756 | default = options.one_entry_default.lower() |
| 757 | else: |
| 758 | return |
| 759 | if 'here'.startswith(default): |
| 760 | self.permanent_policy = EXTRACT_HERE |
| 761 | elif 'rename'.startswith(default): |
| 762 | self.permanent_policy = EXTRACT_RENAME |
| 763 | elif 'inside'.startswith(default): |
| 764 | self.permanent_policy = EXTRACT_WRAP |
| 765 | elif default is not None: |
| 766 | raise ValueError("bad value %s for default policy" % (default,)) |
| 767 | |
| 768 | def prep(self, archive_filename, extractor): |
| 769 | question = self.wrap( |
| 770 | "%s contains one %s but its name doesn't match.", |
| 771 | archive_filename, extractor.content_type) |
| 772 | question.append(" Expected: " + extractor.basename()) |
| 773 | question.append(" Actual: " + extractor.content_name) |
| 774 | choice_vars = (extractor.content_type, extractor.basename()) |
| 775 | self.choices = [text % choice_vars[:text.count('%s')] |
| 776 | for text in self.choice_template] |
| 777 | self.current_policy = (self.permanent_policy or |
| 778 | self.ask_question(question)) |
| 779 | |
| 780 | def ok_for_match(self): |
| 781 | return self.current_policy in (EXTRACT_RENAME, EXTRACT_HERE) |
| 782 | |
| 783 | |
| 784 | class RecursionPolicy(BasePolicy): |
| 785 | answers = {'o': RECURSE_ONCE, 'a': RECURSE_ALWAYS, 'n': RECURSE_NOT_NOW, |
| 786 | 'v': RECURSE_NEVER, 'l': RECURSE_LIST, '': RECURSE_NOT_NOW} |
| 787 | choices = ["_A_lways extract included archives during this session", |
| 788 | "extract included archives this _O_nce", |
| 789 | "choose _N_ot to extract included archives this once", |
| 790 | "ne_V_er extract included archives during this session", |
| 791 | "_L_ist included archives"] |
| 792 | prompt = "What do you want to do? (a/o/N/v/l) " |
| 793 | |
| 794 | def __init__(self, options): |
| 795 | BasePolicy.__init__(self, options) |
| 796 | if options.show_list: |
| 797 | self.permanent_policy = RECURSE_NEVER |
| 798 | elif options.recursive: |
| 799 | self.permanent_policy = RECURSE_ALWAYS |
| 800 | |
| 801 | def prep(self, current_filename, target, extractor): |
| 802 | archive_count = len(extractor.included_archives) |
| 803 | if (self.permanent_policy is not None) or (archive_count == 0): |
| 804 | self.current_policy = self.permanent_policy or RECURSE_NOT_NOW |
| 805 | return |
| 806 | question = self.wrap( |
| 807 | "%s contains %s other archive file(s), out of %s file(s) total.", |
| 808 | current_filename, archive_count, extractor.file_count) |
| 809 | if target == '.': |
| 810 | target = '' |
| 811 | included_root = extractor.included_root |
| 812 | if included_root == './': |
| 813 | included_root = '' |
| 814 | while True: |
| 815 | self.current_policy = self.ask_question(question) |
| 816 | if self.current_policy != RECURSE_LIST: |
| 817 | break |
| 818 | print ("\n%s\n" % |
| 819 | '\n'.join([os.path.join(target, included_root, filename) |
| 820 | for filename in extractor.included_archives])) |
| 821 | if self.current_policy in (RECURSE_ALWAYS, RECURSE_NEVER): |
| 822 | self.permanent_policy = self.current_policy |
| 823 | |
| 824 | def ok_to_recurse(self): |
| 825 | return self.current_policy in (RECURSE_ALWAYS, RECURSE_ONCE) |
| 826 | |
| 827 | |
| 828 | class ExtractorBuilder(object): |
| 829 | extractor_map = {'tar': {'extractor': TarExtractor, |
| 830 | 'mimetypes': ('x-tar',), |
| 831 | 'extensions': ('tar',), |
| 832 | 'magic': ('POSIX tar archive',)}, |
| 833 | 'zip': {'extractor': ZipExtractor, |
| 834 | 'mimetypes': ('zip',), |
| 835 | 'extensions': ('zip',), |
| 836 | 'magic': ('(Zip|ZIP self-extracting) archive',)}, |
| 837 | 'rpm': {'extractor': RPMExtractor, |
| 838 | 'mimetypes': ('x-redhat-package-manager', 'x-rpm'), |
| 839 | 'extensions': ('rpm',), |
| 840 | 'magic': ('RPM',)}, |
| 841 | 'deb': {'extractor': DebExtractor, |
| 842 | 'metadata': DebMetadataExtractor, |
| 843 | 'mimetypes': ('x-debian-package',), |
| 844 | 'extensions': ('deb',), |
| 845 | 'magic': ('Debian binary package',)}, |
| 846 | 'cpio': {'extractor': CpioExtractor, |
| 847 | 'mimetypes': ('x-cpio',), |
| 848 | 'extensions': ('cpio',), |
| 849 | 'magic': ('cpio archive',)}, |
| 850 | 'gem': {'extractor': GemExtractor, |
| 851 | 'metadata': GemMetadataExtractor, |
| 852 | 'mimetypes': ('x-ruby-gem',), |
| 853 | 'extensions': ('gem',)}, |
| 854 | '7z': {'extractor': SevenExtractor, |
| 855 | 'mimetypes': ('x-7z-compressed',), |
| 856 | 'extensions': ('7z',), |
| 857 | 'magic': ('7-zip archive',)}, |
| 858 | 'cab': {'extractor': CABExtractor, |
| 859 | 'mimetypes': ('x-cab',), |
| 860 | 'extensions': ('cab',), |
| 861 | 'magic': ('Microsoft Cabinet Archive',)}, |
| 862 | 'rar': {'extractor': RarExtractor, |
| 863 | 'mimetypes': ('rar',), |
| 864 | 'extensions': ('rar',), |
| 865 | 'magic': ('RAR archive',)}, |
| 866 | 'shield': {'extractor': ShieldExtractor, |
| 867 | 'mimetypes': ('x-cab',), |
| 868 | 'extensions': ('cab', 'hdr'), |
| 869 | 'magic': ('InstallShield CAB',)}, |
| 870 | 'compress': {'extractor': CompressionExtractor} |
| 871 | } |
| 872 | |
| 873 | mimetype_map = {} |
| 874 | magic_mime_map = {} |
| 875 | extension_map = {} |
| 876 | for ext_name, ext_info in extractor_map.items(): |
| 877 | for mimetype in ext_info.get('mimetypes', ()): |
| 878 | if '/' not in mimetype: |
| 879 | mimetype = 'application/' + mimetype |
| 880 | mimetype_map[mimetype] = ext_name |
| 881 | for magic_re in ext_info.get('magic', ()): |
| 882 | magic_mime_map[re.compile(magic_re)] = ext_name |
| 883 | for extension in ext_info.get('extensions', ()): |
| 884 | extension_map.setdefault(extension, []).append((ext_name, None)) |
| 885 | |
| 886 | for mapping in (('tar', 'bzip2', 'tar.bz2', 'tbz2', 'tb2', 'tbz'), |
| 887 | ('tar', 'gzip', 'tar.gz', 'tgz'), |
| 888 | ('tar', 'lzma', 'tar.lzma', 'tlz'), |
| 889 | ('tar', 'xz', 'tar.xz'), |
| 890 | ('tar', 'compress', 'tar.Z', 'taz'), |
| 891 | ('compress', 'gzip', 'Z', 'gz'), |
| 892 | ('compress', 'bzip2', 'bz2'), |
| 893 | ('compress', 'lzma', 'lzma'), |
| 894 | ('compress', 'xz', 'xz')): |
| 895 | for extension in mapping[2:]: |
| 896 | extension_map.setdefault(extension, []).append(mapping[:2]) |
| 897 | |
| 898 | magic_encoding_map = {} |
| 899 | for mapping in (('bzip2', 'bzip2 compressed'), |
| 900 | ('gzip', 'gzip compressed'), |
| 901 | ('lzma', 'LZMA compressed'), |
| 902 | ('xz', 'xz compressed')): |
| 903 | for pattern in mapping[1:]: |
| 904 | magic_encoding_map[re.compile(pattern)] = mapping[0] |
| 905 | |
| 906 | def __init__(self, filename, options): |
| 907 | self.filename = filename |
| 908 | self.options = options |
| 909 | |
| 910 | def build_extractor(self, archive_type, encoding): |
| 911 | extractors = self.extractor_map[archive_type] |
| 912 | if self.options.metadata and extractors.has_key('metadata'): |
| 913 | extractor = extractors['metadata'] |
| 914 | else: |
| 915 | extractor = extractors['extractor'] |
| 916 | return extractor(self.filename, encoding) |
| 917 | |
| 918 | def get_extractor(self): |
| 919 | tried_types = set() |
| 920 | # As smart as it is, the magic test can't go first, because at least |
| 921 | # on my system it just recognizes gem files as tar files. I guess |
| 922 | # it's possible for the opposite problem to occur -- where the mimetype |
| 923 | # or extension suggests something less than ideal -- but it seems less |
| 924 | # likely so I'm sticking with this. |
| 925 | for func_name in ('mimetype', 'extension', 'magic'): |
| 926 | logger.debug("getting extractors by %s" % (func_name,)) |
| 927 | extractor_types = \ |
| 928 | getattr(self, 'try_by_' + func_name)(self.filename) |
| 929 | logger.debug("done getting extractors") |
| 930 | for ext_args in extractor_types: |
| 931 | if ext_args in tried_types: |
| 932 | continue |
| 933 | tried_types.add(ext_args) |
| 934 | logger.debug("trying %s extractor from %s" % |
| 935 | (ext_args, func_name)) |
| 936 | yield self.build_extractor(*ext_args) |
| 937 | |
| 938 | def try_by_mimetype(cls, filename): |
| 939 | mimetype, encoding = mimetypes.guess_type(filename) |
| 940 | try: |
| 941 | return [(cls.mimetype_map[mimetype], encoding)] |
| 942 | except KeyError: |
| 943 | if encoding: |
| 944 | return [('compress', encoding)] |
| 945 | return [] |
| 946 | try_by_mimetype = classmethod(try_by_mimetype) |
| 947 | |
| 948 | def magic_map_matches(cls, output, magic_map): |
| 949 | return [result for regexp, result in magic_map.items() |
| 950 | if regexp.search(output)] |
| 951 | magic_map_matches = classmethod(magic_map_matches) |
| 952 | |
| 953 | def try_by_magic(cls, filename): |
| 954 | process = subprocess.Popen(['file', '-z', filename], |
| 955 | stdout=subprocess.PIPE) |
| 956 | status = process.wait() |
| 957 | if status != 0: |
| 958 | return [] |
| 959 | output = process.stdout.readline() |
| 960 | process.stdout.close() |
| 961 | if output.startswith('%s: ' % filename): |
| 962 | output = output[len(filename) + 2:] |
| 963 | mimes = cls.magic_map_matches(output, cls.magic_mime_map) |
| 964 | encodings = cls.magic_map_matches(output, cls.magic_encoding_map) |
| 965 | if mimes and not encodings: |
| 966 | encodings = [None] |
| 967 | elif encodings and not mimes: |
| 968 | mimes = ['compress'] |
| 969 | return [(m, e) for m in mimes for e in encodings] |
| 970 | try_by_magic = classmethod(try_by_magic) |
| 971 | |
| 972 | def try_by_extension(cls, filename): |
| 973 | parts = filename.split('.')[-2:] |
| 974 | results = [] |
| 975 | while parts: |
| 976 | results.extend(cls.extension_map.get('.'.join(parts), [])) |
| 977 | del parts[0] |
| 978 | return results |
| 979 | try_by_extension = classmethod(try_by_extension) |
| 980 | |
| 981 | |
| 982 | class BaseAction(object): |
| 983 | def __init__(self, options, filenames): |
| 984 | self.options = options |
| 985 | self.filenames = filenames |
| 986 | self.target = None |
| 987 | self.do_print = False |
| 988 | |
| 989 | def report(self, function, *args): |
| 990 | try: |
| 991 | error = function(*args) |
| 992 | except EXTRACTION_ERRORS, exception: |
| 993 | error = str(exception) |
| 994 | logger.debug(''.join(traceback.format_exception(*sys.exc_info()))) |
| 995 | return error |
| 996 | |
| 997 | def show_filename(self, filename): |
| 998 | if len(self.filenames) < 2: |
| 999 | return |
| 1000 | elif self.do_print: |
| 1001 | print |
| 1002 | else: |
| 1003 | self.do_print = True |
| 1004 | print "%s:" % (filename,) |
| 1005 | |
| 1006 | |
| 1007 | class ExtractionAction(BaseAction): |
| 1008 | handlers = [FlatHandler, OverwriteHandler, MatchHandler, EmptyHandler, |
| 1009 | BombHandler] |
| 1010 | |
| 1011 | def get_handler(self, extractor): |
| 1012 | if extractor.content_type in ONE_ENTRY_UNKNOWN: |
| 1013 | self.options.one_entry_policy.prep(self.current_filename, |
| 1014 | extractor) |
| 1015 | for handler in self.handlers: |
| 1016 | if handler.can_handle(extractor.content_type, self.options): |
| 1017 | logger.debug("using %s handler" % (handler.__name__,)) |
| 1018 | self.current_handler = handler(extractor, self.options) |
| 1019 | break |
| 1020 | |
| 1021 | def show_extraction(self, extractor): |
| 1022 | if self.options.log_level > logging.INFO: |
| 1023 | return |
| 1024 | self.show_filename(self.current_filename) |
| 1025 | if extractor.contents is None: |
| 1026 | print self.current_handler.target |
| 1027 | return |
| 1028 | def reverser(x, y): |
| 1029 | return cmp(y, x) |
| 1030 | if self.current_handler.target == '.': |
| 1031 | filenames = extractor.contents |
| 1032 | filenames.sort(reverser) |
| 1033 | else: |
| 1034 | filenames = [self.current_handler.target] |
| 1035 | pathjoin = os.path.join |
| 1036 | isdir = os.path.isdir |
| 1037 | while filenames: |
| 1038 | filename = filenames.pop() |
| 1039 | if isdir(filename): |
| 1040 | print "%s/" % (filename,) |
| 1041 | new_filenames = os.listdir(filename) |
| 1042 | new_filenames.sort(reverser) |
| 1043 | filenames.extend([pathjoin(filename, new_filename) |
| 1044 | for new_filename in new_filenames]) |
| 1045 | else: |
| 1046 | print filename |
| 1047 | |
| 1048 | def run(self, filename, extractor): |
| 1049 | self.current_filename = filename |
| 1050 | error = (self.report(extractor.extract) or |
| 1051 | self.report(self.get_handler, extractor) or |
| 1052 | self.report(self.current_handler.handle) or |
| 1053 | self.report(self.show_extraction, extractor)) |
| 1054 | if not error: |
| 1055 | self.target = self.current_handler.target |
| 1056 | return error |
| 1057 | |
| 1058 | |
| 1059 | class ListAction(BaseAction): |
| 1060 | def list_filenames(self, extractor, filename): |
| 1061 | # We get a line first to make sure there's not going to be some |
| 1062 | # basic error before we show what filename we're listing. |
| 1063 | filename_lister = extractor.get_filenames() |
| 1064 | try: |
| 1065 | first_line = filename_lister.next() |
| 1066 | except StopIteration: |
| 1067 | self.show_filename(filename) |
| 1068 | else: |
| 1069 | self.did_list = True |
| 1070 | self.show_filename(filename) |
| 1071 | print first_line |
| 1072 | for line in filename_lister: |
| 1073 | print line |
| 1074 | |
| 1075 | def run(self, filename, extractor): |
| 1076 | self.did_list = False |
| 1077 | error = self.report(self.list_filenames, extractor, filename) |
| 1078 | if error and self.did_list: |
| 1079 | logger.error("lister failed: ignore above listing for %s" % |
| 1080 | (filename,)) |
| 1081 | return error |
| 1082 | |
| 1083 | class UrlHandler(urllib.FancyURLopener): |
| 1084 | def http_error_default(self, url, fp, errcode, errmsg, headers): |
| 1085 | urllib.URLopener.http_error_default(self, url, fp, errcode, errmsg, headers) |
| 1086 | |
| 1087 | def is_url(self, url): |
| 1088 | if url.startswith("http://"): |
| 1089 | return True |
| 1090 | |
| 1091 | def fetch(self, url): |
| 1092 | i = url.rfind('/') |
| 1093 | filename = url[i+1:] |
| 1094 | try: |
| 1095 | self.retrieve(url, filename) |
| 1096 | except IOError: |
| 1097 | return False, "Failed to fetch "+url |
| 1098 | return True, filename |
| 1099 | |
| 1100 | |
| 1101 | class ExtractorApplication(object): |
| 1102 | def __init__(self, arguments): |
| 1103 | for signal_num in (signal.SIGINT, signal.SIGTERM): |
| 1104 | signal.signal(signal_num, self.abort) |
| 1105 | signal.signal(signal.SIGPIPE, signal.SIG_DFL) |
| 1106 | self.parse_options(arguments) |
| 1107 | self.setup_logger() |
| 1108 | self.successes = [] |
| 1109 | self.failures = [] |
| 1110 | |
| 1111 | def clean_destination(self, dest_name): |
| 1112 | try: |
| 1113 | os.unlink(dest_name) |
| 1114 | except OSError, error: |
| 1115 | if error.errno == errno.EISDIR: |
| 1116 | shutil.rmtree(dest_name, ignore_errors=True) |
| 1117 | |
| 1118 | def abort(self, signal_num, frame): |
| 1119 | signal.signal(signal_num, signal.SIG_IGN) |
| 1120 | print |
| 1121 | logger.debug("traceback:\n" + |
| 1122 | ''.join(traceback.format_stack(frame)).rstrip()) |
| 1123 | logger.debug("got signal %s" % (signal_num,)) |
| 1124 | try: |
| 1125 | basename = self.current_extractor.target |
| 1126 | except AttributeError: |
| 1127 | basename = None |
| 1128 | if basename is not None: |
| 1129 | logger.debug("cleaning up %s" % (basename,)) |
| 1130 | clean_targets = set([os.path.realpath('.')]) |
| 1131 | if hasattr(self, 'current_directory'): |
| 1132 | clean_targets.add(os.path.realpath(self.current_directory)) |
| 1133 | for directory in clean_targets: |
| 1134 | self.clean_destination(os.path.join(directory, basename)) |
| 1135 | sys.exit(1) |
| 1136 | |
| 1137 | def parse_options(self, arguments): |
| 1138 | parser = optparse.OptionParser( |
| 1139 | usage="%prog [options] archive [archive2 ...]", |
| 1140 | description="Intelligent archive extractor", |
| 1141 | version=VERSION_BANNER |
| 1142 | ) |
| 1143 | parser.add_option('-l', '-t', '--list', '--table', dest='show_list', |
| 1144 | action='store_true', default=False, |
| 1145 | help="list contents of archives on standard output") |
| 1146 | parser.add_option('-m', '--metadata', dest='metadata', |
| 1147 | action='store_true', default=False, |
| 1148 | help="extract metadata from a .deb/.gem") |
| 1149 | parser.add_option('-r', '--recursive', dest='recursive', |
| 1150 | action='store_true', default=False, |
| 1151 | help="extract archives contained in the ones listed") |
| 1152 | parser.add_option('--one', '--one-entry', dest='one_entry_default', |
| 1153 | default=None, |
| 1154 | help=("specify extraction policy for one-entry " + |
| 1155 | "archives: inside/rename/here")) |
| 1156 | parser.add_option('-n', '--noninteractive', dest='batch', |
| 1157 | action='store_true', default=False, |
| 1158 | help="don't ask how to handle special cases") |
| 1159 | parser.add_option('-o', '--overwrite', dest='overwrite', |
| 1160 | action='store_true', default=False, |
| 1161 | help="overwrite any existing target output") |
| 1162 | parser.add_option('-f', '--flat', '--no-directory', dest='flat', |
| 1163 | action='store_true', default=False, |
| 1164 | help="extract everything to the current directory") |
| 1165 | parser.add_option('-v', '--verbose', dest='verbose', |
| 1166 | action='count', default=0, |
| 1167 | help="be verbose/print debugging information") |
| 1168 | parser.add_option('-q', '--quiet', dest='quiet', |
| 1169 | action='count', default=3, |
| 1170 | help="suppress warning/error messages") |
| 1171 | self.options, filenames = parser.parse_args(arguments) |
| 1172 | if not filenames: |
| 1173 | parser.error("you did not list any archives") |
| 1174 | # This makes WARNING is the default. |
| 1175 | self.options.log_level = (10 * (self.options.quiet - |
| 1176 | self.options.verbose)) |
| 1177 | try: |
| 1178 | self.options.one_entry_policy = OneEntryPolicy(self.options) |
| 1179 | except ValueError: |
| 1180 | parser.error("invalid value for --one-entry option") |
| 1181 | self.options.recursion_policy = RecursionPolicy(self.options) |
| 1182 | self.archives = {os.path.realpath(os.curdir): filenames} |
| 1183 | |
| 1184 | def setup_logger(self): |
| 1185 | logging.getLogger().setLevel(self.options.log_level) |
| 1186 | handler = logging.StreamHandler() |
| 1187 | handler.setLevel(self.options.log_level) |
| 1188 | formatter = logging.Formatter("dtrx: %(levelname)s: %(message)s") |
| 1189 | handler.setFormatter(formatter) |
| 1190 | logger.addHandler(handler) |
| 1191 | logger.debug("logger is set up") |
| 1192 | |
| 1193 | def recurse(self, filename, extractor, action): |
| 1194 | self.options.recursion_policy.prep(filename, action.target, extractor) |
| 1195 | if self.options.recursion_policy.ok_to_recurse(): |
| 1196 | for filename in extractor.included_archives: |
| 1197 | logger.debug("recursing with %s archive" % |
| 1198 | (extractor.content_type,)) |
| 1199 | tail_path, basename = os.path.split(filename) |
| 1200 | path_args = [self.current_directory, extractor.included_root, |
| 1201 | tail_path] |
| 1202 | logger.debug("included root: %s" % (extractor.included_root,)) |
| 1203 | logger.debug("tail path: %s" % (tail_path,)) |
| 1204 | if os.path.isdir(action.target): |
| 1205 | logger.debug("action target: %s" % (action.target,)) |
| 1206 | path_args.insert(1, action.target) |
| 1207 | directory = os.path.join(*path_args) |
| 1208 | self.archives.setdefault(directory, []).append(basename) |
| 1209 | |
| 1210 | def check_file(self, filename): |
| 1211 | try: |
| 1212 | result = os.stat(filename) |
| 1213 | except OSError, error: |
| 1214 | return error.strerror |
| 1215 | if stat.S_ISDIR(result.st_mode): |
| 1216 | return "cannot work with a directory" |
| 1217 | |
| 1218 | def show_stderr(self, logger_func, stderr): |
| 1219 | if stderr: |
| 1220 | logger_func("Error output from this process:\n" + |
| 1221 | stderr.rstrip('\n')) |
| 1222 | |
| 1223 | def try_extractors(self, filename, builder): |
| 1224 | errors = [] |
| 1225 | for extractor in builder: |
| 1226 | self.current_extractor = extractor # For the abort() method. |
| 1227 | error = self.action.run(filename, extractor) |
| 1228 | if error: |
| 1229 | errors.append((extractor.file_type, extractor.encoding, error, |
| 1230 | extractor.get_stderr())) |
| 1231 | if extractor.target is not None: |
| 1232 | self.clean_destination(extractor.target) |
| 1233 | else: |
| 1234 | self.show_stderr(logger.warn, extractor.get_stderr()) |
| 1235 | self.recurse(filename, extractor, self.action) |
| 1236 | return |
| 1237 | logger.error("could not handle %s" % (filename,)) |
| 1238 | if not errors: |
| 1239 | logger.error("not a known archive type") |
| 1240 | return True |
| 1241 | for file_type, encoding, error, stderr in errors: |
| 1242 | message = ["treating as", file_type, "failed:", error] |
| 1243 | if encoding: |
| 1244 | message.insert(1, "%s-encoded" % (encoding,)) |
| 1245 | logger.error(' '.join(message)) |
| 1246 | self.show_stderr(logger.error, stderr) |
| 1247 | return True |
| 1248 | |
| 1249 | def run(self): |
| 1250 | urlhandler = UrlHandler(); |
| 1251 | if self.options.show_list: |
| 1252 | action = ListAction |
| 1253 | else: |
| 1254 | action = ExtractionAction |
| 1255 | self.action = action(self.options, self.archives.values()[0]) |
| 1256 | while self.archives: |
| 1257 | self.current_directory, self.filenames = self.archives.popitem() |
| 1258 | os.chdir(self.current_directory) |
| 1259 | for filename in self.filenames: |
| 1260 | if urlhandler.is_url(filename): |
| 1261 | error, filename = urlhandler.fetch(filename) |
| 1262 | builder = ExtractorBuilder(filename, self.options) |
| 1263 | error = (self.check_file(filename) or |
| 1264 | self.try_extractors(filename, builder.get_extractor())) |
| 1265 | if error: |
| 1266 | if error != True: |
| 1267 | logger.error("%s: %s" % (filename, error)) |
| 1268 | self.failures.append(filename) |
| 1269 | else: |
| 1270 | self.successes.append(filename) |
| 1271 | self.options.one_entry_policy.permanent_policy = EXTRACT_WRAP |
| 1272 | if self.failures: |
| 1273 | return 1 |
| 1274 | return 0 |
| 1275 | |
| 1276 | |
| 1277 | if __name__ == '__main__': |
| 1278 | app = ExtractorApplication(sys.argv[1:]) |
| 1279 | sys.exit(app.run()) |