Mercurial > ~astiob > upreckon > hgweb
annotate 2.00/zipfile31.py @ 29:a8cc383b787c
Clean up zipfiles and diff them to stock ones
author | Oleg Oshmyan <chortos@inbox.lv> |
---|---|
date | Wed, 24 Nov 2010 23:21:31 +0000 |
parents | 2.00/zipfile3.py@3d535503161f |
children |
rev | line source |
---|---|
21 | 1 """ |
2 Read and write ZIP files. | |
3 | |
4 XXX references to utf-8 need further investigation. | |
5 """ | |
6 # Improved by Chortos-2 in 2010 (added bzip2 support) | |
7 import struct, os, time, sys, shutil | |
8 import binascii, io, stat | |
9 | |
10 try: | |
11 import zlib # We may need its compression method | |
12 crc32 = zlib.crc32 | |
13 except ImportError: | |
14 zlib = None | |
15 crc32 = binascii.crc32 | |
16 | |
17 try: | |
18 import bz2 # We may need its compression method | |
19 except ImportError: | |
20 bz2 = None | |
21 | |
22 __all__ = ["BadZipfile", "error", "ZIP_STORED", "ZIP_DEFLATED", "is_zipfile", | |
23 "ZipInfo", "ZipFile", "PyZipFile", "LargeZipFile", "ZIP_BZIP2" ] | |
24 | |
25 class BadZipfile(Exception): | |
26 pass | |
27 | |
28 | |
29 class LargeZipFile(Exception): | |
30 """ | |
31 Raised when writing a zipfile, the zipfile requires ZIP64 extensions | |
32 and those extensions are disabled. | |
33 """ | |
34 | |
35 error = BadZipfile # The exception raised by this module | |
36 | |
37 ZIP64_LIMIT = (1 << 31) - 1 | |
38 ZIP_FILECOUNT_LIMIT = 1 << 16 | |
39 ZIP_MAX_COMMENT = (1 << 16) - 1 | |
40 | |
41 # constants for Zip file compression methods | |
42 ZIP_STORED = 0 | |
43 ZIP_DEFLATED = 8 | |
44 ZIP_BZIP2 = 12 | |
45 # Other ZIP compression methods not supported | |
46 | |
47 # Below are some formats and associated data for reading/writing headers using | |
48 # the struct module. The names and structures of headers/records are those used | |
49 # in the PKWARE description of the ZIP file format: | |
50 # http://www.pkware.com/documents/casestudies/APPNOTE.TXT | |
51 # (URL valid as of January 2008) | |
52 | |
53 # The "end of central directory" structure, magic number, size, and indices | |
54 # (section V.I in the format document) | |
55 structEndArchive = b"<4s4H2LH" | |
56 stringEndArchive = b"PK\005\006" | |
57 sizeEndCentDir = struct.calcsize(structEndArchive) | |
58 | |
59 _ECD_SIGNATURE = 0 | |
60 _ECD_DISK_NUMBER = 1 | |
61 _ECD_DISK_START = 2 | |
62 _ECD_ENTRIES_THIS_DISK = 3 | |
63 _ECD_ENTRIES_TOTAL = 4 | |
64 _ECD_SIZE = 5 | |
65 _ECD_OFFSET = 6 | |
66 _ECD_COMMENT_SIZE = 7 | |
67 # These last two indices are not part of the structure as defined in the | |
68 # spec, but they are used internally by this module as a convenience | |
69 _ECD_COMMENT = 8 | |
70 _ECD_LOCATION = 9 | |
71 | |
72 # The "central directory" structure, magic number, size, and indices | |
73 # of entries in the structure (section V.F in the format document) | |
74 structCentralDir = "<4s4B4HL2L5H2L" | |
75 stringCentralDir = b"PK\001\002" | |
76 sizeCentralDir = struct.calcsize(structCentralDir) | |
77 | |
78 # indexes of entries in the central directory structure | |
79 _CD_SIGNATURE = 0 | |
80 _CD_CREATE_VERSION = 1 | |
81 _CD_CREATE_SYSTEM = 2 | |
82 _CD_EXTRACT_VERSION = 3 | |
83 _CD_EXTRACT_SYSTEM = 4 | |
84 _CD_FLAG_BITS = 5 | |
85 _CD_COMPRESS_TYPE = 6 | |
86 _CD_TIME = 7 | |
87 _CD_DATE = 8 | |
88 _CD_CRC = 9 | |
89 _CD_COMPRESSED_SIZE = 10 | |
90 _CD_UNCOMPRESSED_SIZE = 11 | |
91 _CD_FILENAME_LENGTH = 12 | |
92 _CD_EXTRA_FIELD_LENGTH = 13 | |
93 _CD_COMMENT_LENGTH = 14 | |
94 _CD_DISK_NUMBER_START = 15 | |
95 _CD_INTERNAL_FILE_ATTRIBUTES = 16 | |
96 _CD_EXTERNAL_FILE_ATTRIBUTES = 17 | |
97 _CD_LOCAL_HEADER_OFFSET = 18 | |
98 | |
99 # The "local file header" structure, magic number, size, and indices | |
100 # (section V.A in the format document) | |
101 structFileHeader = "<4s2B4HL2L2H" | |
102 stringFileHeader = b"PK\003\004" | |
103 sizeFileHeader = struct.calcsize(structFileHeader) | |
104 | |
105 _FH_SIGNATURE = 0 | |
106 _FH_EXTRACT_VERSION = 1 | |
107 _FH_EXTRACT_SYSTEM = 2 | |
108 _FH_GENERAL_PURPOSE_FLAG_BITS = 3 | |
109 _FH_COMPRESSION_METHOD = 4 | |
110 _FH_LAST_MOD_TIME = 5 | |
111 _FH_LAST_MOD_DATE = 6 | |
112 _FH_CRC = 7 | |
113 _FH_COMPRESSED_SIZE = 8 | |
114 _FH_UNCOMPRESSED_SIZE = 9 | |
115 _FH_FILENAME_LENGTH = 10 | |
116 _FH_EXTRA_FIELD_LENGTH = 11 | |
117 | |
118 # The "Zip64 end of central directory locator" structure, magic number, and size | |
119 structEndArchive64Locator = "<4sLQL" | |
120 stringEndArchive64Locator = b"PK\x06\x07" | |
121 sizeEndCentDir64Locator = struct.calcsize(structEndArchive64Locator) | |
122 | |
123 # The "Zip64 end of central directory" record, magic number, size, and indices | |
124 # (section V.G in the format document) | |
125 structEndArchive64 = "<4sQ2H2L4Q" | |
126 stringEndArchive64 = b"PK\x06\x06" | |
127 sizeEndCentDir64 = struct.calcsize(structEndArchive64) | |
128 | |
129 _CD64_SIGNATURE = 0 | |
130 _CD64_DIRECTORY_RECSIZE = 1 | |
131 _CD64_CREATE_VERSION = 2 | |
132 _CD64_EXTRACT_VERSION = 3 | |
133 _CD64_DISK_NUMBER = 4 | |
134 _CD64_DISK_NUMBER_START = 5 | |
135 _CD64_NUMBER_ENTRIES_THIS_DISK = 6 | |
136 _CD64_NUMBER_ENTRIES_TOTAL = 7 | |
137 _CD64_DIRECTORY_SIZE = 8 | |
138 _CD64_OFFSET_START_CENTDIR = 9 | |
139 | |
140 def _check_zipfile(fp): | |
141 try: | |
142 if _EndRecData(fp): | |
143 return True # file has correct magic number | |
144 except IOError: | |
145 pass | |
146 return False | |
147 | |
148 def is_zipfile(filename): | |
149 """Quickly see if a file is a ZIP file by checking the magic number. | |
150 | |
151 The filename argument may be a file or file-like object too. | |
152 """ | |
153 result = False | |
154 try: | |
155 if hasattr(filename, "read"): | |
156 result = _check_zipfile(fp=filename) | |
157 else: | |
158 with open(filename, "rb") as fp: | |
159 result = _check_zipfile(fp) | |
160 except IOError: | |
161 pass | |
162 return result | |
163 | |
164 def _EndRecData64(fpin, offset, endrec): | |
165 """ | |
166 Read the ZIP64 end-of-archive records and use that to update endrec | |
167 """ | |
168 fpin.seek(offset - sizeEndCentDir64Locator, 2) | |
169 data = fpin.read(sizeEndCentDir64Locator) | |
170 sig, diskno, reloff, disks = struct.unpack(structEndArchive64Locator, data) | |
171 if sig != stringEndArchive64Locator: | |
172 return endrec | |
173 | |
174 if diskno != 0 or disks != 1: | |
175 raise BadZipfile("zipfiles that span multiple disks are not supported") | |
176 | |
177 # Assume no 'zip64 extensible data' | |
178 fpin.seek(offset - sizeEndCentDir64Locator - sizeEndCentDir64, 2) | |
179 data = fpin.read(sizeEndCentDir64) | |
180 sig, sz, create_version, read_version, disk_num, disk_dir, \ | |
181 dircount, dircount2, dirsize, diroffset = \ | |
182 struct.unpack(structEndArchive64, data) | |
183 if sig != stringEndArchive64: | |
184 return endrec | |
185 | |
186 # Update the original endrec using data from the ZIP64 record | |
187 endrec[_ECD_SIGNATURE] = sig | |
188 endrec[_ECD_DISK_NUMBER] = disk_num | |
189 endrec[_ECD_DISK_START] = disk_dir | |
190 endrec[_ECD_ENTRIES_THIS_DISK] = dircount | |
191 endrec[_ECD_ENTRIES_TOTAL] = dircount2 | |
192 endrec[_ECD_SIZE] = dirsize | |
193 endrec[_ECD_OFFSET] = diroffset | |
194 return endrec | |
195 | |
196 | |
197 def _EndRecData(fpin): | |
198 """Return data from the "End of Central Directory" record, or None. | |
199 | |
200 The data is a list of the nine items in the ZIP "End of central dir" | |
201 record followed by a tenth item, the file seek offset of this record.""" | |
202 | |
203 # Determine file size | |
204 fpin.seek(0, 2) | |
205 filesize = fpin.tell() | |
206 | |
207 # Check to see if this is ZIP file with no archive comment (the | |
208 # "end of central directory" structure should be the last item in the | |
209 # file if this is the case). | |
210 try: | |
211 fpin.seek(-sizeEndCentDir, 2) | |
212 except IOError: | |
213 return None | |
214 data = fpin.read() | |
215 if data[0:4] == stringEndArchive and data[-2:] == b"\000\000": | |
216 # the signature is correct and there's no comment, unpack structure | |
217 endrec = struct.unpack(structEndArchive, data) | |
218 endrec=list(endrec) | |
219 | |
220 # Append a blank comment and record start offset | |
221 endrec.append(b"") | |
222 endrec.append(filesize - sizeEndCentDir) | |
223 | |
224 # Try to read the "Zip64 end of central directory" structure | |
225 return _EndRecData64(fpin, -sizeEndCentDir, endrec) | |
226 | |
227 # Either this is not a ZIP file, or it is a ZIP file with an archive | |
228 # comment. Search the end of the file for the "end of central directory" | |
229 # record signature. The comment is the last item in the ZIP file and may be | |
230 # up to 64K long. It is assumed that the "end of central directory" magic | |
231 # number does not appear in the comment. | |
232 maxCommentStart = max(filesize - (1 << 16) - sizeEndCentDir, 0) | |
233 fpin.seek(maxCommentStart, 0) | |
234 data = fpin.read() | |
235 start = data.rfind(stringEndArchive) | |
236 if start >= 0: | |
237 # found the magic number; attempt to unpack and interpret | |
238 recData = data[start:start+sizeEndCentDir] | |
239 endrec = list(struct.unpack(structEndArchive, recData)) | |
240 comment = data[start+sizeEndCentDir:] | |
241 # check that comment length is correct | |
242 if endrec[_ECD_COMMENT_SIZE] == len(comment): | |
243 # Append the archive comment and start offset | |
244 endrec.append(comment) | |
245 endrec.append(maxCommentStart + start) | |
246 | |
247 # Try to read the "Zip64 end of central directory" structure | |
248 return _EndRecData64(fpin, maxCommentStart + start - filesize, | |
249 endrec) | |
250 | |
251 # Unable to find a valid end of central directory structure | |
252 return | |
253 | |
254 | |
255 class ZipInfo (object): | |
256 """Class with attributes describing each file in the ZIP archive.""" | |
257 | |
258 __slots__ = ( | |
259 'orig_filename', | |
260 'filename', | |
261 'date_time', | |
262 'compress_type', | |
263 'comment', | |
264 'extra', | |
265 'create_system', | |
266 'create_version', | |
267 'extract_version', | |
268 'reserved', | |
269 'flag_bits', | |
270 'volume', | |
271 'internal_attr', | |
272 'external_attr', | |
273 'header_offset', | |
274 'CRC', | |
275 'compress_size', | |
276 'file_size', | |
277 '_raw_time', | |
278 ) | |
279 | |
280 def __init__(self, filename="NoName", date_time=(1980,1,1,0,0,0)): | |
281 self.orig_filename = filename # Original file name in archive | |
282 | |
283 # Terminate the file name at the first null byte. Null bytes in file | |
284 # names are used as tricks by viruses in archives. | |
285 null_byte = filename.find(chr(0)) | |
286 if null_byte >= 0: | |
287 filename = filename[0:null_byte] | |
288 # This is used to ensure paths in generated ZIP files always use | |
289 # forward slashes as the directory separator, as required by the | |
290 # ZIP format specification. | |
291 if os.sep != "/" and os.sep in filename: | |
292 filename = filename.replace(os.sep, "/") | |
293 | |
294 self.filename = filename # Normalized file name | |
295 self.date_time = date_time # year, month, day, hour, min, sec | |
296 # Standard values: | |
297 self.compress_type = ZIP_STORED # Type of compression for the file | |
298 self.comment = b"" # Comment for each file | |
299 self.extra = b"" # ZIP extra data | |
300 if sys.platform == 'win32': | |
301 self.create_system = 0 # System which created ZIP archive | |
302 else: | |
303 # Assume everything else is unix-y | |
304 self.create_system = 3 # System which created ZIP archive | |
305 self.create_version = 20 # Version which created ZIP archive | |
306 self.extract_version = 20 # Version needed to extract archive | |
307 self.reserved = 0 # Must be zero | |
308 self.flag_bits = 0 # ZIP flag bits | |
309 self.volume = 0 # Volume number of file header | |
310 self.internal_attr = 0 # Internal attributes | |
311 self.external_attr = 0 # External file attributes | |
312 # Other attributes are set by class ZipFile: | |
313 # header_offset Byte offset to the file header | |
314 # CRC CRC-32 of the uncompressed file | |
315 # compress_size Size of the compressed file | |
316 # file_size Size of the uncompressed file | |
317 | |
318 def FileHeader(self): | |
319 """Return the per-file header as a string.""" | |
320 dt = self.date_time | |
321 dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2] | |
322 dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2) | |
323 if self.flag_bits & 0x08: | |
324 # Set these to zero because we write them after the file data | |
325 CRC = compress_size = file_size = 0 | |
326 else: | |
327 CRC = self.CRC | |
328 compress_size = self.compress_size | |
329 file_size = self.file_size | |
330 | |
331 extra = self.extra | |
332 | |
333 if file_size > ZIP64_LIMIT or compress_size > ZIP64_LIMIT: | |
334 # File is larger than what fits into a 4 byte integer, | |
335 # fall back to the ZIP64 extension | |
336 fmt = '<HHQQ' | |
337 extra = extra + struct.pack(fmt, | |
338 1, struct.calcsize(fmt)-4, file_size, compress_size) | |
339 file_size = 0xffffffff | |
340 compress_size = 0xffffffff | |
341 self.extract_version = max(45, self.extract_version) | |
342 self.create_version = max(45, self.extract_version) | |
343 | |
344 filename, flag_bits = self._encodeFilenameFlags() | |
345 header = struct.pack(structFileHeader, stringFileHeader, | |
346 self.extract_version, self.reserved, flag_bits, | |
347 self.compress_type, dostime, dosdate, CRC, | |
348 compress_size, file_size, | |
349 len(filename), len(extra)) | |
350 return header + filename + extra | |
351 | |
352 def _encodeFilenameFlags(self): | |
353 try: | |
354 return self.filename.encode('ascii'), self.flag_bits | |
355 except UnicodeEncodeError: | |
356 return self.filename.encode('utf-8'), self.flag_bits | 0x800 | |
357 | |
358 def _decodeExtra(self): | |
359 # Try to decode the extra field. | |
360 extra = self.extra | |
361 unpack = struct.unpack | |
362 while extra: | |
363 tp, ln = unpack('<HH', extra[:4]) | |
364 if tp == 1: | |
365 if ln >= 24: | |
366 counts = unpack('<QQQ', extra[4:28]) | |
367 elif ln == 16: | |
368 counts = unpack('<QQ', extra[4:20]) | |
369 elif ln == 8: | |
370 counts = unpack('<Q', extra[4:12]) | |
371 elif ln == 0: | |
372 counts = () | |
373 else: | |
374 raise RuntimeError("Corrupt extra field %s"%(ln,)) | |
375 | |
376 idx = 0 | |
377 | |
378 # ZIP64 extension (large files and/or large archives) | |
379 if self.file_size in (0xffffffffffffffff, 0xffffffff): | |
380 self.file_size = counts[idx] | |
381 idx += 1 | |
382 | |
383 if self.compress_size == 0xFFFFFFFF: | |
384 self.compress_size = counts[idx] | |
385 idx += 1 | |
386 | |
387 if self.header_offset == 0xffffffff: | |
388 old = self.header_offset | |
389 self.header_offset = counts[idx] | |
390 idx+=1 | |
391 | |
392 extra = extra[ln+4:] | |
393 | |
394 | |
395 class _ZipDecrypter: | |
396 """Class to handle decryption of files stored within a ZIP archive. | |
397 | |
398 ZIP supports a password-based form of encryption. Even though known | |
399 plaintext attacks have been found against it, it is still useful | |
400 to be able to get data out of such a file. | |
401 | |
402 Usage: | |
403 zd = _ZipDecrypter(mypwd) | |
404 plain_char = zd(cypher_char) | |
405 plain_text = map(zd, cypher_text) | |
406 """ | |
407 | |
408 def _GenerateCRCTable(): | |
409 """Generate a CRC-32 table. | |
410 | |
411 ZIP encryption uses the CRC32 one-byte primitive for scrambling some | |
412 internal keys. We noticed that a direct implementation is faster than | |
413 relying on binascii.crc32(). | |
414 """ | |
415 poly = 0xedb88320 | |
416 table = [0] * 256 | |
417 for i in range(256): | |
418 crc = i | |
419 for j in range(8): | |
420 if crc & 1: | |
421 crc = ((crc >> 1) & 0x7FFFFFFF) ^ poly | |
422 else: | |
423 crc = ((crc >> 1) & 0x7FFFFFFF) | |
424 table[i] = crc | |
425 return table | |
426 crctable = _GenerateCRCTable() | |
427 | |
428 def _crc32(self, ch, crc): | |
429 """Compute the CRC32 primitive on one byte.""" | |
430 return ((crc >> 8) & 0xffffff) ^ self.crctable[(crc ^ ch) & 0xff] | |
431 | |
432 def __init__(self, pwd): | |
433 self.key0 = 305419896 | |
434 self.key1 = 591751049 | |
435 self.key2 = 878082192 | |
436 for p in pwd: | |
437 self._UpdateKeys(p) | |
438 | |
439 def _UpdateKeys(self, c): | |
440 self.key0 = self._crc32(c, self.key0) | |
441 self.key1 = (self.key1 + (self.key0 & 255)) & 4294967295 | |
442 self.key1 = (self.key1 * 134775813 + 1) & 4294967295 | |
443 self.key2 = self._crc32((self.key1 >> 24) & 255, self.key2) | |
444 | |
445 def __call__(self, c): | |
446 """Decrypt a single character.""" | |
447 assert isinstance(c, int) | |
448 k = self.key2 | 2 | |
449 c = c ^ (((k * (k^1)) >> 8) & 255) | |
450 self._UpdateKeys(c) | |
451 return c | |
452 | |
453 class ZipExtFile: | |
454 """File-like object for reading an archive member. | |
455 Is returned by ZipFile.open(). | |
456 """ | |
457 | |
458 def __init__(self, fileobj, zipinfo, decrypt=None): | |
459 self.fileobj = fileobj | |
460 self.decrypter = decrypt | |
461 self.bytes_read = 0 | |
462 self.rawbuffer = b'' | |
463 self.readbuffer = b'' | |
464 self.linebuffer = b'' | |
465 self.eof = False | |
466 self.univ_newlines = False | |
467 self.nlSeps = (b"\n", ) | |
468 self.lastdiscard = b'' | |
469 | |
470 self.compress_type = zipinfo.compress_type | |
471 self.compress_size = zipinfo.compress_size | |
472 | |
473 self.closed = False | |
474 self.mode = "r" | |
475 self.name = zipinfo.filename | |
476 | |
477 # read from compressed files in 64k blocks | |
478 self.compreadsize = 64*1024 | |
479 if self.compress_type == ZIP_DEFLATED: | |
480 self.dc = zlib.decompressobj(-15) | |
481 elif self.compress_type == ZIP_BZIP2: | |
482 self.dc = bz2.BZ2Decompressor() | |
483 self.compreadsize = 900000 | |
484 | |
29
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
485 if hasattr(zipinfo, 'CRC'): |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
486 self._expected_crc = zipinfo.CRC |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
487 self._running_crc = crc32(b'') & 0xffffffff |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
488 else: |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
489 self._expected_crc = None |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
490 |
21 | 491 def set_univ_newlines(self, univ_newlines): |
492 self.univ_newlines = univ_newlines | |
493 | |
494 # pick line separator char(s) based on universal newlines flag | |
495 self.nlSeps = (b"\n", ) | |
496 if self.univ_newlines: | |
497 self.nlSeps = (b"\r\n", b"\r", b"\n") | |
498 | |
499 def __iter__(self): | |
500 return self | |
501 | |
502 def __next__(self): | |
503 nextline = self.readline() | |
504 if not nextline: | |
505 raise StopIteration() | |
506 | |
507 return nextline | |
508 | |
509 def close(self): | |
510 self.closed = True | |
511 | |
512 def _checkfornewline(self): | |
513 nl, nllen = -1, -1 | |
514 if self.linebuffer: | |
515 # ugly check for cases where half of an \r\n pair was | |
516 # read on the last pass, and the \r was discarded. In this | |
517 # case we just throw away the \n at the start of the buffer. | |
518 if (self.lastdiscard, self.linebuffer[:1]) == (b'\r', b'\n'): | |
519 self.linebuffer = self.linebuffer[1:] | |
520 | |
521 for sep in self.nlSeps: | |
522 nl = self.linebuffer.find(sep) | |
523 if nl >= 0: | |
524 nllen = len(sep) | |
525 return nl, nllen | |
526 | |
527 return nl, nllen | |
528 | |
529 def readline(self, size = -1): | |
530 """Read a line with approx. size. If size is negative, | |
531 read a whole line. | |
532 """ | |
533 if size < 0: | |
534 size = sys.maxsize | |
535 elif size == 0: | |
536 return b'' | |
537 | |
538 # check for a newline already in buffer | |
539 nl, nllen = self._checkfornewline() | |
540 | |
541 if nl >= 0: | |
542 # the next line was already in the buffer | |
543 nl = min(nl, size) | |
544 else: | |
545 # no line break in buffer - try to read more | |
546 size -= len(self.linebuffer) | |
547 while nl < 0 and size > 0: | |
548 buf = self.read(min(size, 100)) | |
549 if not buf: | |
550 break | |
551 self.linebuffer += buf | |
552 size -= len(buf) | |
553 | |
554 # check for a newline in buffer | |
555 nl, nllen = self._checkfornewline() | |
556 | |
557 # we either ran out of bytes in the file, or | |
558 # met the specified size limit without finding a newline, | |
559 # so return current buffer | |
560 if nl < 0: | |
561 s = self.linebuffer | |
562 self.linebuffer = b'' | |
563 return s | |
564 | |
565 buf = self.linebuffer[:nl] | |
566 self.lastdiscard = self.linebuffer[nl:nl + nllen] | |
567 self.linebuffer = self.linebuffer[nl + nllen:] | |
568 | |
569 # line is always returned with \n as newline char (except possibly | |
570 # for a final incomplete line in the file, which is handled above). | |
571 return buf + b"\n" | |
572 | |
573 def readlines(self, sizehint = -1): | |
574 """Return a list with all (following) lines. The sizehint parameter | |
575 is ignored in this implementation. | |
576 """ | |
577 result = [] | |
578 while True: | |
579 line = self.readline() | |
580 if not line: break | |
581 result.append(line) | |
582 return result | |
583 | |
29
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
584 def _update_crc(self, newdata, eof): |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
585 # Update the CRC using the given data. |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
586 if self._expected_crc is None: |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
587 # No need to compute the CRC if we don't have a reference value |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
588 return |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
589 self._running_crc = crc32(newdata, self._running_crc) & 0xffffffff |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
590 # Check the CRC if we're at the end of the file |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
591 if eof and self._running_crc != self._expected_crc: |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
592 raise BadZipfile("Bad CRC-32 for file %r" % self.name) |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
593 |
21 | 594 def read(self, size = None): |
595 # act like file obj and return empty string if size is 0 | |
596 if size == 0: | |
597 return b'' | |
598 | |
599 # determine read size | |
600 bytesToRead = self.compress_size - self.bytes_read | |
601 | |
602 # adjust read size for encrypted files since the first 12 bytes | |
603 # are for the encryption/password information | |
604 if self.decrypter is not None: | |
605 bytesToRead -= 12 | |
606 | |
607 if size is not None and size >= 0: | |
608 if self.compress_type == ZIP_STORED: | |
609 lr = len(self.readbuffer) | |
610 bytesToRead = min(bytesToRead, size - lr) | |
611 else: | |
612 if len(self.readbuffer) > size: | |
613 # the user has requested fewer bytes than we've already | |
614 # pulled through the decompressor; don't read any more | |
615 bytesToRead = 0 | |
616 else: | |
617 # user will use up the buffer, so read some more | |
618 lr = len(self.rawbuffer) | |
619 bytesToRead = min(bytesToRead, self.compreadsize - lr) | |
620 | |
621 # avoid reading past end of file contents | |
622 if bytesToRead + self.bytes_read > self.compress_size: | |
623 bytesToRead = self.compress_size - self.bytes_read | |
624 | |
625 # try to read from file (if necessary) | |
626 if bytesToRead > 0: | |
627 data = self.fileobj.read(bytesToRead) | |
628 self.bytes_read += len(data) | |
629 try: | |
630 self.rawbuffer += data | |
631 except: | |
632 print(repr(self.fileobj), repr(self.rawbuffer), | |
633 repr(data)) | |
634 raise | |
635 | |
636 # handle contents of raw buffer | |
637 if self.rawbuffer: | |
638 newdata = self.rawbuffer | |
639 self.rawbuffer = b'' | |
640 | |
641 # decrypt new data if we were given an object to handle that | |
642 if newdata and self.decrypter is not None: | |
643 newdata = bytes(map(self.decrypter, newdata)) | |
644 | |
645 # decompress newly read data if necessary | |
646 if newdata and self.compress_type != ZIP_STORED: | |
647 newdata = self.dc.decompress(newdata) | |
648 self.rawbuffer = self.dc.unconsumed_tail if self.compress_type == ZIP_DEFLATED else '' | |
649 if self.eof and len(self.rawbuffer) == 0: | |
650 # we're out of raw bytes (both from the file and | |
651 # the local buffer); flush just to make sure the | |
652 # decompressor is done | |
28 | 653 try: |
21 | 654 newdata += self.dc.flush() |
28 | 655 except AttributeError: |
656 pass | |
21 | 657 # prevent decompressor from being used again |
658 self.dc = None | |
659 | |
29
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
660 self._update_crc(newdata, eof=( |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
661 self.compress_size == self.bytes_read and |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
662 len(self.rawbuffer) == 0)) |
21 | 663 self.readbuffer += newdata |
664 | |
665 # return what the user asked for | |
666 if size is None or len(self.readbuffer) <= size: | |
667 data = self.readbuffer | |
668 self.readbuffer = b'' | |
669 else: | |
670 data = self.readbuffer[:size] | |
671 self.readbuffer = self.readbuffer[size:] | |
672 | |
673 return data | |
674 | |
675 | |
676 class ZipFile: | |
677 """ Class with methods to open, read, write, close, list zip files. | |
678 | |
679 z = ZipFile(file, mode="r", compression=ZIP_STORED, allowZip64=False) | |
680 | |
681 file: Either the path to the file, or a file-like object. | |
682 If it is a path, the file will be opened and closed by ZipFile. | |
683 mode: The mode can be either read "r", write "w" or append "a". | |
684 compression: ZIP_STORED (no compression), ZIP_DEFLATED (requires zlib), | |
685 or ZIP_BZIP2 (requires bz2). | |
686 allowZip64: if True ZipFile will create files with ZIP64 extensions when | |
687 needed, otherwise it will raise an exception when this would | |
688 be necessary. | |
689 | |
690 """ | |
691 | |
692 fp = None # Set here since __del__ checks it | |
693 | |
694 def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=False): | |
695 """Open the ZIP file with mode read "r", write "w" or append "a".""" | |
696 if mode not in ("r", "w", "a"): | |
697 raise RuntimeError('ZipFile() requires mode "r", "w", or "a"') | |
698 | |
699 if compression == ZIP_STORED: | |
700 pass | |
701 elif compression == ZIP_DEFLATED: | |
702 if not zlib: | |
703 raise RuntimeError( | |
704 "Compression requires the (missing) zlib module") | |
705 elif compression == ZIP_BZIP2: | |
706 if not bz2: | |
707 raise RuntimeError( | |
708 "Compression requires the (missing) bz2 module") | |
709 else: | |
710 raise RuntimeError("That compression method is not supported") | |
711 | |
712 self._allowZip64 = allowZip64 | |
713 self._didModify = False | |
714 self.debug = 0 # Level of printing: 0 through 3 | |
715 self.NameToInfo = {} # Find file info given name | |
716 self.filelist = [] # List of ZipInfo instances for archive | |
717 self.compression = compression # Method of compression | |
718 self.mode = key = mode.replace('b', '')[0] | |
719 self.pwd = None | |
720 self.comment = b'' | |
721 | |
722 # Check if we were passed a file-like object | |
723 if isinstance(file, str): | |
724 # No, it's a filename | |
725 self._filePassed = 0 | |
726 self.filename = file | |
727 modeDict = {'r' : 'rb', 'w': 'wb', 'a' : 'r+b'} | |
728 try: | |
729 self.fp = io.open(file, modeDict[mode]) | |
730 except IOError: | |
731 if mode == 'a': | |
732 mode = key = 'w' | |
733 self.fp = io.open(file, modeDict[mode]) | |
734 else: | |
735 raise | |
736 else: | |
737 self._filePassed = 1 | |
738 self.fp = file | |
739 self.filename = getattr(file, 'name', None) | |
740 | |
741 if key == 'r': | |
742 self._GetContents() | |
743 elif key == 'w': | |
744 pass | |
745 elif key == 'a': | |
746 try: # See if file is a zip file | |
747 self._RealGetContents() | |
748 # seek to start of directory and overwrite | |
749 self.fp.seek(self.start_dir, 0) | |
750 except BadZipfile: # file is not a zip file, just append | |
751 self.fp.seek(0, 2) | |
752 else: | |
753 if not self._filePassed: | |
754 self.fp.close() | |
755 self.fp = None | |
756 raise RuntimeError('Mode must be "r", "w" or "a"') | |
757 | |
758 def _GetContents(self): | |
759 """Read the directory, making sure we close the file if the format | |
760 is bad.""" | |
761 try: | |
762 self._RealGetContents() | |
763 except BadZipfile: | |
764 if not self._filePassed: | |
765 self.fp.close() | |
766 self.fp = None | |
767 raise | |
768 | |
769 def _RealGetContents(self): | |
770 """Read in the table of contents for the ZIP file.""" | |
771 fp = self.fp | |
772 endrec = _EndRecData(fp) | |
773 if not endrec: | |
774 raise BadZipfile("File is not a zip file") | |
775 if self.debug > 1: | |
776 print(endrec) | |
777 size_cd = endrec[_ECD_SIZE] # bytes in central directory | |
778 offset_cd = endrec[_ECD_OFFSET] # offset of central directory | |
779 self.comment = endrec[_ECD_COMMENT] # archive comment | |
780 | |
781 # "concat" is zero, unless zip was concatenated to another file | |
782 concat = endrec[_ECD_LOCATION] - size_cd - offset_cd | |
783 if endrec[_ECD_SIGNATURE] == stringEndArchive64: | |
784 # If Zip64 extension structures are present, account for them | |
785 concat -= (sizeEndCentDir64 + sizeEndCentDir64Locator) | |
786 | |
787 if self.debug > 2: | |
788 inferred = concat + offset_cd | |
789 print("given, inferred, offset", offset_cd, inferred, concat) | |
790 # self.start_dir: Position of start of central directory | |
791 self.start_dir = offset_cd + concat | |
792 fp.seek(self.start_dir, 0) | |
793 data = fp.read(size_cd) | |
794 fp = io.BytesIO(data) | |
795 total = 0 | |
796 while total < size_cd: | |
797 centdir = fp.read(sizeCentralDir) | |
798 if centdir[0:4] != stringCentralDir: | |
799 raise BadZipfile("Bad magic number for central directory") | |
800 centdir = struct.unpack(structCentralDir, centdir) | |
801 if self.debug > 2: | |
802 print(centdir) | |
803 filename = fp.read(centdir[_CD_FILENAME_LENGTH]) | |
804 flags = centdir[5] | |
805 if flags & 0x800: | |
806 # UTF-8 file names extension | |
807 filename = filename.decode('utf-8') | |
808 else: | |
809 # Historical ZIP filename encoding | |
810 filename = filename.decode('cp437') | |
811 # Create ZipInfo instance to store file information | |
812 x = ZipInfo(filename) | |
813 x.extra = fp.read(centdir[_CD_EXTRA_FIELD_LENGTH]) | |
814 x.comment = fp.read(centdir[_CD_COMMENT_LENGTH]) | |
815 x.header_offset = centdir[_CD_LOCAL_HEADER_OFFSET] | |
816 (x.create_version, x.create_system, x.extract_version, x.reserved, | |
817 x.flag_bits, x.compress_type, t, d, | |
818 x.CRC, x.compress_size, x.file_size) = centdir[1:12] | |
819 x.volume, x.internal_attr, x.external_attr = centdir[15:18] | |
820 # Convert date/time code to (year, month, day, hour, min, sec) | |
821 x._raw_time = t | |
822 x.date_time = ( (d>>9)+1980, (d>>5)&0xF, d&0x1F, | |
823 t>>11, (t>>5)&0x3F, (t&0x1F) * 2 ) | |
824 | |
825 x._decodeExtra() | |
826 x.header_offset = x.header_offset + concat | |
827 self.filelist.append(x) | |
828 self.NameToInfo[x.filename] = x | |
829 | |
830 # update total bytes read from central directory | |
831 total = (total + sizeCentralDir + centdir[_CD_FILENAME_LENGTH] | |
832 + centdir[_CD_EXTRA_FIELD_LENGTH] | |
833 + centdir[_CD_COMMENT_LENGTH]) | |
834 | |
835 if self.debug > 2: | |
836 print("total", total) | |
837 | |
838 | |
839 def namelist(self): | |
840 """Return a list of file names in the archive.""" | |
841 l = [] | |
842 for data in self.filelist: | |
843 l.append(data.filename) | |
844 return l | |
845 | |
846 def infolist(self): | |
847 """Return a list of class ZipInfo instances for files in the | |
848 archive.""" | |
849 return self.filelist | |
850 | |
851 def printdir(self, file=None): | |
852 """Print a table of contents for the zip file.""" | |
853 print("%-46s %19s %12s" % ("File Name", "Modified ", "Size"), | |
854 file=file) | |
855 for zinfo in self.filelist: | |
856 date = "%d-%02d-%02d %02d:%02d:%02d" % zinfo.date_time[:6] | |
857 print("%-46s %s %12d" % (zinfo.filename, date, zinfo.file_size), | |
858 file=file) | |
859 | |
860 def testzip(self): | |
861 """Read all the files and check the CRC.""" | |
862 chunk_size = 2 ** 20 | |
863 for zinfo in self.filelist: | |
864 try: | |
865 # Read by chunks, to avoid an OverflowError or a | |
866 # MemoryError with very large embedded files. | |
867 f = self.open(zinfo.filename, "r") | |
868 while f.read(chunk_size): # Check CRC-32 | |
869 pass | |
870 except BadZipfile: | |
871 return zinfo.filename | |
872 | |
873 def getinfo(self, name): | |
874 """Return the instance of ZipInfo given 'name'.""" | |
875 info = self.NameToInfo.get(name) | |
876 if info is None: | |
877 raise KeyError( | |
878 'There is no item named %r in the archive' % name) | |
879 | |
880 return info | |
881 | |
882 def setpassword(self, pwd): | |
883 """Set default password for encrypted files.""" | |
884 assert isinstance(pwd, bytes) | |
885 self.pwd = pwd | |
886 | |
887 def read(self, name, pwd=None): | |
888 """Return file bytes (as a string) for name.""" | |
889 return self.open(name, "r", pwd).read() | |
890 | |
891 def open(self, name, mode="r", pwd=None): | |
892 """Return file-like object for 'name'.""" | |
893 if mode not in ("r", "U", "rU"): | |
894 raise RuntimeError('open() requires mode "r", "U", or "rU"') | |
895 if not self.fp: | |
896 raise RuntimeError( | |
897 "Attempt to read ZIP archive that was already closed") | |
898 | |
899 # Only open a new file for instances where we were not | |
900 # given a file object in the constructor | |
901 if self._filePassed: | |
902 zef_file = self.fp | |
903 else: | |
904 zef_file = io.open(self.filename, 'rb') | |
905 | |
906 # Make sure we have an info object | |
907 if isinstance(name, ZipInfo): | |
908 # 'name' is already an info object | |
909 zinfo = name | |
910 else: | |
911 # Get info object for name | |
912 zinfo = self.getinfo(name) | |
913 | |
914 zef_file.seek(zinfo.header_offset, 0) | |
915 | |
916 # Skip the file header: | |
917 fheader = zef_file.read(sizeFileHeader) | |
918 if fheader[0:4] != stringFileHeader: | |
919 raise BadZipfile("Bad magic number for file header") | |
920 | |
921 fheader = struct.unpack(structFileHeader, fheader) | |
922 fname = zef_file.read(fheader[_FH_FILENAME_LENGTH]) | |
923 if fheader[_FH_EXTRA_FIELD_LENGTH]: | |
924 zef_file.read(fheader[_FH_EXTRA_FIELD_LENGTH]) | |
925 | |
926 if fname != zinfo.orig_filename.encode("utf-8"): | |
927 raise BadZipfile( | |
928 'File name in directory %r and header %r differ.' | |
929 % (zinfo.orig_filename, fname)) | |
930 | |
931 # check for encrypted flag & handle password | |
932 is_encrypted = zinfo.flag_bits & 0x1 | |
933 zd = None | |
934 if is_encrypted: | |
935 if not pwd: | |
936 pwd = self.pwd | |
937 if not pwd: | |
938 raise RuntimeError("File %s is encrypted, " | |
939 "password required for extraction" % name) | |
940 | |
941 zd = _ZipDecrypter(pwd) | |
942 # The first 12 bytes in the cypher stream is an encryption header | |
943 # used to strengthen the algorithm. The first 11 bytes are | |
944 # completely random, while the 12th contains the MSB of the CRC, | |
945 # or the MSB of the file time depending on the header type | |
946 # and is used to check the correctness of the password. | |
947 bytes = zef_file.read(12) | |
948 h = list(map(zd, bytes[0:12])) | |
949 if zinfo.flag_bits & 0x8: | |
950 # compare against the file type from extended local headers | |
951 check_byte = (zinfo._raw_time >> 8) & 0xff | |
952 else: | |
953 # compare against the CRC otherwise | |
954 check_byte = (zinfo.CRC >> 24) & 0xff | |
955 if h[11] != check_byte: | |
956 raise RuntimeError("Bad password for file", name) | |
957 | |
958 # build and return a ZipExtFile | |
959 if zd is None: | |
960 zef = ZipExtFile(zef_file, zinfo) | |
961 else: | |
962 zef = ZipExtFile(zef_file, zinfo, zd) | |
963 | |
964 # set universal newlines on ZipExtFile if necessary | |
965 if "U" in mode: | |
966 zef.set_univ_newlines(True) | |
967 return zef | |
968 | |
969 def extract(self, member, path=None, pwd=None): | |
970 """Extract a member from the archive to the current working directory, | |
971 using its full name. Its file information is extracted as accurately | |
972 as possible. `member' may be a filename or a ZipInfo object. You can | |
973 specify a different directory using `path'. | |
974 """ | |
975 if not isinstance(member, ZipInfo): | |
976 member = self.getinfo(member) | |
977 | |
978 if path is None: | |
979 path = os.getcwd() | |
980 | |
981 return self._extract_member(member, path, pwd) | |
982 | |
983 def extractall(self, path=None, members=None, pwd=None): | |
984 """Extract all members from the archive to the current working | |
985 directory. `path' specifies a different directory to extract to. | |
986 `members' is optional and must be a subset of the list returned | |
987 by namelist(). | |
988 """ | |
989 if members is None: | |
990 members = self.namelist() | |
991 | |
992 for zipinfo in members: | |
993 self.extract(zipinfo, path, pwd) | |
994 | |
995 def _extract_member(self, member, targetpath, pwd): | |
996 """Extract the ZipInfo object 'member' to a physical | |
997 file on the path targetpath. | |
998 """ | |
999 # build the destination pathname, replacing | |
1000 # forward slashes to platform specific separators. | |
1001 # Strip trailing path separator, unless it represents the root. | |
1002 if (targetpath[-1:] in (os.path.sep, os.path.altsep) | |
1003 and len(os.path.splitdrive(targetpath)[1]) > 1): | |
1004 targetpath = targetpath[:-1] | |
1005 | |
1006 # don't include leading "/" from file name if present | |
1007 if member.filename[0] == '/': | |
1008 targetpath = os.path.join(targetpath, member.filename[1:]) | |
1009 else: | |
1010 targetpath = os.path.join(targetpath, member.filename) | |
1011 | |
1012 targetpath = os.path.normpath(targetpath) | |
1013 | |
1014 # Create all upper directories if necessary. | |
1015 upperdirs = os.path.dirname(targetpath) | |
1016 if upperdirs and not os.path.exists(upperdirs): | |
1017 os.makedirs(upperdirs) | |
1018 | |
1019 if member.filename[-1] == '/': | |
1020 if not os.path.isdir(targetpath): | |
1021 os.mkdir(targetpath) | |
1022 return targetpath | |
1023 | |
1024 source = self.open(member, pwd=pwd) | |
1025 target = open(targetpath, "wb") | |
1026 shutil.copyfileobj(source, target) | |
1027 source.close() | |
1028 target.close() | |
1029 | |
1030 return targetpath | |
1031 | |
1032 def _writecheck(self, zinfo): | |
1033 """Check for errors before writing a file to the archive.""" | |
1034 if zinfo.filename in self.NameToInfo: | |
1035 if self.debug: # Warning for duplicate names | |
1036 print("Duplicate name:", zinfo.filename) | |
1037 if self.mode not in ("w", "a"): | |
1038 raise RuntimeError('write() requires mode "w" or "a"') | |
1039 if not self.fp: | |
1040 raise RuntimeError( | |
1041 "Attempt to write ZIP archive that was already closed") | |
1042 if zinfo.compress_type == ZIP_DEFLATED and not zlib: | |
1043 raise RuntimeError( | |
1044 "Compression requires the (missing) zlib module") | |
1045 if zinfo.compress_type == ZIP_BZIP2 and not bz2: | |
1046 raise RuntimeError( | |
1047 "Compression requires the (missing) bz2 module") | |
1048 if zinfo.compress_type not in (ZIP_STORED, ZIP_DEFLATED, ZIP_BZIP2): | |
1049 raise RuntimeError("That compression method is not supported") | |
1050 if zinfo.file_size > ZIP64_LIMIT: | |
1051 if not self._allowZip64: | |
1052 raise LargeZipFile("Filesize would require ZIP64 extensions") | |
1053 if zinfo.header_offset > ZIP64_LIMIT: | |
1054 if not self._allowZip64: | |
1055 raise LargeZipFile( | |
1056 "Zipfile size would require ZIP64 extensions") | |
1057 | |
1058 def write(self, filename, arcname=None, compress_type=None): | |
1059 """Put the bytes from filename into the archive under the name | |
1060 arcname.""" | |
1061 if not self.fp: | |
1062 raise RuntimeError( | |
1063 "Attempt to write to ZIP archive that was already closed") | |
1064 | |
1065 st = os.stat(filename) | |
1066 isdir = stat.S_ISDIR(st.st_mode) | |
1067 mtime = time.localtime(st.st_mtime) | |
1068 date_time = mtime[0:6] | |
1069 # Create ZipInfo instance to store file information | |
1070 if arcname is None: | |
1071 arcname = filename | |
1072 arcname = os.path.normpath(os.path.splitdrive(arcname)[1]) | |
1073 while arcname[0] in (os.sep, os.altsep): | |
1074 arcname = arcname[1:] | |
1075 if isdir: | |
1076 arcname += '/' | |
1077 zinfo = ZipInfo(arcname, date_time) | |
1078 zinfo.external_attr = (st[0] & 0xFFFF) << 16 # Unix attributes | |
1079 if compress_type is None: | |
1080 zinfo.compress_type = self.compression | |
1081 else: | |
1082 zinfo.compress_type = compress_type | |
1083 | |
1084 zinfo.file_size = st.st_size | |
1085 zinfo.flag_bits = 0x00 | |
1086 zinfo.header_offset = self.fp.tell() # Start of header bytes | |
1087 | |
1088 self._writecheck(zinfo) | |
1089 self._didModify = True | |
1090 | |
1091 if isdir: | |
1092 zinfo.file_size = 0 | |
1093 zinfo.compress_size = 0 | |
1094 zinfo.CRC = 0 | |
1095 self.filelist.append(zinfo) | |
1096 self.NameToInfo[zinfo.filename] = zinfo | |
1097 self.fp.write(zinfo.FileHeader()) | |
1098 return | |
1099 | |
1100 with open(filename, "rb") as fp: | |
1101 # Must overwrite CRC and sizes with correct data later | |
1102 zinfo.CRC = CRC = 0 | |
1103 zinfo.compress_size = compress_size = 0 | |
1104 zinfo.file_size = file_size = 0 | |
1105 self.fp.write(zinfo.FileHeader()) | |
1106 if zinfo.compress_type == ZIP_DEFLATED: | |
1107 cmpr = zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION, | |
1108 zlib.DEFLATED, -15) | |
1109 elif zinfo.compress_type == ZIP_BZIP2: | |
1110 cmpr = bz2.BZ2Compressor() | |
1111 else: | |
1112 cmpr = None | |
1113 while 1: | |
1114 buf = fp.read(1024 * 8) | |
1115 if not buf: | |
1116 break | |
1117 file_size = file_size + len(buf) | |
1118 CRC = crc32(buf, CRC) & 0xffffffff | |
1119 if cmpr: | |
1120 buf = cmpr.compress(buf) | |
1121 compress_size = compress_size + len(buf) | |
1122 self.fp.write(buf) | |
1123 if cmpr: | |
1124 buf = cmpr.flush() | |
1125 compress_size = compress_size + len(buf) | |
1126 self.fp.write(buf) | |
1127 zinfo.compress_size = compress_size | |
1128 else: | |
1129 zinfo.compress_size = file_size | |
1130 zinfo.CRC = CRC | |
1131 zinfo.file_size = file_size | |
1132 # Seek backwards and write CRC and file sizes | |
1133 position = self.fp.tell() # Preserve current position in file | |
1134 self.fp.seek(zinfo.header_offset + 14, 0) | |
1135 self.fp.write(struct.pack("<LLL", zinfo.CRC, zinfo.compress_size, | |
1136 zinfo.file_size)) | |
1137 self.fp.seek(position, 0) | |
1138 self.filelist.append(zinfo) | |
1139 self.NameToInfo[zinfo.filename] = zinfo | |
1140 | |
1141 def writestr(self, zinfo_or_arcname, data): | |
1142 """Write a file into the archive. The contents is 'data', which | |
1143 may be either a 'str' or a 'bytes' instance; if it is a 'str', | |
1144 it is encoded as UTF-8 first. | |
1145 'zinfo_or_arcname' is either a ZipInfo instance or | |
1146 the name of the file in the archive.""" | |
1147 if isinstance(data, str): | |
1148 data = data.encode("utf-8") | |
1149 if not isinstance(zinfo_or_arcname, ZipInfo): | |
1150 zinfo = ZipInfo(filename=zinfo_or_arcname, | |
1151 date_time=time.localtime(time.time())[:6]) | |
1152 zinfo.compress_type = self.compression | |
1153 zinfo.external_attr = 0o600 << 16 | |
1154 else: | |
1155 zinfo = zinfo_or_arcname | |
1156 | |
1157 if not self.fp: | |
1158 raise RuntimeError( | |
1159 "Attempt to write to ZIP archive that was already closed") | |
1160 | |
1161 zinfo.file_size = len(data) # Uncompressed size | |
1162 zinfo.header_offset = self.fp.tell() # Start of header data | |
1163 self._writecheck(zinfo) | |
1164 self._didModify = True | |
1165 zinfo.CRC = crc32(data) & 0xffffffff # CRC-32 checksum | |
1166 if zinfo.compress_type == ZIP_DEFLATED: | |
1167 co = zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION, | |
1168 zlib.DEFLATED, -15) | |
1169 data = co.compress(data) + co.flush() | |
1170 zinfo.compress_size = len(data) # Compressed size | |
1171 elif zinfo.compress_type == ZIP_BZIP2: | |
1172 co = bz2.BZ2Compressor() | |
1173 data = co.compress(data) + co.flush() | |
1174 zinfo.compress_size = len(data) # Compressed size | |
1175 else: | |
1176 zinfo.compress_size = zinfo.file_size | |
1177 zinfo.header_offset = self.fp.tell() # Start of header data | |
1178 self.fp.write(zinfo.FileHeader()) | |
1179 self.fp.write(data) | |
1180 self.fp.flush() | |
1181 if zinfo.flag_bits & 0x08: | |
1182 # Write CRC and file sizes after the file data | |
1183 self.fp.write(struct.pack("<LLL", zinfo.CRC, zinfo.compress_size, | |
1184 zinfo.file_size)) | |
1185 self.filelist.append(zinfo) | |
1186 self.NameToInfo[zinfo.filename] = zinfo | |
1187 | |
1188 def __del__(self): | |
1189 """Call the "close()" method in case the user forgot.""" | |
1190 self.close() | |
1191 | |
1192 def close(self): | |
1193 """Close the file, and for mode "w" and "a" write the ending | |
1194 records.""" | |
1195 if self.fp is None: | |
1196 return | |
1197 | |
1198 if self.mode in ("w", "a") and self._didModify: # write ending records | |
1199 count = 0 | |
1200 pos1 = self.fp.tell() | |
1201 for zinfo in self.filelist: # write central directory | |
1202 count = count + 1 | |
1203 dt = zinfo.date_time | |
1204 dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2] | |
1205 dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2) | |
1206 extra = [] | |
1207 if zinfo.file_size > ZIP64_LIMIT \ | |
1208 or zinfo.compress_size > ZIP64_LIMIT: | |
1209 extra.append(zinfo.file_size) | |
1210 extra.append(zinfo.compress_size) | |
1211 file_size = 0xffffffff | |
1212 compress_size = 0xffffffff | |
1213 else: | |
1214 file_size = zinfo.file_size | |
1215 compress_size = zinfo.compress_size | |
1216 | |
1217 if zinfo.header_offset > ZIP64_LIMIT: | |
1218 extra.append(zinfo.header_offset) | |
1219 header_offset = 0xffffffff | |
1220 else: | |
1221 header_offset = zinfo.header_offset | |
1222 | |
1223 extra_data = zinfo.extra | |
1224 if extra: | |
1225 # Append a ZIP64 field to the extra's | |
1226 extra_data = struct.pack( | |
1227 '<HH' + 'Q'*len(extra), | |
1228 1, 8*len(extra), *extra) + extra_data | |
1229 | |
1230 extract_version = max(45, zinfo.extract_version) | |
1231 create_version = max(45, zinfo.create_version) | |
1232 else: | |
1233 extract_version = zinfo.extract_version | |
1234 create_version = zinfo.create_version | |
1235 | |
1236 try: | |
1237 filename, flag_bits = zinfo._encodeFilenameFlags() | |
1238 centdir = struct.pack(structCentralDir, | |
1239 stringCentralDir, create_version, | |
1240 zinfo.create_system, extract_version, zinfo.reserved, | |
1241 flag_bits, zinfo.compress_type, dostime, dosdate, | |
1242 zinfo.CRC, compress_size, file_size, | |
1243 len(filename), len(extra_data), len(zinfo.comment), | |
1244 0, zinfo.internal_attr, zinfo.external_attr, | |
1245 header_offset) | |
1246 except DeprecationWarning: | |
1247 print((structCentralDir, stringCentralDir, create_version, | |
1248 zinfo.create_system, extract_version, zinfo.reserved, | |
1249 zinfo.flag_bits, zinfo.compress_type, dostime, dosdate, | |
1250 zinfo.CRC, compress_size, file_size, | |
1251 len(zinfo.filename), len(extra_data), len(zinfo.comment), | |
1252 0, zinfo.internal_attr, zinfo.external_attr, | |
1253 header_offset), file=sys.stderr) | |
1254 raise | |
1255 self.fp.write(centdir) | |
1256 self.fp.write(filename) | |
1257 self.fp.write(extra_data) | |
1258 self.fp.write(zinfo.comment) | |
1259 | |
1260 pos2 = self.fp.tell() | |
1261 # Write end-of-zip-archive record | |
1262 centDirCount = count | |
1263 centDirSize = pos2 - pos1 | |
1264 centDirOffset = pos1 | |
1265 if (centDirCount >= ZIP_FILECOUNT_LIMIT or | |
1266 centDirOffset > ZIP64_LIMIT or | |
1267 centDirSize > ZIP64_LIMIT): | |
1268 # Need to write the ZIP64 end-of-archive records | |
1269 zip64endrec = struct.pack( | |
1270 structEndArchive64, stringEndArchive64, | |
1271 44, 45, 45, 0, 0, centDirCount, centDirCount, | |
1272 centDirSize, centDirOffset) | |
1273 self.fp.write(zip64endrec) | |
1274 | |
1275 zip64locrec = struct.pack( | |
1276 structEndArchive64Locator, | |
1277 stringEndArchive64Locator, 0, pos2, 1) | |
1278 self.fp.write(zip64locrec) | |
1279 centDirCount = min(centDirCount, 0xFFFF) | |
1280 centDirSize = min(centDirSize, 0xFFFFFFFF) | |
1281 centDirOffset = min(centDirOffset, 0xFFFFFFFF) | |
1282 | |
1283 # check for valid comment length | |
1284 if len(self.comment) >= ZIP_MAX_COMMENT: | |
1285 if self.debug > 0: | |
1286 msg = 'Archive comment is too long; truncating to %d bytes' \ | |
1287 % ZIP_MAX_COMMENT | |
1288 self.comment = self.comment[:ZIP_MAX_COMMENT] | |
1289 | |
1290 endrec = struct.pack(structEndArchive, stringEndArchive, | |
1291 0, 0, centDirCount, centDirCount, | |
1292 centDirSize, centDirOffset, len(self.comment)) | |
1293 self.fp.write(endrec) | |
1294 self.fp.write(self.comment) | |
1295 self.fp.flush() | |
1296 | |
1297 if not self._filePassed: | |
1298 self.fp.close() | |
1299 self.fp = None | |
1300 | |
1301 | |
1302 class PyZipFile(ZipFile): | |
1303 """Class to create ZIP archives with Python library files and packages.""" | |
1304 | |
1305 def writepy(self, pathname, basename=""): | |
1306 """Add all files from "pathname" to the ZIP archive. | |
1307 | |
1308 If pathname is a package directory, search the directory and | |
1309 all package subdirectories recursively for all *.py and enter | |
1310 the modules into the archive. If pathname is a plain | |
1311 directory, listdir *.py and enter all modules. Else, pathname | |
1312 must be a Python *.py file and the module will be put into the | |
1313 archive. Added modules are always module.pyo or module.pyc. | |
1314 This method will compile the module.py into module.pyc if | |
1315 necessary. | |
1316 """ | |
1317 dir, name = os.path.split(pathname) | |
1318 if os.path.isdir(pathname): | |
1319 initname = os.path.join(pathname, "__init__.py") | |
1320 if os.path.isfile(initname): | |
1321 # This is a package directory, add it | |
1322 if basename: | |
1323 basename = "%s/%s" % (basename, name) | |
1324 else: | |
1325 basename = name | |
1326 if self.debug: | |
1327 print("Adding package in", pathname, "as", basename) | |
1328 fname, arcname = self._get_codename(initname[0:-3], basename) | |
1329 if self.debug: | |
1330 print("Adding", arcname) | |
1331 self.write(fname, arcname) | |
1332 dirlist = os.listdir(pathname) | |
1333 dirlist.remove("__init__.py") | |
1334 # Add all *.py files and package subdirectories | |
1335 for filename in dirlist: | |
1336 path = os.path.join(pathname, filename) | |
1337 root, ext = os.path.splitext(filename) | |
1338 if os.path.isdir(path): | |
1339 if os.path.isfile(os.path.join(path, "__init__.py")): | |
1340 # This is a package directory, add it | |
1341 self.writepy(path, basename) # Recursive call | |
1342 elif ext == ".py": | |
1343 fname, arcname = self._get_codename(path[0:-3], | |
1344 basename) | |
1345 if self.debug: | |
1346 print("Adding", arcname) | |
1347 self.write(fname, arcname) | |
1348 else: | |
1349 # This is NOT a package directory, add its files at top level | |
1350 if self.debug: | |
1351 print("Adding files from directory", pathname) | |
1352 for filename in os.listdir(pathname): | |
1353 path = os.path.join(pathname, filename) | |
1354 root, ext = os.path.splitext(filename) | |
1355 if ext == ".py": | |
1356 fname, arcname = self._get_codename(path[0:-3], | |
1357 basename) | |
1358 if self.debug: | |
1359 print("Adding", arcname) | |
1360 self.write(fname, arcname) | |
1361 else: | |
1362 if pathname[-3:] != ".py": | |
1363 raise RuntimeError( | |
1364 'Files added with writepy() must end with ".py"') | |
1365 fname, arcname = self._get_codename(pathname[0:-3], basename) | |
1366 if self.debug: | |
1367 print("Adding file", arcname) | |
1368 self.write(fname, arcname) | |
1369 | |
1370 def _get_codename(self, pathname, basename): | |
1371 """Return (filename, archivename) for the path. | |
1372 | |
1373 Given a module name path, return the correct file path and | |
1374 archive name, compiling if necessary. For example, given | |
1375 /python/lib/string, return (/python/lib/string.pyc, string). | |
1376 """ | |
1377 file_py = pathname + ".py" | |
1378 file_pyc = pathname + ".pyc" | |
1379 file_pyo = pathname + ".pyo" | |
1380 if os.path.isfile(file_pyo) and \ | |
1381 os.stat(file_pyo).st_mtime >= os.stat(file_py).st_mtime: | |
1382 fname = file_pyo # Use .pyo file | |
1383 elif not os.path.isfile(file_pyc) or \ | |
1384 os.stat(file_pyc).st_mtime < os.stat(file_py).st_mtime: | |
1385 import py_compile | |
1386 if self.debug: | |
1387 print("Compiling", file_py) | |
1388 try: | |
1389 py_compile.compile(file_py, file_pyc, None, True) | |
1390 except py_compile.PyCompileError as err: | |
1391 print(err.msg) | |
1392 fname = file_pyc | |
1393 else: | |
1394 fname = file_pyc | |
1395 archivename = os.path.split(fname)[1] | |
1396 if basename: | |
1397 archivename = "%s/%s" % (basename, archivename) | |
1398 return (fname, archivename) | |
1399 | |
1400 | |
1401 def main(args = None): | |
1402 import textwrap | |
1403 USAGE=textwrap.dedent("""\ | |
1404 Usage: | |
1405 zipfile.py -l zipfile.zip # Show listing of a zipfile | |
1406 zipfile.py -t zipfile.zip # Test if a zipfile is valid | |
1407 zipfile.py -e zipfile.zip target # Extract zipfile into target dir | |
1408 zipfile.py -c zipfile.zip src ... # Create zipfile from sources | |
1409 """) | |
1410 if args is None: | |
1411 args = sys.argv[1:] | |
1412 | |
1413 if not args or args[0] not in ('-l', '-c', '-e', '-t'): | |
1414 print(USAGE) | |
1415 sys.exit(1) | |
1416 | |
1417 if args[0] == '-l': | |
1418 if len(args) != 2: | |
1419 print(USAGE) | |
1420 sys.exit(1) | |
1421 zf = ZipFile(args[1], 'r') | |
1422 zf.printdir() | |
1423 zf.close() | |
1424 | |
1425 elif args[0] == '-t': | |
1426 if len(args) != 2: | |
1427 print(USAGE) | |
1428 sys.exit(1) | |
1429 zf = ZipFile(args[1], 'r') | |
29
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
1430 badfile = zf.testzip() |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
1431 if badfile: |
a8cc383b787c
Clean up zipfiles and diff them to stock ones
Oleg Oshmyan <chortos@inbox.lv>
parents:
28
diff
changeset
|
1432 print("The following enclosed file is corrupted: {!r}".format(badfile)) |
21 | 1433 print("Done testing") |
1434 | |
1435 elif args[0] == '-e': | |
1436 if len(args) != 3: | |
1437 print(USAGE) | |
1438 sys.exit(1) | |
1439 | |
1440 zf = ZipFile(args[1], 'r') | |
1441 out = args[2] | |
1442 for path in zf.namelist(): | |
1443 if path.startswith('./'): | |
1444 tgt = os.path.join(out, path[2:]) | |
1445 else: | |
1446 tgt = os.path.join(out, path) | |
1447 | |
1448 tgtdir = os.path.dirname(tgt) | |
1449 if not os.path.exists(tgtdir): | |
1450 os.makedirs(tgtdir) | |
1451 with open(tgt, 'wb') as fp: | |
1452 fp.write(zf.read(path)) | |
1453 zf.close() | |
1454 | |
1455 elif args[0] == '-c': | |
1456 if len(args) < 3: | |
1457 print(USAGE) | |
1458 sys.exit(1) | |
1459 | |
1460 def addToZip(zf, path, zippath): | |
1461 if os.path.isfile(path): | |
1462 zf.write(path, zippath, ZIP_DEFLATED) | |
1463 elif os.path.isdir(path): | |
1464 for nm in os.listdir(path): | |
1465 addToZip(zf, | |
1466 os.path.join(path, nm), os.path.join(zippath, nm)) | |
1467 # else: ignore | |
1468 | |
1469 zf = ZipFile(args[1], 'w', allowZip64=True) | |
1470 for src in args[2:]: | |
1471 addToZip(zf, src, os.path.basename(src)) | |
1472 | |
1473 zf.close() | |
1474 | |
1475 if __name__ == "__main__": | |
1476 main() |