|
21
|
1 """
|
|
|
2 Read and write ZIP files.
|
|
|
3 """
|
|
|
4 # Improved by Chortos-2 in 2010 (added bzip2 support)
|
|
|
5 import struct, os, time, sys, shutil
|
|
|
6 import binascii, cStringIO, stat
|
|
|
7 import io
|
|
|
8 import re
|
|
|
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 = "<4s4H2LH"
|
|
|
56 stringEndArchive = "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 = "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 = "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 = "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 = "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:] == "\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("")
|
|
|
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 = "" # Comment for each file
|
|
|
299 self.extra = "" # 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 if isinstance(self.filename, unicode):
|
|
|
354 try:
|
|
|
355 return self.filename.encode('ascii'), self.flag_bits
|
|
|
356 except UnicodeEncodeError:
|
|
|
357 return self.filename.encode('utf-8'), self.flag_bits | 0x800
|
|
|
358 else:
|
|
|
359 return self.filename, self.flag_bits
|
|
|
360
|
|
|
361 def _decodeFilename(self):
|
|
|
362 if self.flag_bits & 0x800:
|
|
|
363 return self.filename.decode('utf-8')
|
|
|
364 else:
|
|
|
365 return self.filename
|
|
|
366
|
|
|
367 def _decodeExtra(self):
|
|
|
368 # Try to decode the extra field.
|
|
|
369 extra = self.extra
|
|
|
370 unpack = struct.unpack
|
|
|
371 while extra:
|
|
|
372 tp, ln = unpack('<HH', extra[:4])
|
|
|
373 if tp == 1:
|
|
|
374 if ln >= 24:
|
|
|
375 counts = unpack('<QQQ', extra[4:28])
|
|
|
376 elif ln == 16:
|
|
|
377 counts = unpack('<QQ', extra[4:20])
|
|
|
378 elif ln == 8:
|
|
|
379 counts = unpack('<Q', extra[4:12])
|
|
|
380 elif ln == 0:
|
|
|
381 counts = ()
|
|
|
382 else:
|
|
|
383 raise RuntimeError, "Corrupt extra field %s"%(ln,)
|
|
|
384
|
|
|
385 idx = 0
|
|
|
386
|
|
|
387 # ZIP64 extension (large files and/or large archives)
|
|
|
388 if self.file_size in (0xffffffffffffffffL, 0xffffffffL):
|
|
|
389 self.file_size = counts[idx]
|
|
|
390 idx += 1
|
|
|
391
|
|
|
392 if self.compress_size == 0xFFFFFFFFL:
|
|
|
393 self.compress_size = counts[idx]
|
|
|
394 idx += 1
|
|
|
395
|
|
|
396 if self.header_offset == 0xffffffffL:
|
|
|
397 old = self.header_offset
|
|
|
398 self.header_offset = counts[idx]
|
|
|
399 idx+=1
|
|
|
400
|
|
|
401 extra = extra[ln+4:]
|
|
|
402
|
|
|
403
|
|
|
404 class _ZipDecrypter:
|
|
|
405 """Class to handle decryption of files stored within a ZIP archive.
|
|
|
406
|
|
|
407 ZIP supports a password-based form of encryption. Even though known
|
|
|
408 plaintext attacks have been found against it, it is still useful
|
|
|
409 to be able to get data out of such a file.
|
|
|
410
|
|
|
411 Usage:
|
|
|
412 zd = _ZipDecrypter(mypwd)
|
|
|
413 plain_char = zd(cypher_char)
|
|
|
414 plain_text = map(zd, cypher_text)
|
|
|
415 """
|
|
|
416
|
|
|
417 def _GenerateCRCTable():
|
|
|
418 """Generate a CRC-32 table.
|
|
|
419
|
|
|
420 ZIP encryption uses the CRC32 one-byte primitive for scrambling some
|
|
|
421 internal keys. We noticed that a direct implementation is faster than
|
|
|
422 relying on binascii.crc32().
|
|
|
423 """
|
|
|
424 poly = 0xedb88320
|
|
|
425 table = [0] * 256
|
|
|
426 for i in range(256):
|
|
|
427 crc = i
|
|
|
428 for j in range(8):
|
|
|
429 if crc & 1:
|
|
|
430 crc = ((crc >> 1) & 0x7FFFFFFF) ^ poly
|
|
|
431 else:
|
|
|
432 crc = ((crc >> 1) & 0x7FFFFFFF)
|
|
|
433 table[i] = crc
|
|
|
434 return table
|
|
|
435 crctable = _GenerateCRCTable()
|
|
|
436
|
|
|
437 def _crc32(self, ch, crc):
|
|
|
438 """Compute the CRC32 primitive on one byte."""
|
|
|
439 return ((crc >> 8) & 0xffffff) ^ self.crctable[(crc ^ ord(ch)) & 0xff]
|
|
|
440
|
|
|
441 def __init__(self, pwd):
|
|
|
442 self.key0 = 305419896
|
|
|
443 self.key1 = 591751049
|
|
|
444 self.key2 = 878082192
|
|
|
445 for p in pwd:
|
|
|
446 self._UpdateKeys(p)
|
|
|
447
|
|
|
448 def _UpdateKeys(self, c):
|
|
|
449 self.key0 = self._crc32(c, self.key0)
|
|
|
450 self.key1 = (self.key1 + (self.key0 & 255)) & 4294967295
|
|
|
451 self.key1 = (self.key1 * 134775813 + 1) & 4294967295
|
|
|
452 self.key2 = self._crc32(chr((self.key1 >> 24) & 255), self.key2)
|
|
|
453
|
|
|
454 def __call__(self, c):
|
|
|
455 """Decrypt a single character."""
|
|
|
456 c = ord(c)
|
|
|
457 k = self.key2 | 2
|
|
|
458 c = c ^ (((k * (k^1)) >> 8) & 255)
|
|
|
459 c = chr(c)
|
|
|
460 self._UpdateKeys(c)
|
|
|
461 return c
|
|
|
462
|
|
|
463 class ZipExtFile(io.BufferedIOBase):
|
|
|
464 """File-like object for reading an archive member.
|
|
|
465 Is returned by ZipFile.open().
|
|
|
466 """
|
|
|
467
|
|
|
468 # Max size supported by decompressor.
|
|
|
469 MAX_N = 1 << 31 - 1
|
|
|
470
|
|
|
471 # Read from compressed files in 4k blocks.
|
|
|
472 MIN_READ_SIZE = 4096
|
|
|
473
|
|
|
474 # Search for universal newlines or line chunks.
|
|
|
475 PATTERN = re.compile(r'^(?P<chunk>[^\r\n]+)|(?P<newline>\n|\r\n?)')
|
|
|
476
|
|
|
477 def __init__(self, fileobj, mode, zipinfo, decrypter=None):
|
|
|
478 self._fileobj = fileobj
|
|
|
479 self._decrypter = decrypter
|
|
|
480
|
|
|
481 self._compress_type = zipinfo.compress_type
|
|
|
482 self._compress_size = zipinfo.compress_size
|
|
|
483 self._compress_left = zipinfo.compress_size
|
|
|
484
|
|
|
485 if self._compress_type == ZIP_DEFLATED:
|
|
|
486 self._decompressor = zlib.decompressobj(-15)
|
|
|
487 elif self._compress_type == ZIP_BZIP2:
|
|
|
488 self._decompressor = bz2.BZ2Decompressor()
|
|
|
489 self.MIN_READ_SIZE = 900000
|
|
|
490 self._unconsumed = ''
|
|
|
491
|
|
|
492 self._readbuffer = ''
|
|
|
493 self._offset = 0
|
|
|
494
|
|
|
495 self._universal = 'U' in mode
|
|
|
496 self.newlines = None
|
|
|
497
|
|
|
498 # Adjust read size for encrypted files since the first 12 bytes
|
|
|
499 # are for the encryption/password information.
|
|
|
500 if self._decrypter is not None:
|
|
|
501 self._compress_left -= 12
|
|
|
502
|
|
|
503 self.mode = mode
|
|
|
504 self.name = zipinfo.filename
|
|
|
505
|
|
|
506 def readline(self, limit=-1):
|
|
|
507 """Read and return a line from the stream.
|
|
|
508
|
|
|
509 If limit is specified, at most limit bytes will be read.
|
|
|
510 """
|
|
|
511
|
|
|
512 if not self._universal and limit < 0:
|
|
|
513 # Shortcut common case - newline found in buffer.
|
|
|
514 i = self._readbuffer.find('\n', self._offset) + 1
|
|
|
515 if i > 0:
|
|
|
516 line = self._readbuffer[self._offset: i]
|
|
|
517 self._offset = i
|
|
|
518 return line
|
|
|
519
|
|
|
520 if not self._universal:
|
|
|
521 return io.BufferedIOBase.readline(self, limit)
|
|
|
522
|
|
|
523 line = ''
|
|
|
524 while limit < 0 or len(line) < limit:
|
|
|
525 readahead = self.peek(2)
|
|
|
526 if readahead == '':
|
|
|
527 return line
|
|
|
528
|
|
|
529 #
|
|
|
530 # Search for universal newlines or line chunks.
|
|
|
531 #
|
|
|
532 # The pattern returns either a line chunk or a newline, but not
|
|
|
533 # both. Combined with peek(2), we are assured that the sequence
|
|
|
534 # '\r\n' is always retrieved completely and never split into
|
|
|
535 # separate newlines - '\r', '\n' due to coincidental readaheads.
|
|
|
536 #
|
|
|
537 match = self.PATTERN.search(readahead)
|
|
|
538 newline = match.group('newline')
|
|
|
539 if newline is not None:
|
|
|
540 if self.newlines is None:
|
|
|
541 self.newlines = []
|
|
|
542 if newline not in self.newlines:
|
|
|
543 self.newlines.append(newline)
|
|
|
544 self._offset += len(newline)
|
|
|
545 return line + '\n'
|
|
|
546
|
|
|
547 chunk = match.group('chunk')
|
|
|
548 if limit >= 0:
|
|
|
549 chunk = chunk[: limit - len(line)]
|
|
|
550
|
|
|
551 self._offset += len(chunk)
|
|
|
552 line += chunk
|
|
|
553
|
|
|
554 return line
|
|
|
555
|
|
|
556 def peek(self, n=1):
|
|
|
557 """Returns buffered bytes without advancing the position."""
|
|
|
558 if n > len(self._readbuffer) - self._offset:
|
|
|
559 chunk = self.read(n)
|
|
|
560 self._offset -= len(chunk)
|
|
|
561
|
|
|
562 # Return up to 512 bytes to reduce allocation overhead for tight loops.
|
|
|
563 return self._readbuffer[self._offset: self._offset + 512]
|
|
|
564
|
|
|
565 def readable(self):
|
|
|
566 return True
|
|
|
567
|
|
|
568 def read(self, n=-1):
|
|
|
569 """Read and return up to n bytes.
|
|
|
570 If the argument is omitted, None, or negative, data is read and returned until EOF is reached..
|
|
|
571 """
|
|
|
572
|
|
|
573 buf = ''
|
|
|
574 while n < 0 or n is None or n > len(buf):
|
|
|
575 data = self.read1(n)
|
|
|
576 if len(data) == 0:
|
|
|
577 return buf
|
|
|
578
|
|
|
579 buf += data
|
|
|
580
|
|
|
581 return buf
|
|
|
582
|
|
|
583 def read1(self, n):
|
|
|
584 """Read up to n bytes with at most one read() system call."""
|
|
|
585
|
|
|
586 # Simplify algorithm (branching) by transforming negative n to large n.
|
|
|
587 if n < 0 or n is None:
|
|
|
588 n = self.MAX_N
|
|
|
589
|
|
|
590 # Bytes available in read buffer.
|
|
|
591 len_readbuffer = len(self._readbuffer) - self._offset
|
|
|
592
|
|
|
593 # Read from file.
|
|
|
594 if self._compress_left > 0 and n > len_readbuffer + len(self._unconsumed):
|
|
|
595 nbytes = n - len_readbuffer - len(self._unconsumed)
|
|
|
596 nbytes = max(nbytes, self.MIN_READ_SIZE)
|
|
|
597 nbytes = min(nbytes, self._compress_left)
|
|
|
598
|
|
|
599 data = self._fileobj.read(nbytes)
|
|
|
600 self._compress_left -= len(data)
|
|
|
601
|
|
|
602 if data and self._decrypter is not None:
|
|
|
603 data = ''.join(map(self._decrypter, data))
|
|
|
604
|
|
|
605 if self._compress_type == ZIP_STORED:
|
|
|
606 self._readbuffer = self._readbuffer[self._offset:] + data
|
|
|
607 self._offset = 0
|
|
|
608 else:
|
|
|
609 # Prepare deflated bytes for decompression.
|
|
|
610 self._unconsumed += data
|
|
|
611
|
|
|
612 # Handle unconsumed data.
|
|
|
613 if (len(self._unconsumed) > 0 and n > len_readbuffer and
|
|
|
614 self._compress_type == ZIP_DEFLATED):
|
|
|
615 data = self._decompressor.decompress(
|
|
|
616 self._unconsumed,
|
|
|
617 max(n - len_readbuffer, self.MIN_READ_SIZE)
|
|
|
618 )
|
|
|
619
|
|
|
620 self._unconsumed = self._decompressor.unconsumed_tail
|
|
|
621 if len(self._unconsumed) == 0 and self._compress_left == 0:
|
|
|
622 data += self._decompressor.flush()
|
|
|
623
|
|
|
624 self._readbuffer = self._readbuffer[self._offset:] + data
|
|
|
625 self._offset = 0
|
|
|
626 elif (len(self._unconsumed) > 0 and n > len_readbuffer and
|
|
|
627 self._compress_type == ZIP_BZIP2):
|
|
|
628 data = self._decompressor.decompress(self._unconsumed)
|
|
|
629
|
|
|
630 self._unconsumed = ''
|
|
|
631 self._readbuffer = self._readbuffer[self._offset:] + data
|
|
|
632 self._offset = 0
|
|
|
633
|
|
|
634 # Read from buffer.
|
|
|
635 data = self._readbuffer[self._offset: self._offset + n]
|
|
|
636 self._offset += len(data)
|
|
|
637 return data
|
|
|
638
|
|
|
639
|
|
|
640
|
|
|
641 class ZipFile:
|
|
|
642 """ Class with methods to open, read, write, close, list zip files.
|
|
|
643
|
|
|
644 z = ZipFile(file, mode="r", compression=ZIP_STORED, allowZip64=False)
|
|
|
645
|
|
|
646 file: Either the path to the file, or a file-like object.
|
|
|
647 If it is a path, the file will be opened and closed by ZipFile.
|
|
|
648 mode: The mode can be either read "r", write "w" or append "a".
|
|
|
649 compression: ZIP_STORED (no compression), ZIP_DEFLATED (requires zlib),
|
|
|
650 or ZIP_BZIP2 (requires bz2).
|
|
|
651 allowZip64: if True ZipFile will create files with ZIP64 extensions when
|
|
|
652 needed, otherwise it will raise an exception when this would
|
|
|
653 be necessary.
|
|
|
654
|
|
|
655 """
|
|
|
656
|
|
|
657 fp = None # Set here since __del__ checks it
|
|
|
658
|
|
|
659 def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=False):
|
|
|
660 """Open the ZIP file with mode read "r", write "w" or append "a"."""
|
|
|
661 if mode not in ("r", "w", "a"):
|
|
|
662 raise RuntimeError('ZipFile() requires mode "r", "w", or "a"')
|
|
|
663
|
|
|
664 if compression == ZIP_STORED:
|
|
|
665 pass
|
|
|
666 elif compression == ZIP_DEFLATED:
|
|
|
667 if not zlib:
|
|
|
668 raise RuntimeError,\
|
|
|
669 "Compression requires the (missing) zlib module"
|
|
|
670 elif compression == ZIP_BZIP2:
|
|
|
671 if not bz2:
|
|
|
672 raise RuntimeError,\
|
|
|
673 "Compression requires the (missing) bz2 module"
|
|
|
674 else:
|
|
|
675 raise RuntimeError, "That compression method is not supported"
|
|
|
676
|
|
|
677 self._allowZip64 = allowZip64
|
|
|
678 self._didModify = False
|
|
|
679 self.debug = 0 # Level of printing: 0 through 3
|
|
|
680 self.NameToInfo = {} # Find file info given name
|
|
|
681 self.filelist = [] # List of ZipInfo instances for archive
|
|
|
682 self.compression = compression # Method of compression
|
|
|
683 self.mode = key = mode.replace('b', '')[0]
|
|
|
684 self.pwd = None
|
|
|
685 self.comment = ''
|
|
|
686
|
|
|
687 # Check if we were passed a file-like object
|
|
|
688 if isinstance(file, basestring):
|
|
|
689 self._filePassed = 0
|
|
|
690 self.filename = file
|
|
|
691 modeDict = {'r' : 'rb', 'w': 'wb', 'a' : 'r+b'}
|
|
|
692 try:
|
|
|
693 self.fp = open(file, modeDict[mode])
|
|
|
694 except IOError:
|
|
|
695 if mode == 'a':
|
|
|
696 mode = key = 'w'
|
|
|
697 self.fp = open(file, modeDict[mode])
|
|
|
698 else:
|
|
|
699 raise
|
|
|
700 else:
|
|
|
701 self._filePassed = 1
|
|
|
702 self.fp = file
|
|
|
703 self.filename = getattr(file, 'name', None)
|
|
|
704
|
|
|
705 if key == 'r':
|
|
|
706 self._GetContents()
|
|
|
707 elif key == 'w':
|
|
|
708 pass
|
|
|
709 elif key == 'a':
|
|
|
710 try: # See if file is a zip file
|
|
|
711 self._RealGetContents()
|
|
|
712 # seek to start of directory and overwrite
|
|
|
713 self.fp.seek(self.start_dir, 0)
|
|
|
714 except BadZipfile: # file is not a zip file, just append
|
|
|
715 self.fp.seek(0, 2)
|
|
|
716 else:
|
|
|
717 if not self._filePassed:
|
|
|
718 self.fp.close()
|
|
|
719 self.fp = None
|
|
|
720 raise RuntimeError, 'Mode must be "r", "w" or "a"'
|
|
|
721
|
|
|
722 def __enter__(self):
|
|
|
723 return self
|
|
|
724
|
|
|
725 def __exit__(self, type, value, traceback):
|
|
|
726 self.close()
|
|
|
727
|
|
|
728 def _GetContents(self):
|
|
|
729 """Read the directory, making sure we close the file if the format
|
|
|
730 is bad."""
|
|
|
731 try:
|
|
|
732 self._RealGetContents()
|
|
|
733 except BadZipfile:
|
|
|
734 if not self._filePassed:
|
|
|
735 self.fp.close()
|
|
|
736 self.fp = None
|
|
|
737 raise
|
|
|
738
|
|
|
739 def _RealGetContents(self):
|
|
|
740 """Read in the table of contents for the ZIP file."""
|
|
|
741 fp = self.fp
|
|
|
742 endrec = _EndRecData(fp)
|
|
|
743 if not endrec:
|
|
|
744 raise BadZipfile, "File is not a zip file"
|
|
|
745 if self.debug > 1:
|
|
|
746 print endrec
|
|
|
747 size_cd = endrec[_ECD_SIZE] # bytes in central directory
|
|
|
748 offset_cd = endrec[_ECD_OFFSET] # offset of central directory
|
|
|
749 self.comment = endrec[_ECD_COMMENT] # archive comment
|
|
|
750
|
|
|
751 # "concat" is zero, unless zip was concatenated to another file
|
|
|
752 concat = endrec[_ECD_LOCATION] - size_cd - offset_cd
|
|
|
753 if endrec[_ECD_SIGNATURE] == stringEndArchive64:
|
|
|
754 # If Zip64 extension structures are present, account for them
|
|
|
755 concat -= (sizeEndCentDir64 + sizeEndCentDir64Locator)
|
|
|
756
|
|
|
757 if self.debug > 2:
|
|
|
758 inferred = concat + offset_cd
|
|
|
759 print "given, inferred, offset", offset_cd, inferred, concat
|
|
|
760 # self.start_dir: Position of start of central directory
|
|
|
761 self.start_dir = offset_cd + concat
|
|
|
762 fp.seek(self.start_dir, 0)
|
|
|
763 data = fp.read(size_cd)
|
|
|
764 fp = cStringIO.StringIO(data)
|
|
|
765 total = 0
|
|
|
766 while total < size_cd:
|
|
|
767 centdir = fp.read(sizeCentralDir)
|
|
|
768 if centdir[0:4] != stringCentralDir:
|
|
|
769 raise BadZipfile, "Bad magic number for central directory"
|
|
|
770 centdir = struct.unpack(structCentralDir, centdir)
|
|
|
771 if self.debug > 2:
|
|
|
772 print centdir
|
|
|
773 filename = fp.read(centdir[_CD_FILENAME_LENGTH])
|
|
|
774 # Create ZipInfo instance to store file information
|
|
|
775 x = ZipInfo(filename)
|
|
|
776 x.extra = fp.read(centdir[_CD_EXTRA_FIELD_LENGTH])
|
|
|
777 x.comment = fp.read(centdir[_CD_COMMENT_LENGTH])
|
|
|
778 x.header_offset = centdir[_CD_LOCAL_HEADER_OFFSET]
|
|
|
779 (x.create_version, x.create_system, x.extract_version, x.reserved,
|
|
|
780 x.flag_bits, x.compress_type, t, d,
|
|
|
781 x.CRC, x.compress_size, x.file_size) = centdir[1:12]
|
|
|
782 x.volume, x.internal_attr, x.external_attr = centdir[15:18]
|
|
|
783 # Convert date/time code to (year, month, day, hour, min, sec)
|
|
|
784 x._raw_time = t
|
|
|
785 x.date_time = ( (d>>9)+1980, (d>>5)&0xF, d&0x1F,
|
|
|
786 t>>11, (t>>5)&0x3F, (t&0x1F) * 2 )
|
|
|
787
|
|
|
788 x._decodeExtra()
|
|
|
789 x.header_offset = x.header_offset + concat
|
|
|
790 x.filename = x._decodeFilename()
|
|
|
791 self.filelist.append(x)
|
|
|
792 self.NameToInfo[x.filename] = x
|
|
|
793
|
|
|
794 # update total bytes read from central directory
|
|
|
795 total = (total + sizeCentralDir + centdir[_CD_FILENAME_LENGTH]
|
|
|
796 + centdir[_CD_EXTRA_FIELD_LENGTH]
|
|
|
797 + centdir[_CD_COMMENT_LENGTH])
|
|
|
798
|
|
|
799 if self.debug > 2:
|
|
|
800 print "total", total
|
|
|
801
|
|
|
802
|
|
|
803 def namelist(self):
|
|
|
804 """Return a list of file names in the archive."""
|
|
|
805 l = []
|
|
|
806 for data in self.filelist:
|
|
|
807 l.append(data.filename)
|
|
|
808 return l
|
|
|
809
|
|
|
810 def infolist(self):
|
|
|
811 """Return a list of class ZipInfo instances for files in the
|
|
|
812 archive."""
|
|
|
813 return self.filelist
|
|
|
814
|
|
|
815 def printdir(self):
|
|
|
816 """Print a table of contents for the zip file."""
|
|
|
817 print "%-46s %19s %12s" % ("File Name", "Modified ", "Size")
|
|
|
818 for zinfo in self.filelist:
|
|
|
819 date = "%d-%02d-%02d %02d:%02d:%02d" % zinfo.date_time[:6]
|
|
|
820 print "%-46s %s %12d" % (zinfo.filename, date, zinfo.file_size)
|
|
|
821
|
|
|
822 def testzip(self):
|
|
|
823 """Read all the files and check the CRC."""
|
|
|
824 chunk_size = 2 ** 20
|
|
|
825 for zinfo in self.filelist:
|
|
|
826 try:
|
|
|
827 # Read by chunks, to avoid an OverflowError or a
|
|
|
828 # MemoryError with very large embedded files.
|
|
|
829 f = self.open(zinfo.filename, "r")
|
|
|
830 while f.read(chunk_size): # Check CRC-32
|
|
|
831 pass
|
|
|
832 except BadZipfile:
|
|
|
833 return zinfo.filename
|
|
|
834
|
|
|
835 def getinfo(self, name):
|
|
|
836 """Return the instance of ZipInfo given 'name'."""
|
|
|
837 info = self.NameToInfo.get(name)
|
|
|
838 if info is None:
|
|
|
839 raise KeyError(
|
|
|
840 'There is no item named %r in the archive' % name)
|
|
|
841
|
|
|
842 return info
|
|
|
843
|
|
|
844 def setpassword(self, pwd):
|
|
|
845 """Set default password for encrypted files."""
|
|
|
846 self.pwd = pwd
|
|
|
847
|
|
|
848 def read(self, name, pwd=None):
|
|
|
849 """Return file bytes (as a string) for name."""
|
|
|
850 return self.open(name, "r", pwd).read()
|
|
|
851
|
|
|
852 def open(self, name, mode="r", pwd=None):
|
|
|
853 """Return file-like object for 'name'."""
|
|
|
854 if mode not in ("r", "U", "rU"):
|
|
|
855 raise RuntimeError, 'open() requires mode "r", "U", or "rU"'
|
|
|
856 if not self.fp:
|
|
|
857 raise RuntimeError, \
|
|
|
858 "Attempt to read ZIP archive that was already closed"
|
|
|
859
|
|
|
860 # Only open a new file for instances where we were not
|
|
|
861 # given a file object in the constructor
|
|
|
862 if self._filePassed:
|
|
|
863 zef_file = self.fp
|
|
|
864 else:
|
|
|
865 zef_file = open(self.filename, 'rb')
|
|
|
866
|
|
|
867 # Make sure we have an info object
|
|
|
868 if isinstance(name, ZipInfo):
|
|
|
869 # 'name' is already an info object
|
|
|
870 zinfo = name
|
|
|
871 else:
|
|
|
872 # Get info object for name
|
|
|
873 zinfo = self.getinfo(name)
|
|
|
874
|
|
|
875 zef_file.seek(zinfo.header_offset, 0)
|
|
|
876
|
|
|
877 # Skip the file header:
|
|
|
878 fheader = zef_file.read(sizeFileHeader)
|
|
|
879 if fheader[0:4] != stringFileHeader:
|
|
|
880 raise BadZipfile, "Bad magic number for file header"
|
|
|
881
|
|
|
882 fheader = struct.unpack(structFileHeader, fheader)
|
|
|
883 fname = zef_file.read(fheader[_FH_FILENAME_LENGTH])
|
|
|
884 if fheader[_FH_EXTRA_FIELD_LENGTH]:
|
|
|
885 zef_file.read(fheader[_FH_EXTRA_FIELD_LENGTH])
|
|
|
886
|
|
|
887 if fname != zinfo.orig_filename:
|
|
|
888 raise BadZipfile, \
|
|
|
889 'File name in directory "%s" and header "%s" differ.' % (
|
|
|
890 zinfo.orig_filename, fname)
|
|
|
891
|
|
|
892 # check for encrypted flag & handle password
|
|
|
893 is_encrypted = zinfo.flag_bits & 0x1
|
|
|
894 zd = None
|
|
|
895 if is_encrypted:
|
|
|
896 if not pwd:
|
|
|
897 pwd = self.pwd
|
|
|
898 if not pwd:
|
|
|
899 raise RuntimeError, "File %s is encrypted, " \
|
|
|
900 "password required for extraction" % name
|
|
|
901
|
|
|
902 zd = _ZipDecrypter(pwd)
|
|
|
903 # The first 12 bytes in the cypher stream is an encryption header
|
|
|
904 # used to strengthen the algorithm. The first 11 bytes are
|
|
|
905 # completely random, while the 12th contains the MSB of the CRC,
|
|
|
906 # or the MSB of the file time depending on the header type
|
|
|
907 # and is used to check the correctness of the password.
|
|
|
908 bytes = zef_file.read(12)
|
|
|
909 h = map(zd, bytes[0:12])
|
|
|
910 if zinfo.flag_bits & 0x8:
|
|
|
911 # compare against the file type from extended local headers
|
|
|
912 check_byte = (zinfo._raw_time >> 8) & 0xff
|
|
|
913 else:
|
|
|
914 # compare against the CRC otherwise
|
|
|
915 check_byte = (zinfo.CRC >> 24) & 0xff
|
|
|
916 if ord(h[11]) != check_byte:
|
|
|
917 raise RuntimeError("Bad password for file", name)
|
|
|
918
|
|
|
919 return ZipExtFile(zef_file, mode, zinfo, zd)
|
|
|
920
|
|
|
921 def extract(self, member, path=None, pwd=None):
|
|
|
922 """Extract a member from the archive to the current working directory,
|
|
|
923 using its full name. Its file information is extracted as accurately
|
|
|
924 as possible. `member' may be a filename or a ZipInfo object. You can
|
|
|
925 specify a different directory using `path'.
|
|
|
926 """
|
|
|
927 if not isinstance(member, ZipInfo):
|
|
|
928 member = self.getinfo(member)
|
|
|
929
|
|
|
930 if path is None:
|
|
|
931 path = os.getcwd()
|
|
|
932
|
|
|
933 return self._extract_member(member, path, pwd)
|
|
|
934
|
|
|
935 def extractall(self, path=None, members=None, pwd=None):
|
|
|
936 """Extract all members from the archive to the current working
|
|
|
937 directory. `path' specifies a different directory to extract to.
|
|
|
938 `members' is optional and must be a subset of the list returned
|
|
|
939 by namelist().
|
|
|
940 """
|
|
|
941 if members is None:
|
|
|
942 members = self.namelist()
|
|
|
943
|
|
|
944 for zipinfo in members:
|
|
|
945 self.extract(zipinfo, path, pwd)
|
|
|
946
|
|
|
947 def _extract_member(self, member, targetpath, pwd):
|
|
|
948 """Extract the ZipInfo object 'member' to a physical
|
|
|
949 file on the path targetpath.
|
|
|
950 """
|
|
|
951 # build the destination pathname, replacing
|
|
|
952 # forward slashes to platform specific separators.
|
|
|
953 # Strip trailing path separator, unless it represents the root.
|
|
|
954 if (targetpath[-1:] in (os.path.sep, os.path.altsep)
|
|
|
955 and len(os.path.splitdrive(targetpath)[1]) > 1):
|
|
|
956 targetpath = targetpath[:-1]
|
|
|
957
|
|
|
958 # don't include leading "/" from file name if present
|
|
|
959 if member.filename[0] == '/':
|
|
|
960 targetpath = os.path.join(targetpath, member.filename[1:])
|
|
|
961 else:
|
|
|
962 targetpath = os.path.join(targetpath, member.filename)
|
|
|
963
|
|
|
964 targetpath = os.path.normpath(targetpath)
|
|
|
965
|
|
|
966 # Create all upper directories if necessary.
|
|
|
967 upperdirs = os.path.dirname(targetpath)
|
|
|
968 if upperdirs and not os.path.exists(upperdirs):
|
|
|
969 os.makedirs(upperdirs)
|
|
|
970
|
|
|
971 if member.filename[-1] == '/':
|
|
|
972 if not os.path.isdir(targetpath):
|
|
|
973 os.mkdir(targetpath)
|
|
|
974 return targetpath
|
|
|
975
|
|
|
976 source = self.open(member, pwd=pwd)
|
|
|
977 target = file(targetpath, "wb")
|
|
|
978 shutil.copyfileobj(source, target)
|
|
|
979 source.close()
|
|
|
980 target.close()
|
|
|
981
|
|
|
982 return targetpath
|
|
|
983
|
|
|
984 def _writecheck(self, zinfo):
|
|
|
985 """Check for errors before writing a file to the archive."""
|
|
|
986 if zinfo.filename in self.NameToInfo:
|
|
|
987 if self.debug: # Warning for duplicate names
|
|
|
988 print "Duplicate name:", zinfo.filename
|
|
|
989 if self.mode not in ("w", "a"):
|
|
|
990 raise RuntimeError, 'write() requires mode "w" or "a"'
|
|
|
991 if not self.fp:
|
|
|
992 raise RuntimeError, \
|
|
|
993 "Attempt to write ZIP archive that was already closed"
|
|
|
994 if zinfo.compress_type == ZIP_DEFLATED and not zlib:
|
|
|
995 raise RuntimeError, \
|
|
|
996 "Compression requires the (missing) zlib module"
|
|
|
997 if zinfo.compress_type == ZIP_BZIP2 and not bz2:
|
|
|
998 raise RuntimeError, \
|
|
|
999 "Compression requires the (missing) bz2 module"
|
|
|
1000 if zinfo.compress_type not in (ZIP_STORED, ZIP_DEFLATED, ZIP_BZIP2):
|
|
|
1001 raise RuntimeError, \
|
|
|
1002 "That compression method is not supported"
|
|
|
1003 if zinfo.file_size > ZIP64_LIMIT:
|
|
|
1004 if not self._allowZip64:
|
|
|
1005 raise LargeZipFile("Filesize would require ZIP64 extensions")
|
|
|
1006 if zinfo.header_offset > ZIP64_LIMIT:
|
|
|
1007 if not self._allowZip64:
|
|
|
1008 raise LargeZipFile("Zipfile size would require ZIP64 extensions")
|
|
|
1009
|
|
|
1010 def write(self, filename, arcname=None, compress_type=None):
|
|
|
1011 """Put the bytes from filename into the archive under the name
|
|
|
1012 arcname."""
|
|
|
1013 if not self.fp:
|
|
|
1014 raise RuntimeError(
|
|
|
1015 "Attempt to write to ZIP archive that was already closed")
|
|
|
1016
|
|
|
1017 st = os.stat(filename)
|
|
|
1018 isdir = stat.S_ISDIR(st.st_mode)
|
|
|
1019 mtime = time.localtime(st.st_mtime)
|
|
|
1020 date_time = mtime[0:6]
|
|
|
1021 # Create ZipInfo instance to store file information
|
|
|
1022 if arcname is None:
|
|
|
1023 arcname = filename
|
|
|
1024 arcname = os.path.normpath(os.path.splitdrive(arcname)[1])
|
|
|
1025 while arcname[0] in (os.sep, os.altsep):
|
|
|
1026 arcname = arcname[1:]
|
|
|
1027 if isdir:
|
|
|
1028 arcname += '/'
|
|
|
1029 zinfo = ZipInfo(arcname, date_time)
|
|
|
1030 zinfo.external_attr = (st[0] & 0xFFFF) << 16L # Unix attributes
|
|
|
1031 if compress_type is None:
|
|
|
1032 zinfo.compress_type = self.compression
|
|
|
1033 else:
|
|
|
1034 zinfo.compress_type = compress_type
|
|
|
1035
|
|
|
1036 zinfo.file_size = st.st_size
|
|
|
1037 zinfo.flag_bits = 0x00
|
|
|
1038 zinfo.header_offset = self.fp.tell() # Start of header bytes
|
|
|
1039
|
|
|
1040 self._writecheck(zinfo)
|
|
|
1041 self._didModify = True
|
|
|
1042
|
|
|
1043 if isdir:
|
|
|
1044 zinfo.file_size = 0
|
|
|
1045 zinfo.compress_size = 0
|
|
|
1046 zinfo.CRC = 0
|
|
|
1047 self.filelist.append(zinfo)
|
|
|
1048 self.NameToInfo[zinfo.filename] = zinfo
|
|
|
1049 self.fp.write(zinfo.FileHeader())
|
|
|
1050 return
|
|
|
1051
|
|
|
1052 with open(filename, "rb") as fp:
|
|
|
1053 # Must overwrite CRC and sizes with correct data later
|
|
|
1054 zinfo.CRC = CRC = 0
|
|
|
1055 zinfo.compress_size = compress_size = 0
|
|
|
1056 zinfo.file_size = file_size = 0
|
|
|
1057 self.fp.write(zinfo.FileHeader())
|
|
|
1058 if zinfo.compress_type == ZIP_DEFLATED:
|
|
|
1059 cmpr = zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION,
|
|
|
1060 zlib.DEFLATED, -15)
|
|
|
1061 elif zinfo.compress_type == ZIP_BZIP2:
|
|
|
1062 cmpr = bz2.BZ2Compressor()
|
|
|
1063 else:
|
|
|
1064 cmpr = None
|
|
|
1065 while 1:
|
|
|
1066 buf = fp.read(1024 * 8)
|
|
|
1067 if not buf:
|
|
|
1068 break
|
|
|
1069 file_size = file_size + len(buf)
|
|
|
1070 CRC = crc32(buf, CRC) & 0xffffffff
|
|
|
1071 if cmpr:
|
|
|
1072 buf = cmpr.compress(buf)
|
|
|
1073 compress_size = compress_size + len(buf)
|
|
|
1074 self.fp.write(buf)
|
|
|
1075 if cmpr:
|
|
|
1076 buf = cmpr.flush()
|
|
|
1077 compress_size = compress_size + len(buf)
|
|
|
1078 self.fp.write(buf)
|
|
|
1079 zinfo.compress_size = compress_size
|
|
|
1080 else:
|
|
|
1081 zinfo.compress_size = file_size
|
|
|
1082 zinfo.CRC = CRC
|
|
|
1083 zinfo.file_size = file_size
|
|
|
1084 # Seek backwards and write CRC and file sizes
|
|
|
1085 position = self.fp.tell() # Preserve current position in file
|
|
|
1086 self.fp.seek(zinfo.header_offset + 14, 0)
|
|
|
1087 self.fp.write(struct.pack("<LLL", zinfo.CRC, zinfo.compress_size,
|
|
|
1088 zinfo.file_size))
|
|
|
1089 self.fp.seek(position, 0)
|
|
|
1090 self.filelist.append(zinfo)
|
|
|
1091 self.NameToInfo[zinfo.filename] = zinfo
|
|
|
1092
|
|
|
1093 def writestr(self, zinfo_or_arcname, bytes, compress_type=None):
|
|
|
1094 """Write a file into the archive. The contents is the string
|
|
|
1095 'bytes'. 'zinfo_or_arcname' is either a ZipInfo instance or
|
|
|
1096 the name of the file in the archive."""
|
|
|
1097 if not isinstance(zinfo_or_arcname, ZipInfo):
|
|
|
1098 zinfo = ZipInfo(filename=zinfo_or_arcname,
|
|
|
1099 date_time=time.localtime(time.time())[:6])
|
|
|
1100
|
|
|
1101 zinfo.compress_type = self.compression
|
|
|
1102 zinfo.external_attr = 0600 << 16
|
|
|
1103 else:
|
|
|
1104 zinfo = zinfo_or_arcname
|
|
|
1105
|
|
|
1106 if not self.fp:
|
|
|
1107 raise RuntimeError(
|
|
|
1108 "Attempt to write to ZIP archive that was already closed")
|
|
|
1109
|
|
|
1110 if compress_type is not None:
|
|
|
1111 zinfo.compress_type = compress_type
|
|
|
1112
|
|
|
1113 zinfo.file_size = len(bytes) # Uncompressed size
|
|
|
1114 zinfo.header_offset = self.fp.tell() # Start of header bytes
|
|
|
1115 self._writecheck(zinfo)
|
|
|
1116 self._didModify = True
|
|
|
1117 zinfo.CRC = crc32(bytes) & 0xffffffff # CRC-32 checksum
|
|
|
1118 if zinfo.compress_type == ZIP_DEFLATED:
|
|
|
1119 co = zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION,
|
|
|
1120 zlib.DEFLATED, -15)
|
|
|
1121 bytes = co.compress(bytes) + co.flush()
|
|
|
1122 zinfo.compress_size = len(bytes) # Compressed size
|
|
|
1123 elif zinfo.compress_type == ZIP_BZIP2:
|
|
|
1124 co = bz2.BZ2Compressor()
|
|
|
1125 bytes = co.compress(bytes) + co.flush()
|
|
|
1126 zinfo.compress_size = len(bytes) # Compressed size
|
|
|
1127 else:
|
|
|
1128 zinfo.compress_size = zinfo.file_size
|
|
|
1129 zinfo.header_offset = self.fp.tell() # Start of header bytes
|
|
|
1130 self.fp.write(zinfo.FileHeader())
|
|
|
1131 self.fp.write(bytes)
|
|
|
1132 self.fp.flush()
|
|
|
1133 if zinfo.flag_bits & 0x08:
|
|
|
1134 # Write CRC and file sizes after the file data
|
|
|
1135 self.fp.write(struct.pack("<LLL", zinfo.CRC, zinfo.compress_size,
|
|
|
1136 zinfo.file_size))
|
|
|
1137 self.filelist.append(zinfo)
|
|
|
1138 self.NameToInfo[zinfo.filename] = zinfo
|
|
|
1139
|
|
|
1140 def __del__(self):
|
|
|
1141 """Call the "close()" method in case the user forgot."""
|
|
|
1142 self.close()
|
|
|
1143
|
|
|
1144 def close(self):
|
|
|
1145 """Close the file, and for mode "w" and "a" write the ending
|
|
|
1146 records."""
|
|
|
1147 if self.fp is None:
|
|
|
1148 return
|
|
|
1149
|
|
|
1150 if self.mode in ("w", "a") and self._didModify: # write ending records
|
|
|
1151 count = 0
|
|
|
1152 pos1 = self.fp.tell()
|
|
|
1153 for zinfo in self.filelist: # write central directory
|
|
|
1154 count = count + 1
|
|
|
1155 dt = zinfo.date_time
|
|
|
1156 dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2]
|
|
|
1157 dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2)
|
|
|
1158 extra = []
|
|
|
1159 if zinfo.file_size > ZIP64_LIMIT \
|
|
|
1160 or zinfo.compress_size > ZIP64_LIMIT:
|
|
|
1161 extra.append(zinfo.file_size)
|
|
|
1162 extra.append(zinfo.compress_size)
|
|
|
1163 file_size = 0xffffffff
|
|
|
1164 compress_size = 0xffffffff
|
|
|
1165 else:
|
|
|
1166 file_size = zinfo.file_size
|
|
|
1167 compress_size = zinfo.compress_size
|
|
|
1168
|
|
|
1169 if zinfo.header_offset > ZIP64_LIMIT:
|
|
|
1170 extra.append(zinfo.header_offset)
|
|
|
1171 header_offset = 0xffffffffL
|
|
|
1172 else:
|
|
|
1173 header_offset = zinfo.header_offset
|
|
|
1174
|
|
|
1175 extra_data = zinfo.extra
|
|
|
1176 if extra:
|
|
|
1177 # Append a ZIP64 field to the extra's
|
|
|
1178 extra_data = struct.pack(
|
|
|
1179 '<HH' + 'Q'*len(extra),
|
|
|
1180 1, 8*len(extra), *extra) + extra_data
|
|
|
1181
|
|
|
1182 extract_version = max(45, zinfo.extract_version)
|
|
|
1183 create_version = max(45, zinfo.create_version)
|
|
|
1184 else:
|
|
|
1185 extract_version = zinfo.extract_version
|
|
|
1186 create_version = zinfo.create_version
|
|
|
1187
|
|
|
1188 try:
|
|
|
1189 filename, flag_bits = zinfo._encodeFilenameFlags()
|
|
|
1190 centdir = struct.pack(structCentralDir,
|
|
|
1191 stringCentralDir, create_version,
|
|
|
1192 zinfo.create_system, extract_version, zinfo.reserved,
|
|
|
1193 flag_bits, zinfo.compress_type, dostime, dosdate,
|
|
|
1194 zinfo.CRC, compress_size, file_size,
|
|
|
1195 len(filename), len(extra_data), len(zinfo.comment),
|
|
|
1196 0, zinfo.internal_attr, zinfo.external_attr,
|
|
|
1197 header_offset)
|
|
|
1198 except DeprecationWarning:
|
|
|
1199 print >>sys.stderr, (structCentralDir,
|
|
|
1200 stringCentralDir, create_version,
|
|
|
1201 zinfo.create_system, extract_version, zinfo.reserved,
|
|
|
1202 zinfo.flag_bits, zinfo.compress_type, dostime, dosdate,
|
|
|
1203 zinfo.CRC, compress_size, file_size,
|
|
|
1204 len(zinfo.filename), len(extra_data), len(zinfo.comment),
|
|
|
1205 0, zinfo.internal_attr, zinfo.external_attr,
|
|
|
1206 header_offset)
|
|
|
1207 raise
|
|
|
1208 self.fp.write(centdir)
|
|
|
1209 self.fp.write(filename)
|
|
|
1210 self.fp.write(extra_data)
|
|
|
1211 self.fp.write(zinfo.comment)
|
|
|
1212
|
|
|
1213 pos2 = self.fp.tell()
|
|
|
1214 # Write end-of-zip-archive record
|
|
|
1215 centDirCount = count
|
|
|
1216 centDirSize = pos2 - pos1
|
|
|
1217 centDirOffset = pos1
|
|
|
1218 if (centDirCount >= ZIP_FILECOUNT_LIMIT or
|
|
|
1219 centDirOffset > ZIP64_LIMIT or
|
|
|
1220 centDirSize > ZIP64_LIMIT):
|
|
|
1221 # Need to write the ZIP64 end-of-archive records
|
|
|
1222 zip64endrec = struct.pack(
|
|
|
1223 structEndArchive64, stringEndArchive64,
|
|
|
1224 44, 45, 45, 0, 0, centDirCount, centDirCount,
|
|
|
1225 centDirSize, centDirOffset)
|
|
|
1226 self.fp.write(zip64endrec)
|
|
|
1227
|
|
|
1228 zip64locrec = struct.pack(
|
|
|
1229 structEndArchive64Locator,
|
|
|
1230 stringEndArchive64Locator, 0, pos2, 1)
|
|
|
1231 self.fp.write(zip64locrec)
|
|
|
1232 centDirCount = min(centDirCount, 0xFFFF)
|
|
|
1233 centDirSize = min(centDirSize, 0xFFFFFFFF)
|
|
|
1234 centDirOffset = min(centDirOffset, 0xFFFFFFFF)
|
|
|
1235
|
|
|
1236 # check for valid comment length
|
|
|
1237 if len(self.comment) >= ZIP_MAX_COMMENT:
|
|
|
1238 if self.debug > 0:
|
|
|
1239 msg = 'Archive comment is too long; truncating to %d bytes' \
|
|
|
1240 % ZIP_MAX_COMMENT
|
|
|
1241 self.comment = self.comment[:ZIP_MAX_COMMENT]
|
|
|
1242
|
|
|
1243 endrec = struct.pack(structEndArchive, stringEndArchive,
|
|
|
1244 0, 0, centDirCount, centDirCount,
|
|
|
1245 centDirSize, centDirOffset, len(self.comment))
|
|
|
1246 self.fp.write(endrec)
|
|
|
1247 self.fp.write(self.comment)
|
|
|
1248 self.fp.flush()
|
|
|
1249
|
|
|
1250 if not self._filePassed:
|
|
|
1251 self.fp.close()
|
|
|
1252 self.fp = None
|
|
|
1253
|
|
|
1254
|
|
|
1255 class PyZipFile(ZipFile):
|
|
|
1256 """Class to create ZIP archives with Python library files and packages."""
|
|
|
1257
|
|
|
1258 def writepy(self, pathname, basename = ""):
|
|
|
1259 """Add all files from "pathname" to the ZIP archive.
|
|
|
1260
|
|
|
1261 If pathname is a package directory, search the directory and
|
|
|
1262 all package subdirectories recursively for all *.py and enter
|
|
|
1263 the modules into the archive. If pathname is a plain
|
|
|
1264 directory, listdir *.py and enter all modules. Else, pathname
|
|
|
1265 must be a Python *.py file and the module will be put into the
|
|
|
1266 archive. Added modules are always module.pyo or module.pyc.
|
|
|
1267 This method will compile the module.py into module.pyc if
|
|
|
1268 necessary.
|
|
|
1269 """
|
|
|
1270 dir, name = os.path.split(pathname)
|
|
|
1271 if os.path.isdir(pathname):
|
|
|
1272 initname = os.path.join(pathname, "__init__.py")
|
|
|
1273 if os.path.isfile(initname):
|
|
|
1274 # This is a package directory, add it
|
|
|
1275 if basename:
|
|
|
1276 basename = "%s/%s" % (basename, name)
|
|
|
1277 else:
|
|
|
1278 basename = name
|
|
|
1279 if self.debug:
|
|
|
1280 print "Adding package in", pathname, "as", basename
|
|
|
1281 fname, arcname = self._get_codename(initname[0:-3], basename)
|
|
|
1282 if self.debug:
|
|
|
1283 print "Adding", arcname
|
|
|
1284 self.write(fname, arcname)
|
|
|
1285 dirlist = os.listdir(pathname)
|
|
|
1286 dirlist.remove("__init__.py")
|
|
|
1287 # Add all *.py files and package subdirectories
|
|
|
1288 for filename in dirlist:
|
|
|
1289 path = os.path.join(pathname, filename)
|
|
|
1290 root, ext = os.path.splitext(filename)
|
|
|
1291 if os.path.isdir(path):
|
|
|
1292 if os.path.isfile(os.path.join(path, "__init__.py")):
|
|
|
1293 # This is a package directory, add it
|
|
|
1294 self.writepy(path, basename) # Recursive call
|
|
|
1295 elif ext == ".py":
|
|
|
1296 fname, arcname = self._get_codename(path[0:-3],
|
|
|
1297 basename)
|
|
|
1298 if self.debug:
|
|
|
1299 print "Adding", arcname
|
|
|
1300 self.write(fname, arcname)
|
|
|
1301 else:
|
|
|
1302 # This is NOT a package directory, add its files at top level
|
|
|
1303 if self.debug:
|
|
|
1304 print "Adding files from directory", pathname
|
|
|
1305 for filename in os.listdir(pathname):
|
|
|
1306 path = os.path.join(pathname, filename)
|
|
|
1307 root, ext = os.path.splitext(filename)
|
|
|
1308 if ext == ".py":
|
|
|
1309 fname, arcname = self._get_codename(path[0:-3],
|
|
|
1310 basename)
|
|
|
1311 if self.debug:
|
|
|
1312 print "Adding", arcname
|
|
|
1313 self.write(fname, arcname)
|
|
|
1314 else:
|
|
|
1315 if pathname[-3:] != ".py":
|
|
|
1316 raise RuntimeError, \
|
|
|
1317 'Files added with writepy() must end with ".py"'
|
|
|
1318 fname, arcname = self._get_codename(pathname[0:-3], basename)
|
|
|
1319 if self.debug:
|
|
|
1320 print "Adding file", arcname
|
|
|
1321 self.write(fname, arcname)
|
|
|
1322
|
|
|
1323 def _get_codename(self, pathname, basename):
|
|
|
1324 """Return (filename, archivename) for the path.
|
|
|
1325
|
|
|
1326 Given a module name path, return the correct file path and
|
|
|
1327 archive name, compiling if necessary. For example, given
|
|
|
1328 /python/lib/string, return (/python/lib/string.pyc, string).
|
|
|
1329 """
|
|
|
1330 file_py = pathname + ".py"
|
|
|
1331 file_pyc = pathname + ".pyc"
|
|
|
1332 file_pyo = pathname + ".pyo"
|
|
|
1333 if os.path.isfile(file_pyo) and \
|
|
|
1334 os.stat(file_pyo).st_mtime >= os.stat(file_py).st_mtime:
|
|
|
1335 fname = file_pyo # Use .pyo file
|
|
|
1336 elif not os.path.isfile(file_pyc) or \
|
|
|
1337 os.stat(file_pyc).st_mtime < os.stat(file_py).st_mtime:
|
|
|
1338 import py_compile
|
|
|
1339 if self.debug:
|
|
|
1340 print "Compiling", file_py
|
|
|
1341 try:
|
|
|
1342 py_compile.compile(file_py, file_pyc, None, True)
|
|
|
1343 except py_compile.PyCompileError,err:
|
|
|
1344 print err.msg
|
|
|
1345 fname = file_pyc
|
|
|
1346 else:
|
|
|
1347 fname = file_pyc
|
|
|
1348 archivename = os.path.split(fname)[1]
|
|
|
1349 if basename:
|
|
|
1350 archivename = "%s/%s" % (basename, archivename)
|
|
|
1351 return (fname, archivename)
|
|
|
1352
|
|
|
1353
|
|
|
1354 def main(args = None):
|
|
|
1355 import textwrap
|
|
|
1356 USAGE=textwrap.dedent("""\
|
|
|
1357 Usage:
|
|
|
1358 zipfile.py -l zipfile.zip # Show listing of a zipfile
|
|
|
1359 zipfile.py -t zipfile.zip # Test if a zipfile is valid
|
|
|
1360 zipfile.py -e zipfile.zip target # Extract zipfile into target dir
|
|
|
1361 zipfile.py -c zipfile.zip src ... # Create zipfile from sources
|
|
|
1362 """)
|
|
|
1363 if args is None:
|
|
|
1364 args = sys.argv[1:]
|
|
|
1365
|
|
|
1366 if not args or args[0] not in ('-l', '-c', '-e', '-t'):
|
|
|
1367 print USAGE
|
|
|
1368 sys.exit(1)
|
|
|
1369
|
|
|
1370 if args[0] == '-l':
|
|
|
1371 if len(args) != 2:
|
|
|
1372 print USAGE
|
|
|
1373 sys.exit(1)
|
|
|
1374 zf = ZipFile(args[1], 'r')
|
|
|
1375 zf.printdir()
|
|
|
1376 zf.close()
|
|
|
1377
|
|
|
1378 elif args[0] == '-t':
|
|
|
1379 if len(args) != 2:
|
|
|
1380 print USAGE
|
|
|
1381 sys.exit(1)
|
|
|
1382 zf = ZipFile(args[1], 'r')
|
|
|
1383 zf.testzip()
|
|
|
1384 print "Done testing"
|
|
|
1385
|
|
|
1386 elif args[0] == '-e':
|
|
|
1387 if len(args) != 3:
|
|
|
1388 print USAGE
|
|
|
1389 sys.exit(1)
|
|
|
1390
|
|
|
1391 zf = ZipFile(args[1], 'r')
|
|
|
1392 out = args[2]
|
|
|
1393 for path in zf.namelist():
|
|
|
1394 if path.startswith('./'):
|
|
|
1395 tgt = os.path.join(out, path[2:])
|
|
|
1396 else:
|
|
|
1397 tgt = os.path.join(out, path)
|
|
|
1398
|
|
|
1399 tgtdir = os.path.dirname(tgt)
|
|
|
1400 if not os.path.exists(tgtdir):
|
|
|
1401 os.makedirs(tgtdir)
|
|
|
1402 with open(tgt, 'wb') as fp:
|
|
|
1403 fp.write(zf.read(path))
|
|
|
1404 zf.close()
|
|
|
1405
|
|
|
1406 elif args[0] == '-c':
|
|
|
1407 if len(args) < 3:
|
|
|
1408 print USAGE
|
|
|
1409 sys.exit(1)
|
|
|
1410
|
|
|
1411 def addToZip(zf, path, zippath):
|
|
|
1412 if os.path.isfile(path):
|
|
|
1413 zf.write(path, zippath, ZIP_DEFLATED)
|
|
|
1414 elif os.path.isdir(path):
|
|
|
1415 for nm in os.listdir(path):
|
|
|
1416 addToZip(zf,
|
|
|
1417 os.path.join(path, nm), os.path.join(zippath, nm))
|
|
|
1418 # else: ignore
|
|
|
1419
|
|
|
1420 zf = ZipFile(args[1], 'w', allowZip64=True)
|
|
|
1421 for src in args[2:]:
|
|
|
1422 addToZip(zf, src, os.path.basename(src))
|
|
|
1423
|
|
|
1424 zf.close()
|
|
|
1425
|
|
|
1426 if __name__ == "__main__":
|
|
|
1427 main()
|
