Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1 | """ |
| 2 | Read and write ZIP files. |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 3 | |
| 4 | XXX references to utf-8 need further investigation. |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 5 | """ |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 6 | import io |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 7 | import os |
Brett Cannon | b57a085 | 2013-06-15 17:32:30 -0400 | [diff] [blame] | 8 | import importlib.util |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 9 | import sys |
| 10 | import time |
| 11 | import stat |
| 12 | import shutil |
| 13 | import struct |
| 14 | import binascii |
| 15 | |
Serhiy Storchaka | 9e77773 | 2015-10-10 19:43:32 +0300 | [diff] [blame] | 16 | try: |
| 17 | import threading |
| 18 | except ImportError: |
| 19 | import dummy_threading as threading |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 20 | |
| 21 | try: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 22 | import zlib # We may need its compression method |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 23 | crc32 = zlib.crc32 |
Brett Cannon | 260fbe8 | 2013-07-04 18:16:15 -0400 | [diff] [blame] | 24 | except ImportError: |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 25 | zlib = None |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 26 | crc32 = binascii.crc32 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 27 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 28 | try: |
| 29 | import bz2 # We may need its compression method |
Brett Cannon | 260fbe8 | 2013-07-04 18:16:15 -0400 | [diff] [blame] | 30 | except ImportError: |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 31 | bz2 = None |
| 32 | |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 33 | try: |
| 34 | import lzma # We may need its compression method |
Brett Cannon | 260fbe8 | 2013-07-04 18:16:15 -0400 | [diff] [blame] | 35 | except ImportError: |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 36 | lzma = None |
| 37 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 38 | __all__ = ["BadZipFile", "BadZipfile", "error", |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 39 | "ZIP_STORED", "ZIP_DEFLATED", "ZIP_BZIP2", "ZIP_LZMA", |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 40 | "is_zipfile", "ZipInfo", "ZipFile", "PyZipFile", "LargeZipFile"] |
Skip Montanaro | 40fc160 | 2001-03-01 04:27:19 +0000 | [diff] [blame] | 41 | |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 42 | class BadZipFile(Exception): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 43 | pass |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 44 | |
| 45 | |
| 46 | class LargeZipFile(Exception): |
| 47 | """ |
| 48 | Raised when writing a zipfile, the zipfile requires ZIP64 extensions |
| 49 | and those extensions are disabled. |
| 50 | """ |
| 51 | |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 52 | error = BadZipfile = BadZipFile # Pre-3.2 compatibility names |
| 53 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 54 | |
Amaury Forgeot d'Arc | 0c3f8a4 | 2009-01-17 16:42:26 +0000 | [diff] [blame] | 55 | ZIP64_LIMIT = (1 << 31) - 1 |
Serhiy Storchaka | cfbb394 | 2014-09-23 21:34:24 +0300 | [diff] [blame] | 56 | ZIP_FILECOUNT_LIMIT = (1 << 16) - 1 |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 57 | ZIP_MAX_COMMENT = (1 << 16) - 1 |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 58 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 59 | # constants for Zip file compression methods |
| 60 | ZIP_STORED = 0 |
| 61 | ZIP_DEFLATED = 8 |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 62 | ZIP_BZIP2 = 12 |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 63 | ZIP_LZMA = 14 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 64 | # Other ZIP compression methods not supported |
| 65 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 66 | DEFAULT_VERSION = 20 |
| 67 | ZIP64_VERSION = 45 |
| 68 | BZIP2_VERSION = 46 |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 69 | LZMA_VERSION = 63 |
Martin v. Löwis | d099b56 | 2012-05-01 14:08:22 +0200 | [diff] [blame] | 70 | # we recognize (but not necessarily support) all features up to that version |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 71 | MAX_EXTRACT_VERSION = 63 |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 72 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 73 | # Below are some formats and associated data for reading/writing headers using |
| 74 | # the struct module. The names and structures of headers/records are those used |
| 75 | # in the PKWARE description of the ZIP file format: |
| 76 | # http://www.pkware.com/documents/casestudies/APPNOTE.TXT |
| 77 | # (URL valid as of January 2008) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 78 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 79 | # The "end of central directory" structure, magic number, size, and indices |
| 80 | # (section V.I in the format document) |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 81 | structEndArchive = b"<4s4H2LH" |
| 82 | stringEndArchive = b"PK\005\006" |
| 83 | sizeEndCentDir = struct.calcsize(structEndArchive) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 84 | |
| 85 | _ECD_SIGNATURE = 0 |
| 86 | _ECD_DISK_NUMBER = 1 |
| 87 | _ECD_DISK_START = 2 |
| 88 | _ECD_ENTRIES_THIS_DISK = 3 |
| 89 | _ECD_ENTRIES_TOTAL = 4 |
| 90 | _ECD_SIZE = 5 |
| 91 | _ECD_OFFSET = 6 |
| 92 | _ECD_COMMENT_SIZE = 7 |
| 93 | # These last two indices are not part of the structure as defined in the |
| 94 | # spec, but they are used internally by this module as a convenience |
| 95 | _ECD_COMMENT = 8 |
| 96 | _ECD_LOCATION = 9 |
| 97 | |
| 98 | # The "central directory" structure, magic number, size, and indices |
| 99 | # of entries in the structure (section V.F in the format document) |
| 100 | structCentralDir = "<4s4B4HL2L5H2L" |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 101 | stringCentralDir = b"PK\001\002" |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 102 | sizeCentralDir = struct.calcsize(structCentralDir) |
| 103 | |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 104 | # indexes of entries in the central directory structure |
| 105 | _CD_SIGNATURE = 0 |
| 106 | _CD_CREATE_VERSION = 1 |
| 107 | _CD_CREATE_SYSTEM = 2 |
| 108 | _CD_EXTRACT_VERSION = 3 |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 109 | _CD_EXTRACT_SYSTEM = 4 |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 110 | _CD_FLAG_BITS = 5 |
| 111 | _CD_COMPRESS_TYPE = 6 |
| 112 | _CD_TIME = 7 |
| 113 | _CD_DATE = 8 |
| 114 | _CD_CRC = 9 |
| 115 | _CD_COMPRESSED_SIZE = 10 |
| 116 | _CD_UNCOMPRESSED_SIZE = 11 |
| 117 | _CD_FILENAME_LENGTH = 12 |
| 118 | _CD_EXTRA_FIELD_LENGTH = 13 |
| 119 | _CD_COMMENT_LENGTH = 14 |
| 120 | _CD_DISK_NUMBER_START = 15 |
| 121 | _CD_INTERNAL_FILE_ATTRIBUTES = 16 |
| 122 | _CD_EXTERNAL_FILE_ATTRIBUTES = 17 |
| 123 | _CD_LOCAL_HEADER_OFFSET = 18 |
| 124 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 125 | # The "local file header" structure, magic number, size, and indices |
| 126 | # (section V.A in the format document) |
| 127 | structFileHeader = "<4s2B4HL2L2H" |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 128 | stringFileHeader = b"PK\003\004" |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 129 | sizeFileHeader = struct.calcsize(structFileHeader) |
| 130 | |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 131 | _FH_SIGNATURE = 0 |
| 132 | _FH_EXTRACT_VERSION = 1 |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 133 | _FH_EXTRACT_SYSTEM = 2 |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 134 | _FH_GENERAL_PURPOSE_FLAG_BITS = 3 |
| 135 | _FH_COMPRESSION_METHOD = 4 |
| 136 | _FH_LAST_MOD_TIME = 5 |
| 137 | _FH_LAST_MOD_DATE = 6 |
| 138 | _FH_CRC = 7 |
| 139 | _FH_COMPRESSED_SIZE = 8 |
| 140 | _FH_UNCOMPRESSED_SIZE = 9 |
| 141 | _FH_FILENAME_LENGTH = 10 |
| 142 | _FH_EXTRA_FIELD_LENGTH = 11 |
| 143 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 144 | # The "Zip64 end of central directory locator" structure, magic number, and size |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 145 | structEndArchive64Locator = "<4sLQL" |
| 146 | stringEndArchive64Locator = b"PK\x06\x07" |
| 147 | sizeEndCentDir64Locator = struct.calcsize(structEndArchive64Locator) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 148 | |
| 149 | # The "Zip64 end of central directory" record, magic number, size, and indices |
| 150 | # (section V.G in the format document) |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 151 | structEndArchive64 = "<4sQ2H2L4Q" |
| 152 | stringEndArchive64 = b"PK\x06\x06" |
| 153 | sizeEndCentDir64 = struct.calcsize(structEndArchive64) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 154 | |
| 155 | _CD64_SIGNATURE = 0 |
| 156 | _CD64_DIRECTORY_RECSIZE = 1 |
| 157 | _CD64_CREATE_VERSION = 2 |
| 158 | _CD64_EXTRACT_VERSION = 3 |
| 159 | _CD64_DISK_NUMBER = 4 |
| 160 | _CD64_DISK_NUMBER_START = 5 |
| 161 | _CD64_NUMBER_ENTRIES_THIS_DISK = 6 |
| 162 | _CD64_NUMBER_ENTRIES_TOTAL = 7 |
| 163 | _CD64_DIRECTORY_SIZE = 8 |
| 164 | _CD64_OFFSET_START_CENTDIR = 9 |
| 165 | |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 166 | def _check_zipfile(fp): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 167 | try: |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 168 | if _EndRecData(fp): |
| 169 | return True # file has correct magic number |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 170 | except OSError: |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 171 | pass |
Guido van Rossum | 8ca162f | 2002-04-07 06:36:23 +0000 | [diff] [blame] | 172 | return False |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 173 | |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 174 | def is_zipfile(filename): |
| 175 | """Quickly see if a file is a ZIP file by checking the magic number. |
| 176 | |
| 177 | The filename argument may be a file or file-like object too. |
| 178 | """ |
| 179 | result = False |
| 180 | try: |
| 181 | if hasattr(filename, "read"): |
| 182 | result = _check_zipfile(fp=filename) |
| 183 | else: |
| 184 | with open(filename, "rb") as fp: |
| 185 | result = _check_zipfile(fp) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 186 | except OSError: |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 187 | pass |
| 188 | return result |
| 189 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 190 | def _EndRecData64(fpin, offset, endrec): |
| 191 | """ |
| 192 | Read the ZIP64 end-of-archive records and use that to update endrec |
| 193 | """ |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 194 | try: |
| 195 | fpin.seek(offset - sizeEndCentDir64Locator, 2) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 196 | except OSError: |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 197 | # If the seek fails, the file is not large enough to contain a ZIP64 |
| 198 | # end-of-archive record, so just return the end record we were given. |
| 199 | return endrec |
| 200 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 201 | data = fpin.read(sizeEndCentDir64Locator) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 202 | if len(data) != sizeEndCentDir64Locator: |
| 203 | return endrec |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 204 | sig, diskno, reloff, disks = struct.unpack(structEndArchive64Locator, data) |
| 205 | if sig != stringEndArchive64Locator: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 206 | return endrec |
| 207 | |
| 208 | if diskno != 0 or disks != 1: |
Éric Araujo | ae2d832 | 2010-10-28 13:49:17 +0000 | [diff] [blame] | 209 | raise BadZipFile("zipfiles that span multiple disks are not supported") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 210 | |
| 211 | # Assume no 'zip64 extensible data' |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 212 | fpin.seek(offset - sizeEndCentDir64Locator - sizeEndCentDir64, 2) |
| 213 | data = fpin.read(sizeEndCentDir64) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 214 | if len(data) != sizeEndCentDir64: |
| 215 | return endrec |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 216 | sig, sz, create_version, read_version, disk_num, disk_dir, \ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 217 | dircount, dircount2, dirsize, diroffset = \ |
| 218 | struct.unpack(structEndArchive64, data) |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 219 | if sig != stringEndArchive64: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 220 | return endrec |
| 221 | |
| 222 | # Update the original endrec using data from the ZIP64 record |
Antoine Pitrou | 9e4fdf4 | 2008-09-05 23:43:02 +0000 | [diff] [blame] | 223 | endrec[_ECD_SIGNATURE] = sig |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 224 | endrec[_ECD_DISK_NUMBER] = disk_num |
| 225 | endrec[_ECD_DISK_START] = disk_dir |
| 226 | endrec[_ECD_ENTRIES_THIS_DISK] = dircount |
| 227 | endrec[_ECD_ENTRIES_TOTAL] = dircount2 |
| 228 | endrec[_ECD_SIZE] = dirsize |
| 229 | endrec[_ECD_OFFSET] = diroffset |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 230 | return endrec |
| 231 | |
| 232 | |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 233 | def _EndRecData(fpin): |
| 234 | """Return data from the "End of Central Directory" record, or None. |
| 235 | |
| 236 | The data is a list of the nine items in the ZIP "End of central dir" |
| 237 | record followed by a tenth item, the file seek offset of this record.""" |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 238 | |
| 239 | # Determine file size |
| 240 | fpin.seek(0, 2) |
| 241 | filesize = fpin.tell() |
| 242 | |
| 243 | # Check to see if this is ZIP file with no archive comment (the |
| 244 | # "end of central directory" structure should be the last item in the |
| 245 | # file if this is the case). |
Amaury Forgeot d'Arc | bc34780 | 2009-07-28 22:18:57 +0000 | [diff] [blame] | 246 | try: |
| 247 | fpin.seek(-sizeEndCentDir, 2) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 248 | except OSError: |
Amaury Forgeot d'Arc | bc34780 | 2009-07-28 22:18:57 +0000 | [diff] [blame] | 249 | return None |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 250 | data = fpin.read() |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 251 | if (len(data) == sizeEndCentDir and |
| 252 | data[0:4] == stringEndArchive and |
| 253 | data[-2:] == b"\000\000"): |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 254 | # the signature is correct and there's no comment, unpack structure |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 255 | endrec = struct.unpack(structEndArchive, data) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 256 | endrec=list(endrec) |
| 257 | |
| 258 | # Append a blank comment and record start offset |
| 259 | endrec.append(b"") |
| 260 | endrec.append(filesize - sizeEndCentDir) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 261 | |
Amaury Forgeot d'Arc | d3fb4bb | 2009-01-18 00:29:02 +0000 | [diff] [blame] | 262 | # Try to read the "Zip64 end of central directory" structure |
| 263 | return _EndRecData64(fpin, -sizeEndCentDir, endrec) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 264 | |
| 265 | # Either this is not a ZIP file, or it is a ZIP file with an archive |
| 266 | # comment. Search the end of the file for the "end of central directory" |
| 267 | # record signature. The comment is the last item in the ZIP file and may be |
| 268 | # up to 64K long. It is assumed that the "end of central directory" magic |
| 269 | # number does not appear in the comment. |
| 270 | maxCommentStart = max(filesize - (1 << 16) - sizeEndCentDir, 0) |
| 271 | fpin.seek(maxCommentStart, 0) |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 272 | data = fpin.read() |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 273 | start = data.rfind(stringEndArchive) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 274 | if start >= 0: |
| 275 | # found the magic number; attempt to unpack and interpret |
| 276 | recData = data[start:start+sizeEndCentDir] |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 277 | if len(recData) != sizeEndCentDir: |
| 278 | # Zip file is corrupted. |
| 279 | return None |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 280 | endrec = list(struct.unpack(structEndArchive, recData)) |
R David Murray | 4fbb9db | 2011-06-09 15:50:51 -0400 | [diff] [blame] | 281 | commentSize = endrec[_ECD_COMMENT_SIZE] #as claimed by the zip file |
| 282 | comment = data[start+sizeEndCentDir:start+sizeEndCentDir+commentSize] |
| 283 | endrec.append(comment) |
| 284 | endrec.append(maxCommentStart + start) |
Amaury Forgeot d'Arc | d3fb4bb | 2009-01-18 00:29:02 +0000 | [diff] [blame] | 285 | |
R David Murray | 4fbb9db | 2011-06-09 15:50:51 -0400 | [diff] [blame] | 286 | # Try to read the "Zip64 end of central directory" structure |
| 287 | return _EndRecData64(fpin, maxCommentStart + start - filesize, |
| 288 | endrec) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 289 | |
| 290 | # Unable to find a valid end of central directory structure |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 291 | return None |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 292 | |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 293 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 294 | class ZipInfo (object): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 295 | """Class with attributes describing each file in the ZIP archive.""" |
| 296 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 297 | __slots__ = ( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 298 | 'orig_filename', |
| 299 | 'filename', |
| 300 | 'date_time', |
| 301 | 'compress_type', |
| 302 | 'comment', |
| 303 | 'extra', |
| 304 | 'create_system', |
| 305 | 'create_version', |
| 306 | 'extract_version', |
| 307 | 'reserved', |
| 308 | 'flag_bits', |
| 309 | 'volume', |
| 310 | 'internal_attr', |
| 311 | 'external_attr', |
| 312 | 'header_offset', |
| 313 | 'CRC', |
| 314 | 'compress_size', |
| 315 | 'file_size', |
| 316 | '_raw_time', |
| 317 | ) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 318 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 319 | def __init__(self, filename="NoName", date_time=(1980,1,1,0,0,0)): |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 320 | self.orig_filename = filename # Original file name in archive |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 321 | |
| 322 | # Terminate the file name at the first null byte. Null bytes in file |
| 323 | # names are used as tricks by viruses in archives. |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 324 | null_byte = filename.find(chr(0)) |
| 325 | if null_byte >= 0: |
| 326 | filename = filename[0:null_byte] |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 327 | # This is used to ensure paths in generated ZIP files always use |
| 328 | # forward slashes as the directory separator, as required by the |
| 329 | # ZIP format specification. |
| 330 | if os.sep != "/" and os.sep in filename: |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 331 | filename = filename.replace(os.sep, "/") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 332 | |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 333 | self.filename = filename # Normalized file name |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 334 | self.date_time = date_time # year, month, day, hour, min, sec |
Senthil Kumaran | 29fa9d4 | 2011-10-20 01:46:00 +0800 | [diff] [blame] | 335 | |
| 336 | if date_time[0] < 1980: |
| 337 | raise ValueError('ZIP does not support timestamps before 1980') |
| 338 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 339 | # Standard values: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 340 | self.compress_type = ZIP_STORED # Type of compression for the file |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 341 | self.comment = b"" # Comment for each file |
| 342 | self.extra = b"" # ZIP extra data |
Martin v. Löwis | 0075690 | 2006-02-05 17:09:41 +0000 | [diff] [blame] | 343 | if sys.platform == 'win32': |
| 344 | self.create_system = 0 # System which created ZIP archive |
| 345 | else: |
| 346 | # Assume everything else is unix-y |
| 347 | self.create_system = 3 # System which created ZIP archive |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 348 | self.create_version = DEFAULT_VERSION # Version which created ZIP archive |
| 349 | self.extract_version = DEFAULT_VERSION # Version needed to extract archive |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 350 | self.reserved = 0 # Must be zero |
| 351 | self.flag_bits = 0 # ZIP flag bits |
| 352 | self.volume = 0 # Volume number of file header |
| 353 | self.internal_attr = 0 # Internal attributes |
| 354 | self.external_attr = 0 # External file attributes |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 355 | # Other attributes are set by class ZipFile: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 356 | # header_offset Byte offset to the file header |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 357 | # CRC CRC-32 of the uncompressed file |
| 358 | # compress_size Size of the compressed file |
| 359 | # file_size Size of the uncompressed file |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 360 | |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 361 | def __repr__(self): |
| 362 | result = ['<%s filename=%r' % (self.__class__.__name__, self.filename)] |
| 363 | if self.compress_type != ZIP_STORED: |
| 364 | result.append(' compress_type=%s' % |
| 365 | compressor_names.get(self.compress_type, |
| 366 | self.compress_type)) |
| 367 | hi = self.external_attr >> 16 |
| 368 | lo = self.external_attr & 0xFFFF |
| 369 | if hi: |
| 370 | result.append(' filemode=%r' % stat.filemode(hi)) |
| 371 | if lo: |
| 372 | result.append(' external_attr=%#x' % lo) |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 373 | isdir = self.is_dir() |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 374 | if not isdir or self.file_size: |
| 375 | result.append(' file_size=%r' % self.file_size) |
| 376 | if ((not isdir or self.compress_size) and |
| 377 | (self.compress_type != ZIP_STORED or |
| 378 | self.file_size != self.compress_size)): |
| 379 | result.append(' compress_size=%r' % self.compress_size) |
| 380 | result.append('>') |
| 381 | return ''.join(result) |
| 382 | |
Serhiy Storchaka | 182d7cd | 2013-01-15 00:31:39 +0200 | [diff] [blame] | 383 | def FileHeader(self, zip64=None): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 384 | """Return the per-file header as a string.""" |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 385 | dt = self.date_time |
| 386 | dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2] |
Tim Peters | 3caca23 | 2001-12-06 06:23:26 +0000 | [diff] [blame] | 387 | dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 388 | if self.flag_bits & 0x08: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 389 | # Set these to zero because we write them after the file data |
| 390 | CRC = compress_size = file_size = 0 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 391 | else: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 392 | CRC = self.CRC |
| 393 | compress_size = self.compress_size |
| 394 | file_size = self.file_size |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 395 | |
| 396 | extra = self.extra |
| 397 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 398 | min_version = 0 |
Serhiy Storchaka | 182d7cd | 2013-01-15 00:31:39 +0200 | [diff] [blame] | 399 | if zip64 is None: |
| 400 | zip64 = file_size > ZIP64_LIMIT or compress_size > ZIP64_LIMIT |
| 401 | if zip64: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 402 | fmt = '<HHQQ' |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 403 | extra = extra + struct.pack(fmt, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 404 | 1, struct.calcsize(fmt)-4, file_size, compress_size) |
Serhiy Storchaka | 182d7cd | 2013-01-15 00:31:39 +0200 | [diff] [blame] | 405 | if file_size > ZIP64_LIMIT or compress_size > ZIP64_LIMIT: |
| 406 | if not zip64: |
| 407 | raise LargeZipFile("Filesize would require ZIP64 extensions") |
| 408 | # File is larger than what fits into a 4 byte integer, |
| 409 | # fall back to the ZIP64 extension |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 410 | file_size = 0xffffffff |
| 411 | compress_size = 0xffffffff |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 412 | min_version = ZIP64_VERSION |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 413 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 414 | if self.compress_type == ZIP_BZIP2: |
| 415 | min_version = max(BZIP2_VERSION, min_version) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 416 | elif self.compress_type == ZIP_LZMA: |
| 417 | min_version = max(LZMA_VERSION, min_version) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 418 | |
| 419 | self.extract_version = max(min_version, self.extract_version) |
| 420 | self.create_version = max(min_version, self.create_version) |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 421 | filename, flag_bits = self._encodeFilenameFlags() |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 422 | header = struct.pack(structFileHeader, stringFileHeader, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 423 | self.extract_version, self.reserved, flag_bits, |
| 424 | self.compress_type, dostime, dosdate, CRC, |
| 425 | compress_size, file_size, |
| 426 | len(filename), len(extra)) |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 427 | return header + filename + extra |
| 428 | |
| 429 | def _encodeFilenameFlags(self): |
| 430 | try: |
| 431 | return self.filename.encode('ascii'), self.flag_bits |
| 432 | except UnicodeEncodeError: |
| 433 | return self.filename.encode('utf-8'), self.flag_bits | 0x800 |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 434 | |
| 435 | def _decodeExtra(self): |
| 436 | # Try to decode the extra field. |
| 437 | extra = self.extra |
| 438 | unpack = struct.unpack |
Gregory P. Smith | 0af8a86 | 2014-05-29 23:42:14 -0700 | [diff] [blame] | 439 | while len(extra) >= 4: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 440 | tp, ln = unpack('<HH', extra[:4]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 441 | if tp == 1: |
| 442 | if ln >= 24: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 443 | counts = unpack('<QQQ', extra[4:28]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 444 | elif ln == 16: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 445 | counts = unpack('<QQ', extra[4:20]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 446 | elif ln == 8: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 447 | counts = unpack('<Q', extra[4:12]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 448 | elif ln == 0: |
| 449 | counts = () |
| 450 | else: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 451 | raise BadZipFile("Corrupt extra field %04x (size=%d)" % (tp, ln)) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 452 | |
| 453 | idx = 0 |
| 454 | |
| 455 | # ZIP64 extension (large files and/or large archives) |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 456 | if self.file_size in (0xffffffffffffffff, 0xffffffff): |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 457 | self.file_size = counts[idx] |
| 458 | idx += 1 |
| 459 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 460 | if self.compress_size == 0xFFFFFFFF: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 461 | self.compress_size = counts[idx] |
| 462 | idx += 1 |
| 463 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 464 | if self.header_offset == 0xffffffff: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 465 | old = self.header_offset |
| 466 | self.header_offset = counts[idx] |
| 467 | idx+=1 |
| 468 | |
| 469 | extra = extra[ln+4:] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 470 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 471 | @classmethod |
| 472 | def from_file(cls, filename, arcname=None): |
| 473 | """Construct an appropriate ZipInfo for a file on the filesystem. |
| 474 | |
| 475 | filename should be the path to a file or directory on the filesystem. |
| 476 | |
| 477 | arcname is the name which it will have within the archive (by default, |
| 478 | this will be the same as filename, but without a drive letter and with |
| 479 | leading path separators removed). |
| 480 | """ |
| 481 | st = os.stat(filename) |
| 482 | isdir = stat.S_ISDIR(st.st_mode) |
| 483 | mtime = time.localtime(st.st_mtime) |
| 484 | date_time = mtime[0:6] |
| 485 | # Create ZipInfo instance to store file information |
| 486 | if arcname is None: |
| 487 | arcname = filename |
| 488 | arcname = os.path.normpath(os.path.splitdrive(arcname)[1]) |
| 489 | while arcname[0] in (os.sep, os.altsep): |
| 490 | arcname = arcname[1:] |
| 491 | if isdir: |
| 492 | arcname += '/' |
| 493 | zinfo = cls(arcname, date_time) |
| 494 | zinfo.external_attr = (st.st_mode & 0xFFFF) << 16 # Unix attributes |
| 495 | if isdir: |
| 496 | zinfo.file_size = 0 |
| 497 | zinfo.external_attr |= 0x10 # MS-DOS directory flag |
| 498 | else: |
| 499 | zinfo.file_size = st.st_size |
| 500 | |
| 501 | return zinfo |
| 502 | |
| 503 | def is_dir(self): |
Serhiy Storchaka | f47fc55 | 2016-05-15 12:27:16 +0300 | [diff] [blame] | 504 | """Return True if this archive member is a directory.""" |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 505 | return self.filename[-1] == '/' |
| 506 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 507 | |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 508 | class _ZipDecrypter: |
| 509 | """Class to handle decryption of files stored within a ZIP archive. |
| 510 | |
| 511 | ZIP supports a password-based form of encryption. Even though known |
| 512 | plaintext attacks have been found against it, it is still useful |
Christian Heimes | fdab48e | 2008-01-20 09:06:41 +0000 | [diff] [blame] | 513 | to be able to get data out of such a file. |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 514 | |
| 515 | Usage: |
| 516 | zd = _ZipDecrypter(mypwd) |
| 517 | plain_char = zd(cypher_char) |
| 518 | plain_text = map(zd, cypher_text) |
| 519 | """ |
| 520 | |
| 521 | def _GenerateCRCTable(): |
| 522 | """Generate a CRC-32 table. |
| 523 | |
| 524 | ZIP encryption uses the CRC32 one-byte primitive for scrambling some |
| 525 | internal keys. We noticed that a direct implementation is faster than |
| 526 | relying on binascii.crc32(). |
| 527 | """ |
| 528 | poly = 0xedb88320 |
| 529 | table = [0] * 256 |
| 530 | for i in range(256): |
| 531 | crc = i |
| 532 | for j in range(8): |
| 533 | if crc & 1: |
| 534 | crc = ((crc >> 1) & 0x7FFFFFFF) ^ poly |
| 535 | else: |
| 536 | crc = ((crc >> 1) & 0x7FFFFFFF) |
| 537 | table[i] = crc |
| 538 | return table |
Daniel Holth | 9dee304 | 2014-01-02 23:17:21 -0500 | [diff] [blame] | 539 | crctable = None |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 540 | |
| 541 | def _crc32(self, ch, crc): |
| 542 | """Compute the CRC32 primitive on one byte.""" |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 543 | return ((crc >> 8) & 0xffffff) ^ self.crctable[(crc ^ ch) & 0xff] |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 544 | |
| 545 | def __init__(self, pwd): |
Daniel Holth | 9dee304 | 2014-01-02 23:17:21 -0500 | [diff] [blame] | 546 | if _ZipDecrypter.crctable is None: |
| 547 | _ZipDecrypter.crctable = _ZipDecrypter._GenerateCRCTable() |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 548 | self.key0 = 305419896 |
| 549 | self.key1 = 591751049 |
| 550 | self.key2 = 878082192 |
| 551 | for p in pwd: |
| 552 | self._UpdateKeys(p) |
| 553 | |
| 554 | def _UpdateKeys(self, c): |
| 555 | self.key0 = self._crc32(c, self.key0) |
| 556 | self.key1 = (self.key1 + (self.key0 & 255)) & 4294967295 |
| 557 | self.key1 = (self.key1 * 134775813 + 1) & 4294967295 |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 558 | self.key2 = self._crc32((self.key1 >> 24) & 255, self.key2) |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 559 | |
| 560 | def __call__(self, c): |
| 561 | """Decrypt a single character.""" |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 562 | assert isinstance(c, int) |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 563 | k = self.key2 | 2 |
| 564 | c = c ^ (((k * (k^1)) >> 8) & 255) |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 565 | self._UpdateKeys(c) |
| 566 | return c |
| 567 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 568 | |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 569 | class LZMACompressor: |
| 570 | |
| 571 | def __init__(self): |
| 572 | self._comp = None |
| 573 | |
| 574 | def _init(self): |
Nadeem Vawda | a425c3d | 2012-06-21 23:36:48 +0200 | [diff] [blame] | 575 | props = lzma._encode_filter_properties({'id': lzma.FILTER_LZMA1}) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 576 | self._comp = lzma.LZMACompressor(lzma.FORMAT_RAW, filters=[ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 577 | lzma._decode_filter_properties(lzma.FILTER_LZMA1, props) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 578 | ]) |
| 579 | return struct.pack('<BBH', 9, 4, len(props)) + props |
| 580 | |
| 581 | def compress(self, data): |
| 582 | if self._comp is None: |
| 583 | return self._init() + self._comp.compress(data) |
| 584 | return self._comp.compress(data) |
| 585 | |
| 586 | def flush(self): |
| 587 | if self._comp is None: |
| 588 | return self._init() + self._comp.flush() |
| 589 | return self._comp.flush() |
| 590 | |
| 591 | |
| 592 | class LZMADecompressor: |
| 593 | |
| 594 | def __init__(self): |
| 595 | self._decomp = None |
| 596 | self._unconsumed = b'' |
| 597 | self.eof = False |
| 598 | |
| 599 | def decompress(self, data): |
| 600 | if self._decomp is None: |
| 601 | self._unconsumed += data |
| 602 | if len(self._unconsumed) <= 4: |
| 603 | return b'' |
| 604 | psize, = struct.unpack('<H', self._unconsumed[2:4]) |
| 605 | if len(self._unconsumed) <= 4 + psize: |
| 606 | return b'' |
| 607 | |
| 608 | self._decomp = lzma.LZMADecompressor(lzma.FORMAT_RAW, filters=[ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 609 | lzma._decode_filter_properties(lzma.FILTER_LZMA1, |
| 610 | self._unconsumed[4:4 + psize]) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 611 | ]) |
| 612 | data = self._unconsumed[4 + psize:] |
| 613 | del self._unconsumed |
| 614 | |
| 615 | result = self._decomp.decompress(data) |
| 616 | self.eof = self._decomp.eof |
| 617 | return result |
| 618 | |
| 619 | |
| 620 | compressor_names = { |
| 621 | 0: 'store', |
| 622 | 1: 'shrink', |
| 623 | 2: 'reduce', |
| 624 | 3: 'reduce', |
| 625 | 4: 'reduce', |
| 626 | 5: 'reduce', |
| 627 | 6: 'implode', |
| 628 | 7: 'tokenize', |
| 629 | 8: 'deflate', |
| 630 | 9: 'deflate64', |
| 631 | 10: 'implode', |
| 632 | 12: 'bzip2', |
| 633 | 14: 'lzma', |
| 634 | 18: 'terse', |
| 635 | 19: 'lz77', |
| 636 | 97: 'wavpack', |
| 637 | 98: 'ppmd', |
| 638 | } |
| 639 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 640 | def _check_compression(compression): |
| 641 | if compression == ZIP_STORED: |
| 642 | pass |
| 643 | elif compression == ZIP_DEFLATED: |
| 644 | if not zlib: |
| 645 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 646 | "Compression requires the (missing) zlib module") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 647 | elif compression == ZIP_BZIP2: |
| 648 | if not bz2: |
| 649 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 650 | "Compression requires the (missing) bz2 module") |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 651 | elif compression == ZIP_LZMA: |
| 652 | if not lzma: |
| 653 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 654 | "Compression requires the (missing) lzma module") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 655 | else: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 656 | raise NotImplementedError("That compression method is not supported") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 657 | |
| 658 | |
| 659 | def _get_compressor(compress_type): |
| 660 | if compress_type == ZIP_DEFLATED: |
| 661 | return zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 662 | zlib.DEFLATED, -15) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 663 | elif compress_type == ZIP_BZIP2: |
| 664 | return bz2.BZ2Compressor() |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 665 | elif compress_type == ZIP_LZMA: |
| 666 | return LZMACompressor() |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 667 | else: |
| 668 | return None |
| 669 | |
| 670 | |
| 671 | def _get_decompressor(compress_type): |
Martin v. Löwis | b3260f0 | 2012-05-01 08:38:01 +0200 | [diff] [blame] | 672 | if compress_type == ZIP_STORED: |
| 673 | return None |
| 674 | elif compress_type == ZIP_DEFLATED: |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 675 | return zlib.decompressobj(-15) |
| 676 | elif compress_type == ZIP_BZIP2: |
| 677 | return bz2.BZ2Decompressor() |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 678 | elif compress_type == ZIP_LZMA: |
| 679 | return LZMADecompressor() |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 680 | else: |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 681 | descr = compressor_names.get(compress_type) |
Martin v. Löwis | b3260f0 | 2012-05-01 08:38:01 +0200 | [diff] [blame] | 682 | if descr: |
| 683 | raise NotImplementedError("compression type %d (%s)" % (compress_type, descr)) |
| 684 | else: |
| 685 | raise NotImplementedError("compression type %d" % (compress_type,)) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 686 | |
| 687 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 688 | class _SharedFile: |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 689 | def __init__(self, file, pos, close, lock, writing): |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 690 | self._file = file |
| 691 | self._pos = pos |
| 692 | self._close = close |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 693 | self._lock = lock |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 694 | self._writing = writing |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 695 | |
| 696 | def read(self, n=-1): |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 697 | with self._lock: |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 698 | if self._writing(): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 699 | raise ValueError("Can't read from the ZIP file while there " |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 700 | "is an open writing handle on it. " |
| 701 | "Close the writing handle before trying to read.") |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 702 | self._file.seek(self._pos) |
| 703 | data = self._file.read(n) |
| 704 | self._pos = self._file.tell() |
| 705 | return data |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 706 | |
| 707 | def close(self): |
| 708 | if self._file is not None: |
| 709 | fileobj = self._file |
| 710 | self._file = None |
| 711 | self._close(fileobj) |
| 712 | |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 713 | # Provide the tell method for unseekable stream |
| 714 | class _Tellable: |
| 715 | def __init__(self, fp): |
| 716 | self.fp = fp |
| 717 | self.offset = 0 |
| 718 | |
| 719 | def write(self, data): |
| 720 | n = self.fp.write(data) |
| 721 | self.offset += n |
| 722 | return n |
| 723 | |
| 724 | def tell(self): |
| 725 | return self.offset |
| 726 | |
| 727 | def flush(self): |
| 728 | self.fp.flush() |
| 729 | |
| 730 | def close(self): |
| 731 | self.fp.close() |
| 732 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 733 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 734 | class ZipExtFile(io.BufferedIOBase): |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 735 | """File-like object for reading an archive member. |
| 736 | Is returned by ZipFile.open(). |
| 737 | """ |
| 738 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 739 | # Max size supported by decompressor. |
| 740 | MAX_N = 1 << 31 - 1 |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 741 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 742 | # Read from compressed files in 4k blocks. |
| 743 | MIN_READ_SIZE = 4096 |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 744 | |
Łukasz Langa | e94980a | 2010-11-22 23:31:26 +0000 | [diff] [blame] | 745 | def __init__(self, fileobj, mode, zipinfo, decrypter=None, |
| 746 | close_fileobj=False): |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 747 | self._fileobj = fileobj |
| 748 | self._decrypter = decrypter |
Łukasz Langa | e94980a | 2010-11-22 23:31:26 +0000 | [diff] [blame] | 749 | self._close_fileobj = close_fileobj |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 750 | |
Ezio Melotti | 92b4743 | 2010-01-28 01:44:41 +0000 | [diff] [blame] | 751 | self._compress_type = zipinfo.compress_type |
Ezio Melotti | 92b4743 | 2010-01-28 01:44:41 +0000 | [diff] [blame] | 752 | self._compress_left = zipinfo.compress_size |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 753 | self._left = zipinfo.file_size |
Ezio Melotti | 92b4743 | 2010-01-28 01:44:41 +0000 | [diff] [blame] | 754 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 755 | self._decompressor = _get_decompressor(self._compress_type) |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 756 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 757 | self._eof = False |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 758 | self._readbuffer = b'' |
| 759 | self._offset = 0 |
| 760 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 761 | self.newlines = None |
| 762 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 763 | # Adjust read size for encrypted files since the first 12 bytes |
| 764 | # are for the encryption/password information. |
| 765 | if self._decrypter is not None: |
| 766 | self._compress_left -= 12 |
| 767 | |
| 768 | self.mode = mode |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 769 | self.name = zipinfo.filename |
| 770 | |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 771 | if hasattr(zipinfo, 'CRC'): |
| 772 | self._expected_crc = zipinfo.CRC |
Martin Panter | b82032f | 2015-12-11 05:19:29 +0000 | [diff] [blame] | 773 | self._running_crc = crc32(b'') |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 774 | else: |
| 775 | self._expected_crc = None |
| 776 | |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 777 | def __repr__(self): |
| 778 | result = ['<%s.%s' % (self.__class__.__module__, |
| 779 | self.__class__.__qualname__)] |
| 780 | if not self.closed: |
| 781 | result.append(' name=%r mode=%r' % (self.name, self.mode)) |
| 782 | if self._compress_type != ZIP_STORED: |
| 783 | result.append(' compress_type=%s' % |
| 784 | compressor_names.get(self._compress_type, |
| 785 | self._compress_type)) |
| 786 | else: |
| 787 | result.append(' [closed]') |
| 788 | result.append('>') |
| 789 | return ''.join(result) |
| 790 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 791 | def readline(self, limit=-1): |
| 792 | """Read and return a line from the stream. |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 793 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 794 | If limit is specified, at most limit bytes will be read. |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 795 | """ |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 796 | |
Serhiy Storchaka | e670be2 | 2016-06-11 19:32:44 +0300 | [diff] [blame] | 797 | if limit < 0: |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 798 | # Shortcut common case - newline found in buffer. |
| 799 | i = self._readbuffer.find(b'\n', self._offset) + 1 |
| 800 | if i > 0: |
| 801 | line = self._readbuffer[self._offset: i] |
| 802 | self._offset = i |
| 803 | return line |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 804 | |
Serhiy Storchaka | e670be2 | 2016-06-11 19:32:44 +0300 | [diff] [blame] | 805 | return io.BufferedIOBase.readline(self, limit) |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 806 | |
| 807 | def peek(self, n=1): |
| 808 | """Returns buffered bytes without advancing the position.""" |
| 809 | if n > len(self._readbuffer) - self._offset: |
| 810 | chunk = self.read(n) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 811 | if len(chunk) > self._offset: |
| 812 | self._readbuffer = chunk + self._readbuffer[self._offset:] |
| 813 | self._offset = 0 |
| 814 | else: |
| 815 | self._offset -= len(chunk) |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 816 | |
| 817 | # Return up to 512 bytes to reduce allocation overhead for tight loops. |
| 818 | return self._readbuffer[self._offset: self._offset + 512] |
| 819 | |
| 820 | def readable(self): |
| 821 | return True |
| 822 | |
| 823 | def read(self, n=-1): |
| 824 | """Read and return up to n bytes. |
| 825 | If the argument is omitted, None, or negative, data is read and returned until EOF is reached.. |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 826 | """ |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 827 | if n is None or n < 0: |
| 828 | buf = self._readbuffer[self._offset:] |
| 829 | self._readbuffer = b'' |
| 830 | self._offset = 0 |
| 831 | while not self._eof: |
| 832 | buf += self._read1(self.MAX_N) |
| 833 | return buf |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 834 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 835 | end = n + self._offset |
| 836 | if end < len(self._readbuffer): |
| 837 | buf = self._readbuffer[self._offset:end] |
| 838 | self._offset = end |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 839 | return buf |
| 840 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 841 | n = end - len(self._readbuffer) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 842 | buf = self._readbuffer[self._offset:] |
| 843 | self._readbuffer = b'' |
| 844 | self._offset = 0 |
| 845 | while n > 0 and not self._eof: |
| 846 | data = self._read1(n) |
| 847 | if n < len(data): |
| 848 | self._readbuffer = data |
| 849 | self._offset = n |
| 850 | buf += data[:n] |
| 851 | break |
| 852 | buf += data |
| 853 | n -= len(data) |
| 854 | return buf |
| 855 | |
| 856 | def _update_crc(self, newdata): |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 857 | # Update the CRC using the given data. |
| 858 | if self._expected_crc is None: |
| 859 | # No need to compute the CRC if we don't have a reference value |
| 860 | return |
Martin Panter | b82032f | 2015-12-11 05:19:29 +0000 | [diff] [blame] | 861 | self._running_crc = crc32(newdata, self._running_crc) |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 862 | # Check the CRC if we're at the end of the file |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 863 | if self._eof and self._running_crc != self._expected_crc: |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 864 | raise BadZipFile("Bad CRC-32 for file %r" % self.name) |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 865 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 866 | def read1(self, n): |
| 867 | """Read up to n bytes with at most one read() system call.""" |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 868 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 869 | if n is None or n < 0: |
| 870 | buf = self._readbuffer[self._offset:] |
| 871 | self._readbuffer = b'' |
| 872 | self._offset = 0 |
Serhiy Storchaka | d2c07a5 | 2013-09-27 22:11:57 +0300 | [diff] [blame] | 873 | while not self._eof: |
| 874 | data = self._read1(self.MAX_N) |
| 875 | if data: |
| 876 | buf += data |
| 877 | break |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 878 | return buf |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 879 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 880 | end = n + self._offset |
| 881 | if end < len(self._readbuffer): |
| 882 | buf = self._readbuffer[self._offset:end] |
| 883 | self._offset = end |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 884 | return buf |
| 885 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 886 | n = end - len(self._readbuffer) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 887 | buf = self._readbuffer[self._offset:] |
| 888 | self._readbuffer = b'' |
| 889 | self._offset = 0 |
| 890 | if n > 0: |
Serhiy Storchaka | d2c07a5 | 2013-09-27 22:11:57 +0300 | [diff] [blame] | 891 | while not self._eof: |
| 892 | data = self._read1(n) |
| 893 | if n < len(data): |
| 894 | self._readbuffer = data |
| 895 | self._offset = n |
| 896 | buf += data[:n] |
| 897 | break |
| 898 | if data: |
| 899 | buf += data |
| 900 | break |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 901 | return buf |
| 902 | |
| 903 | def _read1(self, n): |
| 904 | # Read up to n compressed bytes with at most one read() system call, |
| 905 | # decrypt and decompress them. |
| 906 | if self._eof or n <= 0: |
| 907 | return b'' |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 908 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 909 | # Read from file. |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 910 | if self._compress_type == ZIP_DEFLATED: |
| 911 | ## Handle unconsumed data. |
| 912 | data = self._decompressor.unconsumed_tail |
| 913 | if n > len(data): |
| 914 | data += self._read2(n - len(data)) |
| 915 | else: |
| 916 | data = self._read2(n) |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 917 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 918 | if self._compress_type == ZIP_STORED: |
| 919 | self._eof = self._compress_left <= 0 |
| 920 | elif self._compress_type == ZIP_DEFLATED: |
| 921 | n = max(n, self.MIN_READ_SIZE) |
| 922 | data = self._decompressor.decompress(data, n) |
| 923 | self._eof = (self._decompressor.eof or |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 924 | self._compress_left <= 0 and |
| 925 | not self._decompressor.unconsumed_tail) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 926 | if self._eof: |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 927 | data += self._decompressor.flush() |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 928 | else: |
| 929 | data = self._decompressor.decompress(data) |
| 930 | self._eof = self._decompressor.eof or self._compress_left <= 0 |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 931 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 932 | data = data[:self._left] |
| 933 | self._left -= len(data) |
| 934 | if self._left <= 0: |
| 935 | self._eof = True |
| 936 | self._update_crc(data) |
| 937 | return data |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 938 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 939 | def _read2(self, n): |
| 940 | if self._compress_left <= 0: |
| 941 | return b'' |
| 942 | |
| 943 | n = max(n, self.MIN_READ_SIZE) |
| 944 | n = min(n, self._compress_left) |
| 945 | |
| 946 | data = self._fileobj.read(n) |
| 947 | self._compress_left -= len(data) |
Serhiy Storchaka | 5ce3f10 | 2014-01-09 14:50:20 +0200 | [diff] [blame] | 948 | if not data: |
| 949 | raise EOFError |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 950 | |
| 951 | if self._decrypter is not None: |
| 952 | data = bytes(map(self._decrypter, data)) |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 953 | return data |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 954 | |
Łukasz Langa | e94980a | 2010-11-22 23:31:26 +0000 | [diff] [blame] | 955 | def close(self): |
| 956 | try: |
| 957 | if self._close_fileobj: |
| 958 | self._fileobj.close() |
| 959 | finally: |
| 960 | super().close() |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 961 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 962 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 963 | class _ZipWriteFile(io.BufferedIOBase): |
| 964 | def __init__(self, zf, zinfo, zip64): |
| 965 | self._zinfo = zinfo |
| 966 | self._zip64 = zip64 |
| 967 | self._zipfile = zf |
| 968 | self._compressor = _get_compressor(zinfo.compress_type) |
| 969 | self._file_size = 0 |
| 970 | self._compress_size = 0 |
| 971 | self._crc = 0 |
| 972 | |
| 973 | @property |
| 974 | def _fileobj(self): |
| 975 | return self._zipfile.fp |
| 976 | |
| 977 | def writable(self): |
| 978 | return True |
| 979 | |
| 980 | def write(self, data): |
| 981 | nbytes = len(data) |
| 982 | self._file_size += nbytes |
| 983 | self._crc = crc32(data, self._crc) |
| 984 | if self._compressor: |
| 985 | data = self._compressor.compress(data) |
| 986 | self._compress_size += len(data) |
| 987 | self._fileobj.write(data) |
| 988 | return nbytes |
| 989 | |
| 990 | def close(self): |
| 991 | super().close() |
| 992 | # Flush any data from the compressor, and update header info |
| 993 | if self._compressor: |
| 994 | buf = self._compressor.flush() |
| 995 | self._compress_size += len(buf) |
| 996 | self._fileobj.write(buf) |
| 997 | self._zinfo.compress_size = self._compress_size |
| 998 | else: |
| 999 | self._zinfo.compress_size = self._file_size |
| 1000 | self._zinfo.CRC = self._crc |
| 1001 | self._zinfo.file_size = self._file_size |
| 1002 | |
| 1003 | # Write updated header info |
| 1004 | if self._zinfo.flag_bits & 0x08: |
| 1005 | # Write CRC and file sizes after the file data |
| 1006 | fmt = '<LQQ' if self._zip64 else '<LLL' |
| 1007 | self._fileobj.write(struct.pack(fmt, self._zinfo.CRC, |
| 1008 | self._zinfo.compress_size, self._zinfo.file_size)) |
| 1009 | self._zipfile.start_dir = self._fileobj.tell() |
| 1010 | else: |
| 1011 | if not self._zip64: |
| 1012 | if self._file_size > ZIP64_LIMIT: |
| 1013 | raise RuntimeError('File size unexpectedly exceeded ZIP64 ' |
| 1014 | 'limit') |
| 1015 | if self._compress_size > ZIP64_LIMIT: |
| 1016 | raise RuntimeError('Compressed size unexpectedly exceeded ' |
| 1017 | 'ZIP64 limit') |
| 1018 | # Seek backwards and write file header (which will now include |
| 1019 | # correct CRC and file sizes) |
| 1020 | |
| 1021 | # Preserve current position in file |
| 1022 | self._zipfile.start_dir = self._fileobj.tell() |
| 1023 | self._fileobj.seek(self._zinfo.header_offset) |
| 1024 | self._fileobj.write(self._zinfo.FileHeader(self._zip64)) |
| 1025 | self._fileobj.seek(self._zipfile.start_dir) |
| 1026 | |
| 1027 | self._zipfile._writing = False |
| 1028 | |
| 1029 | # Successfully written: Add file to our caches |
| 1030 | self._zipfile.filelist.append(self._zinfo) |
| 1031 | self._zipfile.NameToInfo[self._zinfo.filename] = self._zinfo |
| 1032 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1033 | class ZipFile: |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1034 | """ Class with methods to open, read, write, close, list zip files. |
| 1035 | |
Serhiy Storchaka | 235c5e0 | 2013-11-23 15:55:38 +0200 | [diff] [blame] | 1036 | z = ZipFile(file, mode="r", compression=ZIP_STORED, allowZip64=True) |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1037 | |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1038 | file: Either the path to the file, or a file-like object. |
| 1039 | If it is a path, the file will be opened and closed by ZipFile. |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1040 | mode: The mode can be either read 'r', write 'w', exclusive create 'x', |
| 1041 | or append 'a'. |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 1042 | compression: ZIP_STORED (no compression), ZIP_DEFLATED (requires zlib), |
| 1043 | ZIP_BZIP2 (requires bz2) or ZIP_LZMA (requires lzma). |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1044 | allowZip64: if True ZipFile will create files with ZIP64 extensions when |
| 1045 | needed, otherwise it will raise an exception when this would |
| 1046 | be necessary. |
| 1047 | |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1048 | """ |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1049 | |
Fred Drake | 90eac28 | 2001-02-28 05:29:34 +0000 | [diff] [blame] | 1050 | fp = None # Set here since __del__ checks it |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1051 | _windows_illegal_name_trans_table = None |
Fred Drake | 90eac28 | 2001-02-28 05:29:34 +0000 | [diff] [blame] | 1052 | |
Serhiy Storchaka | 235c5e0 | 2013-11-23 15:55:38 +0200 | [diff] [blame] | 1053 | def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=True): |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1054 | """Open the ZIP file with mode read 'r', write 'w', exclusive create 'x', |
| 1055 | or append 'a'.""" |
| 1056 | if mode not in ('r', 'w', 'x', 'a'): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1057 | raise ValueError("ZipFile requires mode 'r', 'w', 'x', or 'a'") |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1058 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 1059 | _check_compression(compression) |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1060 | |
| 1061 | self._allowZip64 = allowZip64 |
| 1062 | self._didModify = False |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 1063 | self.debug = 0 # Level of printing: 0 through 3 |
| 1064 | self.NameToInfo = {} # Find file info given name |
| 1065 | self.filelist = [] # List of ZipInfo instances for archive |
| 1066 | self.compression = compression # Method of compression |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1067 | self.mode = mode |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1068 | self.pwd = None |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1069 | self._comment = b'' |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1070 | |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1071 | # Check if we were passed a file-like object |
Guido van Rossum | 3172c5d | 2007-10-16 18:12:55 +0000 | [diff] [blame] | 1072 | if isinstance(file, str): |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1073 | # No, it's a filename |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1074 | self._filePassed = 0 |
| 1075 | self.filename = file |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1076 | modeDict = {'r' : 'rb', 'w': 'w+b', 'x': 'x+b', 'a' : 'r+b', |
| 1077 | 'r+b': 'w+b', 'w+b': 'wb', 'x+b': 'xb'} |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1078 | filemode = modeDict[mode] |
| 1079 | while True: |
| 1080 | try: |
| 1081 | self.fp = io.open(file, filemode) |
| 1082 | except OSError: |
| 1083 | if filemode in modeDict: |
| 1084 | filemode = modeDict[filemode] |
| 1085 | continue |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1086 | raise |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1087 | break |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1088 | else: |
| 1089 | self._filePassed = 1 |
| 1090 | self.fp = file |
| 1091 | self.filename = getattr(file, 'name', None) |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1092 | self._fileRefCnt = 1 |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1093 | self._lock = threading.RLock() |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1094 | self._seekable = True |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1095 | self._writing = False |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1096 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1097 | try: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1098 | if mode == 'r': |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 1099 | self._RealGetContents() |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1100 | elif mode in ('w', 'x'): |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 1101 | # set the modified flag so central directory gets written |
| 1102 | # even if no files are added to the archive |
| 1103 | self._didModify = True |
Serhiy Storchaka | 34cba33 | 2017-01-01 19:00:30 +0200 | [diff] [blame] | 1104 | self._start_disk = 0 |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1105 | try: |
Serhiy Storchaka | 34cba33 | 2017-01-01 19:00:30 +0200 | [diff] [blame] | 1106 | self.start_dir = self.fp.tell() |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1107 | except (AttributeError, OSError): |
| 1108 | self.fp = _Tellable(self.fp) |
Serhiy Storchaka | 34cba33 | 2017-01-01 19:00:30 +0200 | [diff] [blame] | 1109 | self.start_dir = 0 |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1110 | self._seekable = False |
| 1111 | else: |
| 1112 | # Some file-like objects can provide tell() but not seek() |
| 1113 | try: |
| 1114 | self.fp.seek(self.start_dir) |
| 1115 | except (AttributeError, OSError): |
| 1116 | self._seekable = False |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1117 | elif mode == 'a': |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1118 | try: |
| 1119 | # See if file is a zip file |
| 1120 | self._RealGetContents() |
| 1121 | # seek to start of directory and overwrite |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1122 | self.fp.seek(self.start_dir) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1123 | except BadZipFile: |
| 1124 | # file is not a zip file, just append |
| 1125 | self.fp.seek(0, 2) |
| 1126 | |
| 1127 | # set the modified flag so central directory gets written |
| 1128 | # even if no files are added to the archive |
| 1129 | self._didModify = True |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1130 | self.start_dir = self._start_disk = self.fp.tell() |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1131 | else: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1132 | raise ValueError("Mode must be 'r', 'w', 'x', or 'a'") |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1133 | except: |
| 1134 | fp = self.fp |
| 1135 | self.fp = None |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1136 | self._fpclose(fp) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1137 | raise |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1138 | |
Ezio Melotti | faa6b7f | 2009-12-30 12:34:59 +0000 | [diff] [blame] | 1139 | def __enter__(self): |
| 1140 | return self |
| 1141 | |
| 1142 | def __exit__(self, type, value, traceback): |
| 1143 | self.close() |
| 1144 | |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 1145 | def __repr__(self): |
| 1146 | result = ['<%s.%s' % (self.__class__.__module__, |
| 1147 | self.__class__.__qualname__)] |
| 1148 | if self.fp is not None: |
| 1149 | if self._filePassed: |
| 1150 | result.append(' file=%r' % self.fp) |
| 1151 | elif self.filename is not None: |
| 1152 | result.append(' filename=%r' % self.filename) |
| 1153 | result.append(' mode=%r' % self.mode) |
| 1154 | else: |
| 1155 | result.append(' [closed]') |
| 1156 | result.append('>') |
| 1157 | return ''.join(result) |
| 1158 | |
Tim Peters | 7d3bad6 | 2001-04-04 18:56:49 +0000 | [diff] [blame] | 1159 | def _RealGetContents(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1160 | """Read in the table of contents for the ZIP file.""" |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1161 | fp = self.fp |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 1162 | try: |
| 1163 | endrec = _EndRecData(fp) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 1164 | except OSError: |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 1165 | raise BadZipFile("File is not a zip file") |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 1166 | if not endrec: |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 1167 | raise BadZipFile("File is not a zip file") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1168 | if self.debug > 1: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1169 | print(endrec) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1170 | size_cd = endrec[_ECD_SIZE] # bytes in central directory |
| 1171 | offset_cd = endrec[_ECD_OFFSET] # offset of central directory |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1172 | self._comment = endrec[_ECD_COMMENT] # archive comment |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1173 | |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1174 | # self._start_disk: Position of the start of ZIP archive |
| 1175 | # It is zero, unless ZIP was concatenated to another file |
| 1176 | self._start_disk = endrec[_ECD_LOCATION] - size_cd - offset_cd |
Antoine Pitrou | 9e4fdf4 | 2008-09-05 23:43:02 +0000 | [diff] [blame] | 1177 | if endrec[_ECD_SIGNATURE] == stringEndArchive64: |
| 1178 | # If Zip64 extension structures are present, account for them |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1179 | self._start_disk -= (sizeEndCentDir64 + sizeEndCentDir64Locator) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1180 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1181 | if self.debug > 2: |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1182 | inferred = self._start_disk + offset_cd |
| 1183 | print("given, inferred, offset", offset_cd, inferred, self._start_disk) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1184 | # self.start_dir: Position of start of central directory |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1185 | self.start_dir = offset_cd + self._start_disk |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1186 | fp.seek(self.start_dir, 0) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1187 | data = fp.read(size_cd) |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1188 | fp = io.BytesIO(data) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1189 | total = 0 |
| 1190 | while total < size_cd: |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1191 | centdir = fp.read(sizeCentralDir) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 1192 | if len(centdir) != sizeCentralDir: |
| 1193 | raise BadZipFile("Truncated central directory") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1194 | centdir = struct.unpack(structCentralDir, centdir) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 1195 | if centdir[_CD_SIGNATURE] != stringCentralDir: |
| 1196 | raise BadZipFile("Bad magic number for central directory") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1197 | if self.debug > 2: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1198 | print(centdir) |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 1199 | filename = fp.read(centdir[_CD_FILENAME_LENGTH]) |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 1200 | flags = centdir[5] |
| 1201 | if flags & 0x800: |
| 1202 | # UTF-8 file names extension |
| 1203 | filename = filename.decode('utf-8') |
| 1204 | else: |
| 1205 | # Historical ZIP filename encoding |
| 1206 | filename = filename.decode('cp437') |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1207 | # Create ZipInfo instance to store file information |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 1208 | x = ZipInfo(filename) |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 1209 | x.extra = fp.read(centdir[_CD_EXTRA_FIELD_LENGTH]) |
| 1210 | x.comment = fp.read(centdir[_CD_COMMENT_LENGTH]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1211 | x.header_offset = centdir[_CD_LOCAL_HEADER_OFFSET] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1212 | (x.create_version, x.create_system, x.extract_version, x.reserved, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1213 | x.flag_bits, x.compress_type, t, d, |
| 1214 | x.CRC, x.compress_size, x.file_size) = centdir[1:12] |
Martin v. Löwis | d099b56 | 2012-05-01 14:08:22 +0200 | [diff] [blame] | 1215 | if x.extract_version > MAX_EXTRACT_VERSION: |
| 1216 | raise NotImplementedError("zip file version %.1f" % |
| 1217 | (x.extract_version / 10)) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1218 | x.volume, x.internal_attr, x.external_attr = centdir[15:18] |
| 1219 | # Convert date/time code to (year, month, day, hour, min, sec) |
Christian Heimes | fdab48e | 2008-01-20 09:06:41 +0000 | [diff] [blame] | 1220 | x._raw_time = t |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1221 | x.date_time = ( (d>>9)+1980, (d>>5)&0xF, d&0x1F, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1222 | t>>11, (t>>5)&0x3F, (t&0x1F) * 2 ) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1223 | |
| 1224 | x._decodeExtra() |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1225 | x.header_offset = x.header_offset + self._start_disk |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1226 | self.filelist.append(x) |
| 1227 | self.NameToInfo[x.filename] = x |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1228 | |
| 1229 | # update total bytes read from central directory |
| 1230 | total = (total + sizeCentralDir + centdir[_CD_FILENAME_LENGTH] |
| 1231 | + centdir[_CD_EXTRA_FIELD_LENGTH] |
| 1232 | + centdir[_CD_COMMENT_LENGTH]) |
| 1233 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1234 | if self.debug > 2: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1235 | print("total", total) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1236 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1237 | |
| 1238 | def namelist(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1239 | """Return a list of file names in the archive.""" |
Ezio Melotti | 006917e | 2012-04-16 21:34:24 -0600 | [diff] [blame] | 1240 | return [data.filename for data in self.filelist] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1241 | |
| 1242 | def infolist(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1243 | """Return a list of class ZipInfo instances for files in the |
| 1244 | archive.""" |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1245 | return self.filelist |
| 1246 | |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1247 | def printdir(self, file=None): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1248 | """Print a table of contents for the zip file.""" |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1249 | print("%-46s %19s %12s" % ("File Name", "Modified ", "Size"), |
| 1250 | file=file) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1251 | for zinfo in self.filelist: |
Guido van Rossum | 7736b5b | 2008-01-15 21:44:53 +0000 | [diff] [blame] | 1252 | date = "%d-%02d-%02d %02d:%02d:%02d" % zinfo.date_time[:6] |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1253 | print("%-46s %s %12d" % (zinfo.filename, date, zinfo.file_size), |
| 1254 | file=file) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1255 | |
| 1256 | def testzip(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1257 | """Read all the files and check the CRC.""" |
Benjamin Peterson | 4cd6a95 | 2008-08-17 20:23:46 +0000 | [diff] [blame] | 1258 | chunk_size = 2 ** 20 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1259 | for zinfo in self.filelist: |
| 1260 | try: |
Benjamin Peterson | 4cd6a95 | 2008-08-17 20:23:46 +0000 | [diff] [blame] | 1261 | # Read by chunks, to avoid an OverflowError or a |
| 1262 | # MemoryError with very large embedded files. |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1263 | with self.open(zinfo.filename, "r") as f: |
| 1264 | while f.read(chunk_size): # Check CRC-32 |
| 1265 | pass |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 1266 | except BadZipFile: |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1267 | return zinfo.filename |
| 1268 | |
| 1269 | def getinfo(self, name): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1270 | """Return the instance of ZipInfo given 'name'.""" |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1271 | info = self.NameToInfo.get(name) |
| 1272 | if info is None: |
| 1273 | raise KeyError( |
| 1274 | 'There is no item named %r in the archive' % name) |
| 1275 | |
| 1276 | return info |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1277 | |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1278 | def setpassword(self, pwd): |
| 1279 | """Set default password for encrypted files.""" |
R. David Murray | 8d855d8 | 2010-12-21 21:53:37 +0000 | [diff] [blame] | 1280 | if pwd and not isinstance(pwd, bytes): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1281 | raise TypeError("pwd: expected bytes, got %s" % type(pwd).__name__) |
R. David Murray | 8d855d8 | 2010-12-21 21:53:37 +0000 | [diff] [blame] | 1282 | if pwd: |
| 1283 | self.pwd = pwd |
| 1284 | else: |
| 1285 | self.pwd = None |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1286 | |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1287 | @property |
| 1288 | def comment(self): |
| 1289 | """The comment text associated with the ZIP file.""" |
| 1290 | return self._comment |
| 1291 | |
| 1292 | @comment.setter |
| 1293 | def comment(self, comment): |
| 1294 | if not isinstance(comment, bytes): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1295 | raise TypeError("comment: expected bytes, got %s" % type(comment).__name__) |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1296 | # check for valid comment length |
Serhiy Storchaka | 9b7a1a1 | 2014-01-20 21:57:40 +0200 | [diff] [blame] | 1297 | if len(comment) > ZIP_MAX_COMMENT: |
| 1298 | import warnings |
| 1299 | warnings.warn('Archive comment is too long; truncating to %d bytes' |
| 1300 | % ZIP_MAX_COMMENT, stacklevel=2) |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1301 | comment = comment[:ZIP_MAX_COMMENT] |
| 1302 | self._comment = comment |
| 1303 | self._didModify = True |
| 1304 | |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1305 | def read(self, name, pwd=None): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1306 | """Return file bytes (as a string) for name.""" |
Benjamin Peterson | d285bdb | 2010-10-31 17:57:22 +0000 | [diff] [blame] | 1307 | with self.open(name, "r", pwd) as fp: |
| 1308 | return fp.read() |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1309 | |
Serhiy Storchaka | f47fc55 | 2016-05-15 12:27:16 +0300 | [diff] [blame] | 1310 | def open(self, name, mode="r", pwd=None, *, force_zip64=False): |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1311 | """Return file-like object for 'name'. |
| 1312 | |
| 1313 | name is a string for the file name within the ZIP file, or a ZipInfo |
| 1314 | object. |
| 1315 | |
| 1316 | mode should be 'r' to read a file already in the ZIP file, or 'w' to |
| 1317 | write to a file newly added to the archive. |
| 1318 | |
| 1319 | pwd is the password to decrypt files (only used for reading). |
| 1320 | |
| 1321 | When writing, if the file size is not known in advance but may exceed |
| 1322 | 2 GiB, pass force_zip64 to use the ZIP64 format, which can handle large |
| 1323 | files. If the size is known in advance, it is best to pass a ZipInfo |
| 1324 | instance for name, with zinfo.file_size set. |
| 1325 | """ |
Serhiy Storchaka | e670be2 | 2016-06-11 19:32:44 +0300 | [diff] [blame] | 1326 | if mode not in {"r", "w"}: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1327 | raise ValueError('open() requires mode "r" or "w"') |
R. David Murray | 8d855d8 | 2010-12-21 21:53:37 +0000 | [diff] [blame] | 1328 | if pwd and not isinstance(pwd, bytes): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1329 | raise TypeError("pwd: expected bytes, got %s" % type(pwd).__name__) |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1330 | if pwd and (mode == "w"): |
| 1331 | raise ValueError("pwd is only supported for reading files") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1332 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1333 | raise ValueError( |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1334 | "Attempt to use ZIP archive that was already closed") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1335 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1336 | # Make sure we have an info object |
| 1337 | if isinstance(name, ZipInfo): |
| 1338 | # 'name' is already an info object |
| 1339 | zinfo = name |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1340 | elif mode == 'w': |
| 1341 | zinfo = ZipInfo(name) |
| 1342 | zinfo.compress_type = self.compression |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1343 | else: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1344 | # Get info object for name |
| 1345 | zinfo = self.getinfo(name) |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1346 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1347 | if mode == 'w': |
| 1348 | return self._open_to_write(zinfo, force_zip64=force_zip64) |
| 1349 | |
| 1350 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1351 | raise ValueError("Can't read from the ZIP file while there " |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1352 | "is an open writing handle on it. " |
| 1353 | "Close the writing handle before trying to read.") |
| 1354 | |
| 1355 | # Open for reading: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1356 | self._fileRefCnt += 1 |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1357 | zef_file = _SharedFile(self.fp, zinfo.header_offset, |
| 1358 | self._fpclose, self._lock, lambda: self._writing) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1359 | try: |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1360 | # Skip the file header: |
| 1361 | fheader = zef_file.read(sizeFileHeader) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 1362 | if len(fheader) != sizeFileHeader: |
| 1363 | raise BadZipFile("Truncated file header") |
| 1364 | fheader = struct.unpack(structFileHeader, fheader) |
| 1365 | if fheader[_FH_SIGNATURE] != stringFileHeader: |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1366 | raise BadZipFile("Bad magic number for file header") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1367 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1368 | fname = zef_file.read(fheader[_FH_FILENAME_LENGTH]) |
| 1369 | if fheader[_FH_EXTRA_FIELD_LENGTH]: |
| 1370 | zef_file.read(fheader[_FH_EXTRA_FIELD_LENGTH]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1371 | |
Antoine Pitrou | 8572da5 | 2012-11-17 23:52:05 +0100 | [diff] [blame] | 1372 | if zinfo.flag_bits & 0x20: |
| 1373 | # Zip 2.7: compressed patched data |
| 1374 | raise NotImplementedError("compressed patched data (flag bit 5)") |
Martin v. Löwis | 2a2ce32 | 2012-05-01 08:44:08 +0200 | [diff] [blame] | 1375 | |
Antoine Pitrou | 8572da5 | 2012-11-17 23:52:05 +0100 | [diff] [blame] | 1376 | if zinfo.flag_bits & 0x40: |
| 1377 | # strong encryption |
| 1378 | raise NotImplementedError("strong encryption (flag bit 6)") |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 1379 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1380 | if zinfo.flag_bits & 0x800: |
| 1381 | # UTF-8 filename |
| 1382 | fname_str = fname.decode("utf-8") |
| 1383 | else: |
| 1384 | fname_str = fname.decode("cp437") |
Georg Brandl | 5ba11de | 2011-01-01 10:09:32 +0000 | [diff] [blame] | 1385 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1386 | if fname_str != zinfo.orig_filename: |
| 1387 | raise BadZipFile( |
| 1388 | 'File name in directory %r and header %r differ.' |
| 1389 | % (zinfo.orig_filename, fname)) |
| 1390 | |
| 1391 | # check for encrypted flag & handle password |
| 1392 | is_encrypted = zinfo.flag_bits & 0x1 |
| 1393 | zd = None |
| 1394 | if is_encrypted: |
| 1395 | if not pwd: |
| 1396 | pwd = self.pwd |
| 1397 | if not pwd: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1398 | raise RuntimeError("File %r is encrypted, password " |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1399 | "required for extraction" % name) |
| 1400 | |
| 1401 | zd = _ZipDecrypter(pwd) |
| 1402 | # The first 12 bytes in the cypher stream is an encryption header |
| 1403 | # used to strengthen the algorithm. The first 11 bytes are |
| 1404 | # completely random, while the 12th contains the MSB of the CRC, |
| 1405 | # or the MSB of the file time depending on the header type |
| 1406 | # and is used to check the correctness of the password. |
| 1407 | header = zef_file.read(12) |
| 1408 | h = list(map(zd, header[0:12])) |
| 1409 | if zinfo.flag_bits & 0x8: |
| 1410 | # compare against the file type from extended local headers |
| 1411 | check_byte = (zinfo._raw_time >> 8) & 0xff |
| 1412 | else: |
| 1413 | # compare against the CRC otherwise |
| 1414 | check_byte = (zinfo.CRC >> 24) & 0xff |
| 1415 | if h[11] != check_byte: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1416 | raise RuntimeError("Bad password for file %r" % name) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1417 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1418 | return ZipExtFile(zef_file, mode, zinfo, zd, True) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1419 | except: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1420 | zef_file.close() |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1421 | raise |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1422 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1423 | def _open_to_write(self, zinfo, force_zip64=False): |
| 1424 | if force_zip64 and not self._allowZip64: |
| 1425 | raise ValueError( |
| 1426 | "force_zip64 is True, but allowZip64 was False when opening " |
| 1427 | "the ZIP file." |
| 1428 | ) |
| 1429 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1430 | raise ValueError("Can't write to the ZIP file while there is " |
| 1431 | "another write handle open on it. " |
| 1432 | "Close the first handle before opening another.") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1433 | |
| 1434 | # Sizes and CRC are overwritten with correct data after processing the file |
| 1435 | if not hasattr(zinfo, 'file_size'): |
| 1436 | zinfo.file_size = 0 |
| 1437 | zinfo.compress_size = 0 |
| 1438 | zinfo.CRC = 0 |
| 1439 | |
| 1440 | zinfo.flag_bits = 0x00 |
| 1441 | if zinfo.compress_type == ZIP_LZMA: |
| 1442 | # Compressed data includes an end-of-stream (EOS) marker |
| 1443 | zinfo.flag_bits |= 0x02 |
| 1444 | if not self._seekable: |
| 1445 | zinfo.flag_bits |= 0x08 |
| 1446 | |
| 1447 | if not zinfo.external_attr: |
| 1448 | zinfo.external_attr = 0o600 << 16 # permissions: ?rw------- |
| 1449 | |
| 1450 | # Compressed size can be larger than uncompressed size |
| 1451 | zip64 = self._allowZip64 and \ |
| 1452 | (force_zip64 or zinfo.file_size * 1.05 > ZIP64_LIMIT) |
| 1453 | |
| 1454 | if self._seekable: |
| 1455 | self.fp.seek(self.start_dir) |
| 1456 | zinfo.header_offset = self.fp.tell() |
| 1457 | |
| 1458 | self._writecheck(zinfo) |
| 1459 | self._didModify = True |
| 1460 | |
| 1461 | self.fp.write(zinfo.FileHeader(zip64)) |
| 1462 | |
| 1463 | self._writing = True |
| 1464 | return _ZipWriteFile(self, zinfo, zip64) |
| 1465 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1466 | def extract(self, member, path=None, pwd=None): |
| 1467 | """Extract a member from the archive to the current working directory, |
| 1468 | using its full name. Its file information is extracted as accurately |
| 1469 | as possible. `member' may be a filename or a ZipInfo object. You can |
| 1470 | specify a different directory using `path'. |
| 1471 | """ |
| 1472 | if not isinstance(member, ZipInfo): |
| 1473 | member = self.getinfo(member) |
| 1474 | |
| 1475 | if path is None: |
| 1476 | path = os.getcwd() |
| 1477 | |
| 1478 | return self._extract_member(member, path, pwd) |
| 1479 | |
| 1480 | def extractall(self, path=None, members=None, pwd=None): |
| 1481 | """Extract all members from the archive to the current working |
| 1482 | directory. `path' specifies a different directory to extract to. |
| 1483 | `members' is optional and must be a subset of the list returned |
| 1484 | by namelist(). |
| 1485 | """ |
| 1486 | if members is None: |
| 1487 | members = self.namelist() |
| 1488 | |
| 1489 | for zipinfo in members: |
| 1490 | self.extract(zipinfo, path, pwd) |
| 1491 | |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1492 | @classmethod |
| 1493 | def _sanitize_windows_name(cls, arcname, pathsep): |
| 1494 | """Replace bad characters and remove trailing dots from parts.""" |
| 1495 | table = cls._windows_illegal_name_trans_table |
| 1496 | if not table: |
| 1497 | illegal = ':<>|"?*' |
| 1498 | table = str.maketrans(illegal, '_' * len(illegal)) |
| 1499 | cls._windows_illegal_name_trans_table = table |
| 1500 | arcname = arcname.translate(table) |
| 1501 | # remove trailing dots |
| 1502 | arcname = (x.rstrip('.') for x in arcname.split(pathsep)) |
| 1503 | # rejoin, removing empty parts. |
| 1504 | arcname = pathsep.join(x for x in arcname if x) |
| 1505 | return arcname |
| 1506 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1507 | def _extract_member(self, member, targetpath, pwd): |
| 1508 | """Extract the ZipInfo object 'member' to a physical |
| 1509 | file on the path targetpath. |
| 1510 | """ |
| 1511 | # build the destination pathname, replacing |
| 1512 | # forward slashes to platform specific separators. |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1513 | arcname = member.filename.replace('/', os.path.sep) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1514 | |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1515 | if os.path.altsep: |
| 1516 | arcname = arcname.replace(os.path.altsep, os.path.sep) |
| 1517 | # interpret absolute pathname as relative, remove drive letter or |
| 1518 | # UNC path, redundant separators, "." and ".." components. |
| 1519 | arcname = os.path.splitdrive(arcname)[1] |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1520 | invalid_path_parts = ('', os.path.curdir, os.path.pardir) |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1521 | arcname = os.path.sep.join(x for x in arcname.split(os.path.sep) |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1522 | if x not in invalid_path_parts) |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1523 | if os.path.sep == '\\': |
Serhiy Storchaka | e5e6444 | 2013-02-02 19:50:59 +0200 | [diff] [blame] | 1524 | # filter illegal characters on Windows |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1525 | arcname = self._sanitize_windows_name(arcname, os.path.sep) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1526 | |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1527 | targetpath = os.path.join(targetpath, arcname) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1528 | targetpath = os.path.normpath(targetpath) |
| 1529 | |
| 1530 | # Create all upper directories if necessary. |
| 1531 | upperdirs = os.path.dirname(targetpath) |
| 1532 | if upperdirs and not os.path.exists(upperdirs): |
| 1533 | os.makedirs(upperdirs) |
| 1534 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 1535 | if member.is_dir(): |
Martin v. Löwis | 70ccd16 | 2009-05-24 19:47:22 +0000 | [diff] [blame] | 1536 | if not os.path.isdir(targetpath): |
| 1537 | os.mkdir(targetpath) |
Martin v. Löwis | 59e4779 | 2009-01-24 14:10:07 +0000 | [diff] [blame] | 1538 | return targetpath |
| 1539 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1540 | with self.open(member, pwd=pwd) as source, \ |
| 1541 | open(targetpath, "wb") as target: |
| 1542 | shutil.copyfileobj(source, target) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1543 | |
| 1544 | return targetpath |
| 1545 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1546 | def _writecheck(self, zinfo): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1547 | """Check for errors before writing a file to the archive.""" |
Raymond Hettinger | 54f0222 | 2002-06-01 14:18:47 +0000 | [diff] [blame] | 1548 | if zinfo.filename in self.NameToInfo: |
Serhiy Storchaka | 9b7a1a1 | 2014-01-20 21:57:40 +0200 | [diff] [blame] | 1549 | import warnings |
| 1550 | warnings.warn('Duplicate name: %r' % zinfo.filename, stacklevel=3) |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1551 | if self.mode not in ('w', 'x', 'a'): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1552 | raise ValueError("write() requires mode 'w', 'x', or 'a'") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1553 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1554 | raise ValueError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1555 | "Attempt to write ZIP archive that was already closed") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 1556 | _check_compression(zinfo.compress_type) |
Serhiy Storchaka | cfbb394 | 2014-09-23 21:34:24 +0300 | [diff] [blame] | 1557 | if not self._allowZip64: |
| 1558 | requires_zip64 = None |
| 1559 | if len(self.filelist) >= ZIP_FILECOUNT_LIMIT: |
| 1560 | requires_zip64 = "Files count" |
| 1561 | elif zinfo.file_size > ZIP64_LIMIT: |
| 1562 | requires_zip64 = "Filesize" |
| 1563 | elif zinfo.header_offset > ZIP64_LIMIT: |
| 1564 | requires_zip64 = "Zipfile size" |
| 1565 | if requires_zip64: |
| 1566 | raise LargeZipFile(requires_zip64 + |
| 1567 | " would require ZIP64 extensions") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1568 | |
| 1569 | def write(self, filename, arcname=None, compress_type=None): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1570 | """Put the bytes from filename into the archive under the name |
| 1571 | arcname.""" |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1572 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1573 | raise ValueError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1574 | "Attempt to write to ZIP archive that was already closed") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1575 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1576 | raise ValueError( |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1577 | "Can't write to ZIP archive while an open writing handle exists" |
| 1578 | ) |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1579 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 1580 | zinfo = ZipInfo.from_file(filename, arcname) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1581 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 1582 | if zinfo.is_dir(): |
| 1583 | zinfo.compress_size = 0 |
| 1584 | zinfo.CRC = 0 |
| 1585 | else: |
| 1586 | if compress_type is not None: |
| 1587 | zinfo.compress_type = compress_type |
| 1588 | else: |
| 1589 | zinfo.compress_type = self.compression |
| 1590 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1591 | if zinfo.is_dir(): |
| 1592 | with self._lock: |
| 1593 | if self._seekable: |
| 1594 | self.fp.seek(self.start_dir) |
| 1595 | zinfo.header_offset = self.fp.tell() # Start of header bytes |
| 1596 | if zinfo.compress_type == ZIP_LZMA: |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1597 | # Compressed data includes an end-of-stream (EOS) marker |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1598 | zinfo.flag_bits |= 0x02 |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1599 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1600 | self._writecheck(zinfo) |
| 1601 | self._didModify = True |
Martin v. Löwis | 59e4779 | 2009-01-24 14:10:07 +0000 | [diff] [blame] | 1602 | |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1603 | self.filelist.append(zinfo) |
| 1604 | self.NameToInfo[zinfo.filename] = zinfo |
| 1605 | self.fp.write(zinfo.FileHeader(False)) |
| 1606 | self.start_dir = self.fp.tell() |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1607 | else: |
| 1608 | with open(filename, "rb") as src, self.open(zinfo, 'w') as dest: |
| 1609 | shutil.copyfileobj(src, dest, 1024*8) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1610 | |
Ronald Oussoren | ee5c885 | 2010-02-07 20:24:02 +0000 | [diff] [blame] | 1611 | def writestr(self, zinfo_or_arcname, data, compress_type=None): |
Guido van Rossum | 85825dc | 2007-08-27 17:03:28 +0000 | [diff] [blame] | 1612 | """Write a file into the archive. The contents is 'data', which |
| 1613 | may be either a 'str' or a 'bytes' instance; if it is a 'str', |
| 1614 | it is encoded as UTF-8 first. |
| 1615 | 'zinfo_or_arcname' is either a ZipInfo instance or |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1616 | the name of the file in the archive.""" |
Guido van Rossum | 85825dc | 2007-08-27 17:03:28 +0000 | [diff] [blame] | 1617 | if isinstance(data, str): |
| 1618 | data = data.encode("utf-8") |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1619 | if not isinstance(zinfo_or_arcname, ZipInfo): |
| 1620 | zinfo = ZipInfo(filename=zinfo_or_arcname, |
Guido van Rossum | 7736b5b | 2008-01-15 21:44:53 +0000 | [diff] [blame] | 1621 | date_time=time.localtime(time.time())[:6]) |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1622 | zinfo.compress_type = self.compression |
Serhiy Storchaka | 46a3492 | 2014-09-23 22:40:23 +0300 | [diff] [blame] | 1623 | if zinfo.filename[-1] == '/': |
| 1624 | zinfo.external_attr = 0o40775 << 16 # drwxrwxr-x |
| 1625 | zinfo.external_attr |= 0x10 # MS-DOS directory flag |
| 1626 | else: |
| 1627 | zinfo.external_attr = 0o600 << 16 # ?rw------- |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1628 | else: |
| 1629 | zinfo = zinfo_or_arcname |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1630 | |
| 1631 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1632 | raise ValueError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1633 | "Attempt to write to ZIP archive that was already closed") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1634 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1635 | raise ValueError( |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1636 | "Can't write to ZIP archive while an open writing handle exists." |
| 1637 | ) |
| 1638 | |
| 1639 | if compress_type is not None: |
| 1640 | zinfo.compress_type = compress_type |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1641 | |
Guido van Rossum | 85825dc | 2007-08-27 17:03:28 +0000 | [diff] [blame] | 1642 | zinfo.file_size = len(data) # Uncompressed size |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1643 | with self._lock: |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1644 | with self.open(zinfo, mode='w') as dest: |
| 1645 | dest.write(data) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1646 | |
| 1647 | def __del__(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1648 | """Call the "close()" method in case the user forgot.""" |
Tim Peters | d15f8bb | 2001-11-28 23:16:40 +0000 | [diff] [blame] | 1649 | self.close() |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1650 | |
| 1651 | def close(self): |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1652 | """Close the file, and for mode 'w', 'x' and 'a' write the ending |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1653 | records.""" |
Tim Peters | d15f8bb | 2001-11-28 23:16:40 +0000 | [diff] [blame] | 1654 | if self.fp is None: |
| 1655 | return |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1656 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1657 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1658 | raise ValueError("Can't close the ZIP file while there is " |
| 1659 | "an open writing handle on it. " |
| 1660 | "Close the writing handle before closing the zip.") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1661 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1662 | try: |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1663 | if self.mode in ('w', 'x', 'a') and self._didModify: # write ending records |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1664 | with self._lock: |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1665 | if self._seekable: |
Serhiy Storchaka | a14f7d2 | 2015-01-26 14:01:27 +0200 | [diff] [blame] | 1666 | self.fp.seek(self.start_dir) |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1667 | self._write_end_record() |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1668 | finally: |
| 1669 | fp = self.fp |
| 1670 | self.fp = None |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1671 | self._fpclose(fp) |
| 1672 | |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1673 | def _write_end_record(self): |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1674 | for zinfo in self.filelist: # write central directory |
| 1675 | dt = zinfo.date_time |
| 1676 | dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2] |
| 1677 | dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2) |
| 1678 | extra = [] |
| 1679 | if zinfo.file_size > ZIP64_LIMIT \ |
| 1680 | or zinfo.compress_size > ZIP64_LIMIT: |
| 1681 | extra.append(zinfo.file_size) |
| 1682 | extra.append(zinfo.compress_size) |
| 1683 | file_size = 0xffffffff |
| 1684 | compress_size = 0xffffffff |
| 1685 | else: |
| 1686 | file_size = zinfo.file_size |
| 1687 | compress_size = zinfo.compress_size |
| 1688 | |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1689 | header_offset = zinfo.header_offset - self._start_disk |
| 1690 | if header_offset > ZIP64_LIMIT: |
| 1691 | extra.append(header_offset) |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1692 | header_offset = 0xffffffff |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1693 | |
| 1694 | extra_data = zinfo.extra |
| 1695 | min_version = 0 |
| 1696 | if extra: |
| 1697 | # Append a ZIP64 field to the extra's |
| 1698 | extra_data = struct.pack( |
| 1699 | '<HH' + 'Q'*len(extra), |
| 1700 | 1, 8*len(extra), *extra) + extra_data |
| 1701 | |
| 1702 | min_version = ZIP64_VERSION |
| 1703 | |
| 1704 | if zinfo.compress_type == ZIP_BZIP2: |
| 1705 | min_version = max(BZIP2_VERSION, min_version) |
| 1706 | elif zinfo.compress_type == ZIP_LZMA: |
| 1707 | min_version = max(LZMA_VERSION, min_version) |
| 1708 | |
| 1709 | extract_version = max(min_version, zinfo.extract_version) |
| 1710 | create_version = max(min_version, zinfo.create_version) |
| 1711 | try: |
| 1712 | filename, flag_bits = zinfo._encodeFilenameFlags() |
| 1713 | centdir = struct.pack(structCentralDir, |
| 1714 | stringCentralDir, create_version, |
| 1715 | zinfo.create_system, extract_version, zinfo.reserved, |
| 1716 | flag_bits, zinfo.compress_type, dostime, dosdate, |
| 1717 | zinfo.CRC, compress_size, file_size, |
| 1718 | len(filename), len(extra_data), len(zinfo.comment), |
| 1719 | 0, zinfo.internal_attr, zinfo.external_attr, |
| 1720 | header_offset) |
| 1721 | except DeprecationWarning: |
| 1722 | print((structCentralDir, stringCentralDir, create_version, |
| 1723 | zinfo.create_system, extract_version, zinfo.reserved, |
| 1724 | zinfo.flag_bits, zinfo.compress_type, dostime, dosdate, |
| 1725 | zinfo.CRC, compress_size, file_size, |
| 1726 | len(zinfo.filename), len(extra_data), len(zinfo.comment), |
| 1727 | 0, zinfo.internal_attr, zinfo.external_attr, |
| 1728 | header_offset), file=sys.stderr) |
| 1729 | raise |
| 1730 | self.fp.write(centdir) |
| 1731 | self.fp.write(filename) |
| 1732 | self.fp.write(extra_data) |
| 1733 | self.fp.write(zinfo.comment) |
| 1734 | |
| 1735 | pos2 = self.fp.tell() |
| 1736 | # Write end-of-zip-archive record |
| 1737 | centDirCount = len(self.filelist) |
| 1738 | centDirSize = pos2 - self.start_dir |
Serhiy Storchaka | 8793b21 | 2016-10-07 22:20:50 +0300 | [diff] [blame] | 1739 | centDirOffset = self.start_dir - self._start_disk |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1740 | requires_zip64 = None |
| 1741 | if centDirCount > ZIP_FILECOUNT_LIMIT: |
| 1742 | requires_zip64 = "Files count" |
| 1743 | elif centDirOffset > ZIP64_LIMIT: |
| 1744 | requires_zip64 = "Central directory offset" |
| 1745 | elif centDirSize > ZIP64_LIMIT: |
| 1746 | requires_zip64 = "Central directory size" |
| 1747 | if requires_zip64: |
| 1748 | # Need to write the ZIP64 end-of-archive records |
| 1749 | if not self._allowZip64: |
| 1750 | raise LargeZipFile(requires_zip64 + |
| 1751 | " would require ZIP64 extensions") |
| 1752 | zip64endrec = struct.pack( |
| 1753 | structEndArchive64, stringEndArchive64, |
| 1754 | 44, 45, 45, 0, 0, centDirCount, centDirCount, |
| 1755 | centDirSize, centDirOffset) |
| 1756 | self.fp.write(zip64endrec) |
| 1757 | |
| 1758 | zip64locrec = struct.pack( |
| 1759 | structEndArchive64Locator, |
| 1760 | stringEndArchive64Locator, 0, pos2, 1) |
| 1761 | self.fp.write(zip64locrec) |
| 1762 | centDirCount = min(centDirCount, 0xFFFF) |
| 1763 | centDirSize = min(centDirSize, 0xFFFFFFFF) |
| 1764 | centDirOffset = min(centDirOffset, 0xFFFFFFFF) |
| 1765 | |
| 1766 | endrec = struct.pack(structEndArchive, stringEndArchive, |
| 1767 | 0, 0, centDirCount, centDirCount, |
| 1768 | centDirSize, centDirOffset, len(self._comment)) |
| 1769 | self.fp.write(endrec) |
| 1770 | self.fp.write(self._comment) |
| 1771 | self.fp.flush() |
| 1772 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1773 | def _fpclose(self, fp): |
| 1774 | assert self._fileRefCnt > 0 |
| 1775 | self._fileRefCnt -= 1 |
| 1776 | if not self._fileRefCnt and not self._filePassed: |
| 1777 | fp.close() |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1778 | |
| 1779 | |
| 1780 | class PyZipFile(ZipFile): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1781 | """Class to create ZIP archives with Python library files and packages.""" |
| 1782 | |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1783 | def __init__(self, file, mode="r", compression=ZIP_STORED, |
Serhiy Storchaka | 235c5e0 | 2013-11-23 15:55:38 +0200 | [diff] [blame] | 1784 | allowZip64=True, optimize=-1): |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1785 | ZipFile.__init__(self, file, mode=mode, compression=compression, |
| 1786 | allowZip64=allowZip64) |
| 1787 | self._optimize = optimize |
| 1788 | |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1789 | def writepy(self, pathname, basename="", filterfunc=None): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1790 | """Add all files from "pathname" to the ZIP archive. |
| 1791 | |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1792 | If pathname is a package directory, search the directory and |
| 1793 | all package subdirectories recursively for all *.py and enter |
| 1794 | the modules into the archive. If pathname is a plain |
| 1795 | directory, listdir *.py and enter all modules. Else, pathname |
| 1796 | must be a Python *.py file and the module will be put into the |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1797 | archive. Added modules are always module.pyc. |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1798 | This method will compile the module.py into module.pyc if |
| 1799 | necessary. |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1800 | If filterfunc(pathname) is given, it is called with every argument. |
| 1801 | When it is False, the file or directory is skipped. |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1802 | """ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1803 | if filterfunc and not filterfunc(pathname): |
| 1804 | if self.debug: |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1805 | label = 'path' if os.path.isdir(pathname) else 'file' |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1806 | print('%s %r skipped by filterfunc' % (label, pathname)) |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1807 | return |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1808 | dir, name = os.path.split(pathname) |
| 1809 | if os.path.isdir(pathname): |
| 1810 | initname = os.path.join(pathname, "__init__.py") |
| 1811 | if os.path.isfile(initname): |
| 1812 | # This is a package directory, add it |
| 1813 | if basename: |
| 1814 | basename = "%s/%s" % (basename, name) |
| 1815 | else: |
| 1816 | basename = name |
| 1817 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1818 | print("Adding package in", pathname, "as", basename) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1819 | fname, arcname = self._get_codename(initname[0:-3], basename) |
| 1820 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1821 | print("Adding", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1822 | self.write(fname, arcname) |
| 1823 | dirlist = os.listdir(pathname) |
| 1824 | dirlist.remove("__init__.py") |
| 1825 | # Add all *.py files and package subdirectories |
| 1826 | for filename in dirlist: |
| 1827 | path = os.path.join(pathname, filename) |
| 1828 | root, ext = os.path.splitext(filename) |
| 1829 | if os.path.isdir(path): |
| 1830 | if os.path.isfile(os.path.join(path, "__init__.py")): |
| 1831 | # This is a package directory, add it |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1832 | self.writepy(path, basename, |
| 1833 | filterfunc=filterfunc) # Recursive call |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1834 | elif ext == ".py": |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1835 | if filterfunc and not filterfunc(path): |
| 1836 | if self.debug: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1837 | print('file %r skipped by filterfunc' % path) |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1838 | continue |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1839 | fname, arcname = self._get_codename(path[0:-3], |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1840 | basename) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1841 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1842 | print("Adding", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1843 | self.write(fname, arcname) |
| 1844 | else: |
| 1845 | # This is NOT a package directory, add its files at top level |
| 1846 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1847 | print("Adding files from directory", pathname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1848 | for filename in os.listdir(pathname): |
| 1849 | path = os.path.join(pathname, filename) |
| 1850 | root, ext = os.path.splitext(filename) |
| 1851 | if ext == ".py": |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1852 | if filterfunc and not filterfunc(path): |
| 1853 | if self.debug: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1854 | print('file %r skipped by filterfunc' % path) |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1855 | continue |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1856 | fname, arcname = self._get_codename(path[0:-3], |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1857 | basename) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1858 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1859 | print("Adding", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1860 | self.write(fname, arcname) |
| 1861 | else: |
| 1862 | if pathname[-3:] != ".py": |
Collin Winter | ce36ad8 | 2007-08-30 01:19:48 +0000 | [diff] [blame] | 1863 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1864 | 'Files added with writepy() must end with ".py"') |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1865 | fname, arcname = self._get_codename(pathname[0:-3], basename) |
| 1866 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1867 | print("Adding file", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1868 | self.write(fname, arcname) |
| 1869 | |
| 1870 | def _get_codename(self, pathname, basename): |
| 1871 | """Return (filename, archivename) for the path. |
| 1872 | |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1873 | Given a module name path, return the correct file path and |
| 1874 | archive name, compiling if necessary. For example, given |
| 1875 | /python/lib/string, return (/python/lib/string.pyc, string). |
| 1876 | """ |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1877 | def _compile(file, optimize=-1): |
| 1878 | import py_compile |
| 1879 | if self.debug: |
| 1880 | print("Compiling", file) |
| 1881 | try: |
| 1882 | py_compile.compile(file, doraise=True, optimize=optimize) |
Serhiy Storchaka | 45c4375 | 2013-01-29 20:10:28 +0200 | [diff] [blame] | 1883 | except py_compile.PyCompileError as err: |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1884 | print(err.msg) |
| 1885 | return False |
| 1886 | return True |
| 1887 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1888 | file_py = pathname + ".py" |
| 1889 | file_pyc = pathname + ".pyc" |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1890 | pycache_opt0 = importlib.util.cache_from_source(file_py, optimization='') |
| 1891 | pycache_opt1 = importlib.util.cache_from_source(file_py, optimization=1) |
| 1892 | pycache_opt2 = importlib.util.cache_from_source(file_py, optimization=2) |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1893 | if self._optimize == -1: |
| 1894 | # legacy mode: use whatever file is present |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1895 | if (os.path.isfile(file_pyc) and |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1896 | os.stat(file_pyc).st_mtime >= os.stat(file_py).st_mtime): |
| 1897 | # Use .pyc file. |
| 1898 | arcname = fname = file_pyc |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1899 | elif (os.path.isfile(pycache_opt0) and |
| 1900 | os.stat(pycache_opt0).st_mtime >= os.stat(file_py).st_mtime): |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1901 | # Use the __pycache__/*.pyc file, but write it to the legacy pyc |
| 1902 | # file name in the archive. |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1903 | fname = pycache_opt0 |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1904 | arcname = file_pyc |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1905 | elif (os.path.isfile(pycache_opt1) and |
| 1906 | os.stat(pycache_opt1).st_mtime >= os.stat(file_py).st_mtime): |
| 1907 | # Use the __pycache__/*.pyc file, but write it to the legacy pyc |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1908 | # file name in the archive. |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1909 | fname = pycache_opt1 |
| 1910 | arcname = file_pyc |
| 1911 | elif (os.path.isfile(pycache_opt2) and |
| 1912 | os.stat(pycache_opt2).st_mtime >= os.stat(file_py).st_mtime): |
| 1913 | # Use the __pycache__/*.pyc file, but write it to the legacy pyc |
| 1914 | # file name in the archive. |
| 1915 | fname = pycache_opt2 |
| 1916 | arcname = file_pyc |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 1917 | else: |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1918 | # Compile py into PEP 3147 pyc file. |
| 1919 | if _compile(file_py): |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1920 | if sys.flags.optimize == 0: |
| 1921 | fname = pycache_opt0 |
| 1922 | elif sys.flags.optimize == 1: |
| 1923 | fname = pycache_opt1 |
| 1924 | else: |
| 1925 | fname = pycache_opt2 |
| 1926 | arcname = file_pyc |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1927 | else: |
| 1928 | fname = arcname = file_py |
| 1929 | else: |
| 1930 | # new mode: use given optimization level |
| 1931 | if self._optimize == 0: |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1932 | fname = pycache_opt0 |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1933 | arcname = file_pyc |
| 1934 | else: |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 1935 | arcname = file_pyc |
| 1936 | if self._optimize == 1: |
| 1937 | fname = pycache_opt1 |
| 1938 | elif self._optimize == 2: |
| 1939 | fname = pycache_opt2 |
| 1940 | else: |
| 1941 | msg = "invalid value for 'optimize': {!r}".format(self._optimize) |
| 1942 | raise ValueError(msg) |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1943 | if not (os.path.isfile(fname) and |
| 1944 | os.stat(fname).st_mtime >= os.stat(file_py).st_mtime): |
| 1945 | if not _compile(file_py, optimize=self._optimize): |
| 1946 | fname = arcname = file_py |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 1947 | archivename = os.path.split(arcname)[1] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1948 | if basename: |
| 1949 | archivename = "%s/%s" % (basename, archivename) |
| 1950 | return (fname, archivename) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1951 | |
| 1952 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 1953 | def main(args=None): |
| 1954 | import argparse |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1955 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 1956 | description = 'A simple command line interface for zipfile module.' |
| 1957 | parser = argparse.ArgumentParser(description=description) |
| 1958 | group = parser.add_mutually_exclusive_group() |
| 1959 | group.add_argument('-l', '--list', metavar='<zipfile>', |
| 1960 | help='Show listing of a zipfile') |
| 1961 | group.add_argument('-e', '--extract', nargs=2, |
| 1962 | metavar=('<zipfile>', '<output_dir>'), |
| 1963 | help='Extract zipfile into target dir') |
| 1964 | group.add_argument('-c', '--create', nargs='+', |
| 1965 | metavar=('<name>', '<file>'), |
| 1966 | help='Create zipfile from sources') |
| 1967 | group.add_argument('-t', '--test', metavar='<zipfile>', |
| 1968 | help='Test if a zipfile is valid') |
| 1969 | args = parser.parse_args(args) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1970 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 1971 | if args.test is not None: |
| 1972 | src = args.test |
| 1973 | with ZipFile(src, 'r') as zf: |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1974 | badfile = zf.testzip() |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 1975 | if badfile: |
| 1976 | print("The following enclosed file is corrupted: {!r}".format(badfile)) |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1977 | print("Done testing") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1978 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 1979 | elif args.list is not None: |
| 1980 | src = args.list |
| 1981 | with ZipFile(src, 'r') as zf: |
| 1982 | zf.printdir() |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1983 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 1984 | elif args.extract is not None: |
| 1985 | src, curdir = args.extract |
| 1986 | with ZipFile(src, 'r') as zf: |
| 1987 | zf.extractall(curdir) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1988 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 1989 | elif args.create is not None: |
| 1990 | zip_name = args.create.pop(0) |
| 1991 | files = args.create |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1992 | |
| 1993 | def addToZip(zf, path, zippath): |
| 1994 | if os.path.isfile(path): |
| 1995 | zf.write(path, zippath, ZIP_DEFLATED) |
| 1996 | elif os.path.isdir(path): |
Serhiy Storchaka | 518e71b | 2014-10-04 13:39:34 +0300 | [diff] [blame] | 1997 | if zippath: |
| 1998 | zf.write(path, zippath) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1999 | for nm in os.listdir(path): |
| 2000 | addToZip(zf, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 2001 | os.path.join(path, nm), os.path.join(zippath, nm)) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2002 | # else: ignore |
| 2003 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2004 | with ZipFile(zip_name, 'w') as zf: |
| 2005 | for path in files: |
Serhiy Storchaka | 518e71b | 2014-10-04 13:39:34 +0300 | [diff] [blame] | 2006 | zippath = os.path.basename(path) |
| 2007 | if not zippath: |
| 2008 | zippath = os.path.basename(os.path.dirname(path)) |
| 2009 | if zippath in ('', os.curdir, os.pardir): |
| 2010 | zippath = '' |
| 2011 | addToZip(zf, path, zippath) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2012 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2013 | else: |
| 2014 | parser.exit(2, parser.format_usage()) |
| 2015 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2016 | if __name__ == "__main__": |
| 2017 | main() |