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 | """ |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 6 | import binascii |
| 7 | import functools |
| 8 | import importlib.util |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 9 | import io |
Miss Islington (bot) | c410f38 | 2019-08-24 09:03:52 -0700 | [diff] [blame^] | 10 | import itertools |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 11 | import os |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 12 | import posixpath |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 13 | import shutil |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 14 | import stat |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 15 | import struct |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 16 | import sys |
Antoine Pitrou | a6a4dc8 | 2017-09-07 18:56:24 +0200 | [diff] [blame] | 17 | import threading |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 18 | import time |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 19 | |
| 20 | try: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 21 | import zlib # We may need its compression method |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 22 | crc32 = zlib.crc32 |
Brett Cannon | 260fbe8 | 2013-07-04 18:16:15 -0400 | [diff] [blame] | 23 | except ImportError: |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 24 | zlib = None |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 25 | crc32 = binascii.crc32 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 26 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 27 | try: |
| 28 | import bz2 # We may need its compression method |
Brett Cannon | 260fbe8 | 2013-07-04 18:16:15 -0400 | [diff] [blame] | 29 | except ImportError: |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 30 | bz2 = None |
| 31 | |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 32 | try: |
| 33 | import lzma # We may need its compression method |
Brett Cannon | 260fbe8 | 2013-07-04 18:16:15 -0400 | [diff] [blame] | 34 | except ImportError: |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 35 | lzma = None |
| 36 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 37 | __all__ = ["BadZipFile", "BadZipfile", "error", |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 38 | "ZIP_STORED", "ZIP_DEFLATED", "ZIP_BZIP2", "ZIP_LZMA", |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 39 | "is_zipfile", "ZipInfo", "ZipFile", "PyZipFile", "LargeZipFile"] |
Skip Montanaro | 40fc160 | 2001-03-01 04:27:19 +0000 | [diff] [blame] | 40 | |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 41 | class BadZipFile(Exception): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 42 | pass |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 43 | |
| 44 | |
| 45 | class LargeZipFile(Exception): |
| 46 | """ |
| 47 | Raised when writing a zipfile, the zipfile requires ZIP64 extensions |
| 48 | and those extensions are disabled. |
| 49 | """ |
| 50 | |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 51 | error = BadZipfile = BadZipFile # Pre-3.2 compatibility names |
| 52 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 53 | |
Amaury Forgeot d'Arc | 0c3f8a4 | 2009-01-17 16:42:26 +0000 | [diff] [blame] | 54 | ZIP64_LIMIT = (1 << 31) - 1 |
Serhiy Storchaka | cfbb394 | 2014-09-23 21:34:24 +0300 | [diff] [blame] | 55 | ZIP_FILECOUNT_LIMIT = (1 << 16) - 1 |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 56 | ZIP_MAX_COMMENT = (1 << 16) - 1 |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 57 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 58 | # constants for Zip file compression methods |
| 59 | ZIP_STORED = 0 |
| 60 | ZIP_DEFLATED = 8 |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 61 | ZIP_BZIP2 = 12 |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 62 | ZIP_LZMA = 14 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 63 | # Other ZIP compression methods not supported |
| 64 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 65 | DEFAULT_VERSION = 20 |
| 66 | ZIP64_VERSION = 45 |
| 67 | BZIP2_VERSION = 46 |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 68 | LZMA_VERSION = 63 |
Martin v. Löwis | d099b56 | 2012-05-01 14:08:22 +0200 | [diff] [blame] | 69 | # 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] | 70 | MAX_EXTRACT_VERSION = 63 |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 71 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 72 | # Below are some formats and associated data for reading/writing headers using |
| 73 | # the struct module. The names and structures of headers/records are those used |
| 74 | # in the PKWARE description of the ZIP file format: |
| 75 | # http://www.pkware.com/documents/casestudies/APPNOTE.TXT |
| 76 | # (URL valid as of January 2008) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 77 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 78 | # The "end of central directory" structure, magic number, size, and indices |
| 79 | # (section V.I in the format document) |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 80 | structEndArchive = b"<4s4H2LH" |
| 81 | stringEndArchive = b"PK\005\006" |
| 82 | sizeEndCentDir = struct.calcsize(structEndArchive) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 83 | |
| 84 | _ECD_SIGNATURE = 0 |
| 85 | _ECD_DISK_NUMBER = 1 |
| 86 | _ECD_DISK_START = 2 |
| 87 | _ECD_ENTRIES_THIS_DISK = 3 |
| 88 | _ECD_ENTRIES_TOTAL = 4 |
| 89 | _ECD_SIZE = 5 |
| 90 | _ECD_OFFSET = 6 |
| 91 | _ECD_COMMENT_SIZE = 7 |
| 92 | # These last two indices are not part of the structure as defined in the |
| 93 | # spec, but they are used internally by this module as a convenience |
| 94 | _ECD_COMMENT = 8 |
| 95 | _ECD_LOCATION = 9 |
| 96 | |
| 97 | # The "central directory" structure, magic number, size, and indices |
| 98 | # of entries in the structure (section V.F in the format document) |
| 99 | structCentralDir = "<4s4B4HL2L5H2L" |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 100 | stringCentralDir = b"PK\001\002" |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 101 | sizeCentralDir = struct.calcsize(structCentralDir) |
| 102 | |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 103 | # indexes of entries in the central directory structure |
| 104 | _CD_SIGNATURE = 0 |
| 105 | _CD_CREATE_VERSION = 1 |
| 106 | _CD_CREATE_SYSTEM = 2 |
| 107 | _CD_EXTRACT_VERSION = 3 |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 108 | _CD_EXTRACT_SYSTEM = 4 |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 109 | _CD_FLAG_BITS = 5 |
| 110 | _CD_COMPRESS_TYPE = 6 |
| 111 | _CD_TIME = 7 |
| 112 | _CD_DATE = 8 |
| 113 | _CD_CRC = 9 |
| 114 | _CD_COMPRESSED_SIZE = 10 |
| 115 | _CD_UNCOMPRESSED_SIZE = 11 |
| 116 | _CD_FILENAME_LENGTH = 12 |
| 117 | _CD_EXTRA_FIELD_LENGTH = 13 |
| 118 | _CD_COMMENT_LENGTH = 14 |
| 119 | _CD_DISK_NUMBER_START = 15 |
| 120 | _CD_INTERNAL_FILE_ATTRIBUTES = 16 |
| 121 | _CD_EXTERNAL_FILE_ATTRIBUTES = 17 |
| 122 | _CD_LOCAL_HEADER_OFFSET = 18 |
| 123 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 124 | # The "local file header" structure, magic number, size, and indices |
| 125 | # (section V.A in the format document) |
| 126 | structFileHeader = "<4s2B4HL2L2H" |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 127 | stringFileHeader = b"PK\003\004" |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 128 | sizeFileHeader = struct.calcsize(structFileHeader) |
| 129 | |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 130 | _FH_SIGNATURE = 0 |
| 131 | _FH_EXTRACT_VERSION = 1 |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 132 | _FH_EXTRACT_SYSTEM = 2 |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 133 | _FH_GENERAL_PURPOSE_FLAG_BITS = 3 |
| 134 | _FH_COMPRESSION_METHOD = 4 |
| 135 | _FH_LAST_MOD_TIME = 5 |
| 136 | _FH_LAST_MOD_DATE = 6 |
| 137 | _FH_CRC = 7 |
| 138 | _FH_COMPRESSED_SIZE = 8 |
| 139 | _FH_UNCOMPRESSED_SIZE = 9 |
| 140 | _FH_FILENAME_LENGTH = 10 |
| 141 | _FH_EXTRA_FIELD_LENGTH = 11 |
| 142 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 143 | # The "Zip64 end of central directory locator" structure, magic number, and size |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 144 | structEndArchive64Locator = "<4sLQL" |
| 145 | stringEndArchive64Locator = b"PK\x06\x07" |
| 146 | sizeEndCentDir64Locator = struct.calcsize(structEndArchive64Locator) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 147 | |
| 148 | # The "Zip64 end of central directory" record, magic number, size, and indices |
| 149 | # (section V.G in the format document) |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 150 | structEndArchive64 = "<4sQ2H2L4Q" |
| 151 | stringEndArchive64 = b"PK\x06\x06" |
| 152 | sizeEndCentDir64 = struct.calcsize(structEndArchive64) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 153 | |
| 154 | _CD64_SIGNATURE = 0 |
| 155 | _CD64_DIRECTORY_RECSIZE = 1 |
| 156 | _CD64_CREATE_VERSION = 2 |
| 157 | _CD64_EXTRACT_VERSION = 3 |
| 158 | _CD64_DISK_NUMBER = 4 |
| 159 | _CD64_DISK_NUMBER_START = 5 |
| 160 | _CD64_NUMBER_ENTRIES_THIS_DISK = 6 |
| 161 | _CD64_NUMBER_ENTRIES_TOTAL = 7 |
| 162 | _CD64_DIRECTORY_SIZE = 8 |
| 163 | _CD64_OFFSET_START_CENTDIR = 9 |
| 164 | |
Silas Sewell | 4ba3b50 | 2018-09-18 13:00:05 -0400 | [diff] [blame] | 165 | _DD_SIGNATURE = 0x08074b50 |
| 166 | |
Serhiy Storchaka | 9bdb7be | 2018-09-17 15:36:40 +0300 | [diff] [blame] | 167 | _EXTRA_FIELD_STRUCT = struct.Struct('<HH') |
| 168 | |
| 169 | def _strip_extra(extra, xids): |
| 170 | # Remove Extra Fields with specified IDs. |
| 171 | unpack = _EXTRA_FIELD_STRUCT.unpack |
| 172 | modified = False |
| 173 | buffer = [] |
| 174 | start = i = 0 |
| 175 | while i + 4 <= len(extra): |
| 176 | xid, xlen = unpack(extra[i : i + 4]) |
| 177 | j = i + 4 + xlen |
| 178 | if xid in xids: |
| 179 | if i != start: |
| 180 | buffer.append(extra[start : i]) |
| 181 | start = j |
| 182 | modified = True |
| 183 | i = j |
| 184 | if not modified: |
| 185 | return extra |
| 186 | return b''.join(buffer) |
| 187 | |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 188 | def _check_zipfile(fp): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 189 | try: |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 190 | if _EndRecData(fp): |
| 191 | return True # file has correct magic number |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 192 | except OSError: |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 193 | pass |
Guido van Rossum | 8ca162f | 2002-04-07 06:36:23 +0000 | [diff] [blame] | 194 | return False |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 195 | |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 196 | def is_zipfile(filename): |
| 197 | """Quickly see if a file is a ZIP file by checking the magic number. |
| 198 | |
| 199 | The filename argument may be a file or file-like object too. |
| 200 | """ |
| 201 | result = False |
| 202 | try: |
| 203 | if hasattr(filename, "read"): |
| 204 | result = _check_zipfile(fp=filename) |
| 205 | else: |
| 206 | with open(filename, "rb") as fp: |
| 207 | result = _check_zipfile(fp) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 208 | except OSError: |
Antoine Pitrou | db5fe66 | 2008-12-27 15:50:40 +0000 | [diff] [blame] | 209 | pass |
| 210 | return result |
| 211 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 212 | def _EndRecData64(fpin, offset, endrec): |
| 213 | """ |
| 214 | Read the ZIP64 end-of-archive records and use that to update endrec |
| 215 | """ |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 216 | try: |
| 217 | fpin.seek(offset - sizeEndCentDir64Locator, 2) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 218 | except OSError: |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 219 | # If the seek fails, the file is not large enough to contain a ZIP64 |
| 220 | # end-of-archive record, so just return the end record we were given. |
| 221 | return endrec |
| 222 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 223 | data = fpin.read(sizeEndCentDir64Locator) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 224 | if len(data) != sizeEndCentDir64Locator: |
| 225 | return endrec |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 226 | sig, diskno, reloff, disks = struct.unpack(structEndArchive64Locator, data) |
| 227 | if sig != stringEndArchive64Locator: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 228 | return endrec |
| 229 | |
Francisco Facioni | ab0716e | 2019-05-29 00:15:11 +0100 | [diff] [blame] | 230 | if diskno != 0 or disks > 1: |
Éric Araujo | ae2d832 | 2010-10-28 13:49:17 +0000 | [diff] [blame] | 231 | raise BadZipFile("zipfiles that span multiple disks are not supported") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 232 | |
| 233 | # Assume no 'zip64 extensible data' |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 234 | fpin.seek(offset - sizeEndCentDir64Locator - sizeEndCentDir64, 2) |
| 235 | data = fpin.read(sizeEndCentDir64) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 236 | if len(data) != sizeEndCentDir64: |
| 237 | return endrec |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 238 | sig, sz, create_version, read_version, disk_num, disk_dir, \ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 239 | dircount, dircount2, dirsize, diroffset = \ |
| 240 | struct.unpack(structEndArchive64, data) |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 241 | if sig != stringEndArchive64: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 242 | return endrec |
| 243 | |
| 244 | # Update the original endrec using data from the ZIP64 record |
Antoine Pitrou | 9e4fdf4 | 2008-09-05 23:43:02 +0000 | [diff] [blame] | 245 | endrec[_ECD_SIGNATURE] = sig |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 246 | endrec[_ECD_DISK_NUMBER] = disk_num |
| 247 | endrec[_ECD_DISK_START] = disk_dir |
| 248 | endrec[_ECD_ENTRIES_THIS_DISK] = dircount |
| 249 | endrec[_ECD_ENTRIES_TOTAL] = dircount2 |
| 250 | endrec[_ECD_SIZE] = dirsize |
| 251 | endrec[_ECD_OFFSET] = diroffset |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 252 | return endrec |
| 253 | |
| 254 | |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 255 | def _EndRecData(fpin): |
| 256 | """Return data from the "End of Central Directory" record, or None. |
| 257 | |
| 258 | The data is a list of the nine items in the ZIP "End of central dir" |
| 259 | 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] | 260 | |
| 261 | # Determine file size |
| 262 | fpin.seek(0, 2) |
| 263 | filesize = fpin.tell() |
| 264 | |
| 265 | # Check to see if this is ZIP file with no archive comment (the |
| 266 | # "end of central directory" structure should be the last item in the |
| 267 | # file if this is the case). |
Amaury Forgeot d'Arc | bc34780 | 2009-07-28 22:18:57 +0000 | [diff] [blame] | 268 | try: |
| 269 | fpin.seek(-sizeEndCentDir, 2) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 270 | except OSError: |
Amaury Forgeot d'Arc | bc34780 | 2009-07-28 22:18:57 +0000 | [diff] [blame] | 271 | return None |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 272 | data = fpin.read() |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 273 | if (len(data) == sizeEndCentDir and |
| 274 | data[0:4] == stringEndArchive and |
| 275 | data[-2:] == b"\000\000"): |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 276 | # the signature is correct and there's no comment, unpack structure |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 277 | endrec = struct.unpack(structEndArchive, data) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 278 | endrec=list(endrec) |
| 279 | |
| 280 | # Append a blank comment and record start offset |
| 281 | endrec.append(b"") |
| 282 | endrec.append(filesize - sizeEndCentDir) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 283 | |
Amaury Forgeot d'Arc | d3fb4bb | 2009-01-18 00:29:02 +0000 | [diff] [blame] | 284 | # Try to read the "Zip64 end of central directory" structure |
| 285 | return _EndRecData64(fpin, -sizeEndCentDir, endrec) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 286 | |
| 287 | # Either this is not a ZIP file, or it is a ZIP file with an archive |
| 288 | # comment. Search the end of the file for the "end of central directory" |
| 289 | # record signature. The comment is the last item in the ZIP file and may be |
| 290 | # up to 64K long. It is assumed that the "end of central directory" magic |
| 291 | # number does not appear in the comment. |
| 292 | maxCommentStart = max(filesize - (1 << 16) - sizeEndCentDir, 0) |
| 293 | fpin.seek(maxCommentStart, 0) |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 294 | data = fpin.read() |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 295 | start = data.rfind(stringEndArchive) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 296 | if start >= 0: |
| 297 | # found the magic number; attempt to unpack and interpret |
| 298 | recData = data[start:start+sizeEndCentDir] |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 299 | if len(recData) != sizeEndCentDir: |
| 300 | # Zip file is corrupted. |
| 301 | return None |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 302 | endrec = list(struct.unpack(structEndArchive, recData)) |
R David Murray | 4fbb9db | 2011-06-09 15:50:51 -0400 | [diff] [blame] | 303 | commentSize = endrec[_ECD_COMMENT_SIZE] #as claimed by the zip file |
| 304 | comment = data[start+sizeEndCentDir:start+sizeEndCentDir+commentSize] |
| 305 | endrec.append(comment) |
| 306 | endrec.append(maxCommentStart + start) |
Amaury Forgeot d'Arc | d3fb4bb | 2009-01-18 00:29:02 +0000 | [diff] [blame] | 307 | |
R David Murray | 4fbb9db | 2011-06-09 15:50:51 -0400 | [diff] [blame] | 308 | # Try to read the "Zip64 end of central directory" structure |
| 309 | return _EndRecData64(fpin, maxCommentStart + start - filesize, |
| 310 | endrec) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 311 | |
| 312 | # Unable to find a valid end of central directory structure |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 313 | return None |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 314 | |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 315 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 316 | class ZipInfo (object): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 317 | """Class with attributes describing each file in the ZIP archive.""" |
| 318 | |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 319 | __slots__ = ( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 320 | 'orig_filename', |
| 321 | 'filename', |
| 322 | 'date_time', |
| 323 | 'compress_type', |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 324 | '_compresslevel', |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 325 | 'comment', |
| 326 | 'extra', |
| 327 | 'create_system', |
| 328 | 'create_version', |
| 329 | 'extract_version', |
| 330 | 'reserved', |
| 331 | 'flag_bits', |
| 332 | 'volume', |
| 333 | 'internal_attr', |
| 334 | 'external_attr', |
| 335 | 'header_offset', |
| 336 | 'CRC', |
| 337 | 'compress_size', |
| 338 | 'file_size', |
| 339 | '_raw_time', |
| 340 | ) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 341 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 342 | 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] | 343 | self.orig_filename = filename # Original file name in archive |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 344 | |
| 345 | # Terminate the file name at the first null byte. Null bytes in file |
| 346 | # names are used as tricks by viruses in archives. |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 347 | null_byte = filename.find(chr(0)) |
| 348 | if null_byte >= 0: |
| 349 | filename = filename[0:null_byte] |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 350 | # This is used to ensure paths in generated ZIP files always use |
| 351 | # forward slashes as the directory separator, as required by the |
| 352 | # ZIP format specification. |
| 353 | if os.sep != "/" and os.sep in filename: |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 354 | filename = filename.replace(os.sep, "/") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 355 | |
Greg Ward | 8e36d28 | 2003-06-18 00:53:06 +0000 | [diff] [blame] | 356 | self.filename = filename # Normalized file name |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 357 | self.date_time = date_time # year, month, day, hour, min, sec |
Senthil Kumaran | 29fa9d4 | 2011-10-20 01:46:00 +0800 | [diff] [blame] | 358 | |
| 359 | if date_time[0] < 1980: |
| 360 | raise ValueError('ZIP does not support timestamps before 1980') |
| 361 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 362 | # Standard values: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 363 | self.compress_type = ZIP_STORED # Type of compression for the file |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 364 | self._compresslevel = None # Level for the compressor |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 365 | self.comment = b"" # Comment for each file |
| 366 | self.extra = b"" # ZIP extra data |
Martin v. Löwis | 0075690 | 2006-02-05 17:09:41 +0000 | [diff] [blame] | 367 | if sys.platform == 'win32': |
| 368 | self.create_system = 0 # System which created ZIP archive |
| 369 | else: |
| 370 | # Assume everything else is unix-y |
| 371 | self.create_system = 3 # System which created ZIP archive |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 372 | self.create_version = DEFAULT_VERSION # Version which created ZIP archive |
| 373 | self.extract_version = DEFAULT_VERSION # Version needed to extract archive |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 374 | self.reserved = 0 # Must be zero |
| 375 | self.flag_bits = 0 # ZIP flag bits |
| 376 | self.volume = 0 # Volume number of file header |
| 377 | self.internal_attr = 0 # Internal attributes |
| 378 | self.external_attr = 0 # External file attributes |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 379 | # Other attributes are set by class ZipFile: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 380 | # header_offset Byte offset to the file header |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 381 | # CRC CRC-32 of the uncompressed file |
| 382 | # compress_size Size of the compressed file |
| 383 | # file_size Size of the uncompressed file |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 384 | |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 385 | def __repr__(self): |
| 386 | result = ['<%s filename=%r' % (self.__class__.__name__, self.filename)] |
| 387 | if self.compress_type != ZIP_STORED: |
| 388 | result.append(' compress_type=%s' % |
| 389 | compressor_names.get(self.compress_type, |
| 390 | self.compress_type)) |
| 391 | hi = self.external_attr >> 16 |
| 392 | lo = self.external_attr & 0xFFFF |
| 393 | if hi: |
| 394 | result.append(' filemode=%r' % stat.filemode(hi)) |
| 395 | if lo: |
| 396 | result.append(' external_attr=%#x' % lo) |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 397 | isdir = self.is_dir() |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 398 | if not isdir or self.file_size: |
| 399 | result.append(' file_size=%r' % self.file_size) |
| 400 | if ((not isdir or self.compress_size) and |
| 401 | (self.compress_type != ZIP_STORED or |
| 402 | self.file_size != self.compress_size)): |
| 403 | result.append(' compress_size=%r' % self.compress_size) |
| 404 | result.append('>') |
| 405 | return ''.join(result) |
| 406 | |
Serhiy Storchaka | 182d7cd | 2013-01-15 00:31:39 +0200 | [diff] [blame] | 407 | def FileHeader(self, zip64=None): |
Serhiy Storchaka | 4bb186d | 2018-11-25 09:51:14 +0200 | [diff] [blame] | 408 | """Return the per-file header as a bytes object.""" |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 409 | dt = self.date_time |
| 410 | dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2] |
Tim Peters | 3caca23 | 2001-12-06 06:23:26 +0000 | [diff] [blame] | 411 | dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 412 | if self.flag_bits & 0x08: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 413 | # Set these to zero because we write them after the file data |
| 414 | CRC = compress_size = file_size = 0 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 415 | else: |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 416 | CRC = self.CRC |
| 417 | compress_size = self.compress_size |
| 418 | file_size = self.file_size |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 419 | |
| 420 | extra = self.extra |
| 421 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 422 | min_version = 0 |
Serhiy Storchaka | 182d7cd | 2013-01-15 00:31:39 +0200 | [diff] [blame] | 423 | if zip64 is None: |
| 424 | zip64 = file_size > ZIP64_LIMIT or compress_size > ZIP64_LIMIT |
| 425 | if zip64: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 426 | fmt = '<HHQQ' |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 427 | extra = extra + struct.pack(fmt, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 428 | 1, struct.calcsize(fmt)-4, file_size, compress_size) |
Serhiy Storchaka | 182d7cd | 2013-01-15 00:31:39 +0200 | [diff] [blame] | 429 | if file_size > ZIP64_LIMIT or compress_size > ZIP64_LIMIT: |
| 430 | if not zip64: |
| 431 | raise LargeZipFile("Filesize would require ZIP64 extensions") |
| 432 | # File is larger than what fits into a 4 byte integer, |
| 433 | # fall back to the ZIP64 extension |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 434 | file_size = 0xffffffff |
| 435 | compress_size = 0xffffffff |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 436 | min_version = ZIP64_VERSION |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 437 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 438 | if self.compress_type == ZIP_BZIP2: |
| 439 | min_version = max(BZIP2_VERSION, min_version) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 440 | elif self.compress_type == ZIP_LZMA: |
| 441 | min_version = max(LZMA_VERSION, min_version) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 442 | |
| 443 | self.extract_version = max(min_version, self.extract_version) |
| 444 | self.create_version = max(min_version, self.create_version) |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 445 | filename, flag_bits = self._encodeFilenameFlags() |
Georg Brandl | 2ee470f | 2008-07-16 12:55:28 +0000 | [diff] [blame] | 446 | header = struct.pack(structFileHeader, stringFileHeader, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 447 | self.extract_version, self.reserved, flag_bits, |
| 448 | self.compress_type, dostime, dosdate, CRC, |
| 449 | compress_size, file_size, |
| 450 | len(filename), len(extra)) |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 451 | return header + filename + extra |
| 452 | |
| 453 | def _encodeFilenameFlags(self): |
| 454 | try: |
| 455 | return self.filename.encode('ascii'), self.flag_bits |
| 456 | except UnicodeEncodeError: |
| 457 | return self.filename.encode('utf-8'), self.flag_bits | 0x800 |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 458 | |
| 459 | def _decodeExtra(self): |
| 460 | # Try to decode the extra field. |
| 461 | extra = self.extra |
| 462 | unpack = struct.unpack |
Gregory P. Smith | 0af8a86 | 2014-05-29 23:42:14 -0700 | [diff] [blame] | 463 | while len(extra) >= 4: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 464 | tp, ln = unpack('<HH', extra[:4]) |
Serhiy Storchaka | feccdb2 | 2017-03-09 18:34:03 +0200 | [diff] [blame] | 465 | if ln+4 > len(extra): |
| 466 | raise BadZipFile("Corrupt extra field %04x (size=%d)" % (tp, ln)) |
| 467 | if tp == 0x0001: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 468 | if ln >= 24: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 469 | counts = unpack('<QQQ', extra[4:28]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 470 | elif ln == 16: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 471 | counts = unpack('<QQ', extra[4:20]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 472 | elif ln == 8: |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 473 | counts = unpack('<Q', extra[4:12]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 474 | elif ln == 0: |
| 475 | counts = () |
| 476 | else: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 477 | raise BadZipFile("Corrupt extra field %04x (size=%d)" % (tp, ln)) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 478 | |
| 479 | idx = 0 |
| 480 | |
| 481 | # ZIP64 extension (large files and/or large archives) |
Christian Heimes | d5e2b6f | 2008-03-19 21:50:51 +0000 | [diff] [blame] | 482 | if self.file_size in (0xffffffffffffffff, 0xffffffff): |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 483 | self.file_size = counts[idx] |
| 484 | idx += 1 |
| 485 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 486 | if self.compress_size == 0xFFFFFFFF: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 487 | self.compress_size = counts[idx] |
| 488 | idx += 1 |
| 489 | |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 490 | if self.header_offset == 0xffffffff: |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 491 | old = self.header_offset |
| 492 | self.header_offset = counts[idx] |
| 493 | idx+=1 |
| 494 | |
| 495 | extra = extra[ln+4:] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 496 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 497 | @classmethod |
Marcel Plch | a2fe1e5 | 2018-08-02 15:04:52 +0200 | [diff] [blame] | 498 | def from_file(cls, filename, arcname=None, *, strict_timestamps=True): |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 499 | """Construct an appropriate ZipInfo for a file on the filesystem. |
| 500 | |
| 501 | filename should be the path to a file or directory on the filesystem. |
| 502 | |
| 503 | arcname is the name which it will have within the archive (by default, |
| 504 | this will be the same as filename, but without a drive letter and with |
| 505 | leading path separators removed). |
| 506 | """ |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 507 | if isinstance(filename, os.PathLike): |
| 508 | filename = os.fspath(filename) |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 509 | st = os.stat(filename) |
| 510 | isdir = stat.S_ISDIR(st.st_mode) |
| 511 | mtime = time.localtime(st.st_mtime) |
| 512 | date_time = mtime[0:6] |
Marcel Plch | a2fe1e5 | 2018-08-02 15:04:52 +0200 | [diff] [blame] | 513 | if not strict_timestamps and date_time[0] < 1980: |
| 514 | date_time = (1980, 1, 1, 0, 0, 0) |
| 515 | elif not strict_timestamps and date_time[0] > 2107: |
| 516 | date_time = (2107, 12, 31, 23, 59, 59) |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 517 | # Create ZipInfo instance to store file information |
| 518 | if arcname is None: |
| 519 | arcname = filename |
| 520 | arcname = os.path.normpath(os.path.splitdrive(arcname)[1]) |
| 521 | while arcname[0] in (os.sep, os.altsep): |
| 522 | arcname = arcname[1:] |
| 523 | if isdir: |
| 524 | arcname += '/' |
| 525 | zinfo = cls(arcname, date_time) |
| 526 | zinfo.external_attr = (st.st_mode & 0xFFFF) << 16 # Unix attributes |
| 527 | if isdir: |
| 528 | zinfo.file_size = 0 |
| 529 | zinfo.external_attr |= 0x10 # MS-DOS directory flag |
| 530 | else: |
| 531 | zinfo.file_size = st.st_size |
| 532 | |
| 533 | return zinfo |
| 534 | |
| 535 | def is_dir(self): |
Serhiy Storchaka | f47fc55 | 2016-05-15 12:27:16 +0300 | [diff] [blame] | 536 | """Return True if this archive member is a directory.""" |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 537 | return self.filename[-1] == '/' |
| 538 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 539 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 540 | # ZIP encryption uses the CRC32 one-byte primitive for scrambling some |
| 541 | # internal keys. We noticed that a direct implementation is faster than |
| 542 | # relying on binascii.crc32(). |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 543 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 544 | _crctable = None |
| 545 | def _gen_crc(crc): |
| 546 | for j in range(8): |
| 547 | if crc & 1: |
| 548 | crc = (crc >> 1) ^ 0xEDB88320 |
| 549 | else: |
| 550 | crc >>= 1 |
| 551 | return crc |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 552 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 553 | # ZIP supports a password-based form of encryption. Even though known |
| 554 | # plaintext attacks have been found against it, it is still useful |
| 555 | # to be able to get data out of such a file. |
| 556 | # |
| 557 | # Usage: |
| 558 | # zd = _ZipDecrypter(mypwd) |
| 559 | # plain_bytes = zd(cypher_bytes) |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 560 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 561 | def _ZipDecrypter(pwd): |
| 562 | key0 = 305419896 |
| 563 | key1 = 591751049 |
| 564 | key2 = 878082192 |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 565 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 566 | global _crctable |
| 567 | if _crctable is None: |
| 568 | _crctable = list(map(_gen_crc, range(256))) |
| 569 | crctable = _crctable |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 570 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 571 | def crc32(ch, crc): |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 572 | """Compute the CRC32 primitive on one byte.""" |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 573 | return (crc >> 8) ^ crctable[(crc ^ ch) & 0xFF] |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 574 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 575 | def update_keys(c): |
| 576 | nonlocal key0, key1, key2 |
| 577 | key0 = crc32(c, key0) |
| 578 | key1 = (key1 + (key0 & 0xFF)) & 0xFFFFFFFF |
| 579 | key1 = (key1 * 134775813 + 1) & 0xFFFFFFFF |
| 580 | key2 = crc32(key1 >> 24, key2) |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 581 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 582 | for p in pwd: |
| 583 | update_keys(p) |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 584 | |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 585 | def decrypter(data): |
| 586 | """Decrypt a bytes object.""" |
| 587 | result = bytearray() |
| 588 | append = result.append |
| 589 | for c in data: |
| 590 | k = key2 | 2 |
| 591 | c ^= ((k * (k^1)) >> 8) & 0xFF |
| 592 | update_keys(c) |
| 593 | append(c) |
| 594 | return bytes(result) |
| 595 | |
| 596 | return decrypter |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 597 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 598 | |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 599 | class LZMACompressor: |
| 600 | |
| 601 | def __init__(self): |
| 602 | self._comp = None |
| 603 | |
| 604 | def _init(self): |
Nadeem Vawda | a425c3d | 2012-06-21 23:36:48 +0200 | [diff] [blame] | 605 | props = lzma._encode_filter_properties({'id': lzma.FILTER_LZMA1}) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 606 | self._comp = lzma.LZMACompressor(lzma.FORMAT_RAW, filters=[ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 607 | lzma._decode_filter_properties(lzma.FILTER_LZMA1, props) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 608 | ]) |
| 609 | return struct.pack('<BBH', 9, 4, len(props)) + props |
| 610 | |
| 611 | def compress(self, data): |
| 612 | if self._comp is None: |
| 613 | return self._init() + self._comp.compress(data) |
| 614 | return self._comp.compress(data) |
| 615 | |
| 616 | def flush(self): |
| 617 | if self._comp is None: |
| 618 | return self._init() + self._comp.flush() |
| 619 | return self._comp.flush() |
| 620 | |
| 621 | |
| 622 | class LZMADecompressor: |
| 623 | |
| 624 | def __init__(self): |
| 625 | self._decomp = None |
| 626 | self._unconsumed = b'' |
| 627 | self.eof = False |
| 628 | |
| 629 | def decompress(self, data): |
| 630 | if self._decomp is None: |
| 631 | self._unconsumed += data |
| 632 | if len(self._unconsumed) <= 4: |
| 633 | return b'' |
| 634 | psize, = struct.unpack('<H', self._unconsumed[2:4]) |
| 635 | if len(self._unconsumed) <= 4 + psize: |
| 636 | return b'' |
| 637 | |
| 638 | self._decomp = lzma.LZMADecompressor(lzma.FORMAT_RAW, filters=[ |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 639 | lzma._decode_filter_properties(lzma.FILTER_LZMA1, |
| 640 | self._unconsumed[4:4 + psize]) |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 641 | ]) |
| 642 | data = self._unconsumed[4 + psize:] |
| 643 | del self._unconsumed |
| 644 | |
| 645 | result = self._decomp.decompress(data) |
| 646 | self.eof = self._decomp.eof |
| 647 | return result |
| 648 | |
| 649 | |
| 650 | compressor_names = { |
| 651 | 0: 'store', |
| 652 | 1: 'shrink', |
| 653 | 2: 'reduce', |
| 654 | 3: 'reduce', |
| 655 | 4: 'reduce', |
| 656 | 5: 'reduce', |
| 657 | 6: 'implode', |
| 658 | 7: 'tokenize', |
| 659 | 8: 'deflate', |
| 660 | 9: 'deflate64', |
| 661 | 10: 'implode', |
| 662 | 12: 'bzip2', |
| 663 | 14: 'lzma', |
| 664 | 18: 'terse', |
| 665 | 19: 'lz77', |
| 666 | 97: 'wavpack', |
| 667 | 98: 'ppmd', |
| 668 | } |
| 669 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 670 | def _check_compression(compression): |
| 671 | if compression == ZIP_STORED: |
| 672 | pass |
| 673 | elif compression == ZIP_DEFLATED: |
| 674 | if not zlib: |
| 675 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 676 | "Compression requires the (missing) zlib module") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 677 | elif compression == ZIP_BZIP2: |
| 678 | if not bz2: |
| 679 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 680 | "Compression requires the (missing) bz2 module") |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 681 | elif compression == ZIP_LZMA: |
| 682 | if not lzma: |
| 683 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 684 | "Compression requires the (missing) lzma module") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 685 | else: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 686 | raise NotImplementedError("That compression method is not supported") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 687 | |
| 688 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 689 | def _get_compressor(compress_type, compresslevel=None): |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 690 | if compress_type == ZIP_DEFLATED: |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 691 | if compresslevel is not None: |
| 692 | return zlib.compressobj(compresslevel, zlib.DEFLATED, -15) |
| 693 | return zlib.compressobj(zlib.Z_DEFAULT_COMPRESSION, zlib.DEFLATED, -15) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 694 | elif compress_type == ZIP_BZIP2: |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 695 | if compresslevel is not None: |
| 696 | return bz2.BZ2Compressor(compresslevel) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 697 | return bz2.BZ2Compressor() |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 698 | # compresslevel is ignored for ZIP_LZMA |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 699 | elif compress_type == ZIP_LZMA: |
| 700 | return LZMACompressor() |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 701 | else: |
| 702 | return None |
| 703 | |
| 704 | |
| 705 | def _get_decompressor(compress_type): |
Martin v. Löwis | b3260f0 | 2012-05-01 08:38:01 +0200 | [diff] [blame] | 706 | if compress_type == ZIP_STORED: |
| 707 | return None |
| 708 | elif compress_type == ZIP_DEFLATED: |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 709 | return zlib.decompressobj(-15) |
| 710 | elif compress_type == ZIP_BZIP2: |
| 711 | return bz2.BZ2Decompressor() |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 712 | elif compress_type == ZIP_LZMA: |
| 713 | return LZMADecompressor() |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 714 | else: |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 715 | descr = compressor_names.get(compress_type) |
Martin v. Löwis | b3260f0 | 2012-05-01 08:38:01 +0200 | [diff] [blame] | 716 | if descr: |
| 717 | raise NotImplementedError("compression type %d (%s)" % (compress_type, descr)) |
| 718 | else: |
| 719 | raise NotImplementedError("compression type %d" % (compress_type,)) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 720 | |
| 721 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 722 | class _SharedFile: |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 723 | def __init__(self, file, pos, close, lock, writing): |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 724 | self._file = file |
| 725 | self._pos = pos |
| 726 | self._close = close |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 727 | self._lock = lock |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 728 | self._writing = writing |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 729 | self.seekable = file.seekable |
| 730 | self.tell = file.tell |
| 731 | |
| 732 | def seek(self, offset, whence=0): |
| 733 | with self._lock: |
Mickaël Schoentgen | 3f8c691 | 2018-07-29 20:26:52 +0200 | [diff] [blame] | 734 | if self._writing(): |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 735 | raise ValueError("Can't reposition in the ZIP file while " |
| 736 | "there is an open writing handle on it. " |
| 737 | "Close the writing handle before trying to read.") |
Mickaël Schoentgen | 3f8c691 | 2018-07-29 20:26:52 +0200 | [diff] [blame] | 738 | self._file.seek(offset, whence) |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 739 | self._pos = self._file.tell() |
| 740 | return self._pos |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 741 | |
| 742 | def read(self, n=-1): |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 743 | with self._lock: |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 744 | if self._writing(): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 745 | raise ValueError("Can't read from the ZIP file while there " |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 746 | "is an open writing handle on it. " |
| 747 | "Close the writing handle before trying to read.") |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 748 | self._file.seek(self._pos) |
| 749 | data = self._file.read(n) |
| 750 | self._pos = self._file.tell() |
| 751 | return data |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 752 | |
| 753 | def close(self): |
| 754 | if self._file is not None: |
| 755 | fileobj = self._file |
| 756 | self._file = None |
| 757 | self._close(fileobj) |
| 758 | |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 759 | # Provide the tell method for unseekable stream |
| 760 | class _Tellable: |
| 761 | def __init__(self, fp): |
| 762 | self.fp = fp |
| 763 | self.offset = 0 |
| 764 | |
| 765 | def write(self, data): |
| 766 | n = self.fp.write(data) |
| 767 | self.offset += n |
| 768 | return n |
| 769 | |
| 770 | def tell(self): |
| 771 | return self.offset |
| 772 | |
| 773 | def flush(self): |
| 774 | self.fp.flush() |
| 775 | |
| 776 | def close(self): |
| 777 | self.fp.close() |
| 778 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 779 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 780 | class ZipExtFile(io.BufferedIOBase): |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 781 | """File-like object for reading an archive member. |
| 782 | Is returned by ZipFile.open(). |
| 783 | """ |
| 784 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 785 | # Max size supported by decompressor. |
| 786 | MAX_N = 1 << 31 - 1 |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 787 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 788 | # Read from compressed files in 4k blocks. |
| 789 | MIN_READ_SIZE = 4096 |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 790 | |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 791 | # Chunk size to read during seek |
| 792 | MAX_SEEK_READ = 1 << 24 |
| 793 | |
Łukasz Langa | e94980a | 2010-11-22 23:31:26 +0000 | [diff] [blame] | 794 | def __init__(self, fileobj, mode, zipinfo, decrypter=None, |
| 795 | close_fileobj=False): |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 796 | self._fileobj = fileobj |
| 797 | self._decrypter = decrypter |
Łukasz Langa | e94980a | 2010-11-22 23:31:26 +0000 | [diff] [blame] | 798 | self._close_fileobj = close_fileobj |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 799 | |
Ezio Melotti | 92b4743 | 2010-01-28 01:44:41 +0000 | [diff] [blame] | 800 | self._compress_type = zipinfo.compress_type |
Ezio Melotti | 92b4743 | 2010-01-28 01:44:41 +0000 | [diff] [blame] | 801 | self._compress_left = zipinfo.compress_size |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 802 | self._left = zipinfo.file_size |
Ezio Melotti | 92b4743 | 2010-01-28 01:44:41 +0000 | [diff] [blame] | 803 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 804 | self._decompressor = _get_decompressor(self._compress_type) |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 805 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 806 | self._eof = False |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 807 | self._readbuffer = b'' |
| 808 | self._offset = 0 |
| 809 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 810 | self.newlines = None |
| 811 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 812 | # Adjust read size for encrypted files since the first 12 bytes |
| 813 | # are for the encryption/password information. |
| 814 | if self._decrypter is not None: |
| 815 | self._compress_left -= 12 |
| 816 | |
| 817 | self.mode = mode |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 818 | self.name = zipinfo.filename |
| 819 | |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 820 | if hasattr(zipinfo, 'CRC'): |
| 821 | self._expected_crc = zipinfo.CRC |
Martin Panter | b82032f | 2015-12-11 05:19:29 +0000 | [diff] [blame] | 822 | self._running_crc = crc32(b'') |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 823 | else: |
| 824 | self._expected_crc = None |
| 825 | |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 826 | self._seekable = False |
| 827 | try: |
| 828 | if fileobj.seekable(): |
| 829 | self._orig_compress_start = fileobj.tell() |
| 830 | self._orig_compress_size = zipinfo.compress_size |
| 831 | self._orig_file_size = zipinfo.file_size |
| 832 | self._orig_start_crc = self._running_crc |
| 833 | self._seekable = True |
| 834 | except AttributeError: |
| 835 | pass |
| 836 | |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 837 | def __repr__(self): |
| 838 | result = ['<%s.%s' % (self.__class__.__module__, |
| 839 | self.__class__.__qualname__)] |
| 840 | if not self.closed: |
| 841 | result.append(' name=%r mode=%r' % (self.name, self.mode)) |
| 842 | if self._compress_type != ZIP_STORED: |
| 843 | result.append(' compress_type=%s' % |
| 844 | compressor_names.get(self._compress_type, |
| 845 | self._compress_type)) |
| 846 | else: |
| 847 | result.append(' [closed]') |
| 848 | result.append('>') |
| 849 | return ''.join(result) |
| 850 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 851 | def readline(self, limit=-1): |
| 852 | """Read and return a line from the stream. |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 853 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 854 | If limit is specified, at most limit bytes will be read. |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 855 | """ |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 856 | |
Serhiy Storchaka | e670be2 | 2016-06-11 19:32:44 +0300 | [diff] [blame] | 857 | if limit < 0: |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 858 | # Shortcut common case - newline found in buffer. |
| 859 | i = self._readbuffer.find(b'\n', self._offset) + 1 |
| 860 | if i > 0: |
| 861 | line = self._readbuffer[self._offset: i] |
| 862 | self._offset = i |
| 863 | return line |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 864 | |
Serhiy Storchaka | e670be2 | 2016-06-11 19:32:44 +0300 | [diff] [blame] | 865 | return io.BufferedIOBase.readline(self, limit) |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 866 | |
| 867 | def peek(self, n=1): |
| 868 | """Returns buffered bytes without advancing the position.""" |
| 869 | if n > len(self._readbuffer) - self._offset: |
| 870 | chunk = self.read(n) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 871 | if len(chunk) > self._offset: |
| 872 | self._readbuffer = chunk + self._readbuffer[self._offset:] |
| 873 | self._offset = 0 |
| 874 | else: |
| 875 | self._offset -= len(chunk) |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 876 | |
| 877 | # Return up to 512 bytes to reduce allocation overhead for tight loops. |
| 878 | return self._readbuffer[self._offset: self._offset + 512] |
| 879 | |
| 880 | def readable(self): |
| 881 | return True |
| 882 | |
| 883 | def read(self, n=-1): |
| 884 | """Read and return up to n bytes. |
nick sung | 53c2935 | 2019-03-15 03:26:25 +0800 | [diff] [blame] | 885 | 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] | 886 | """ |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 887 | if n is None or n < 0: |
| 888 | buf = self._readbuffer[self._offset:] |
| 889 | self._readbuffer = b'' |
| 890 | self._offset = 0 |
| 891 | while not self._eof: |
| 892 | buf += self._read1(self.MAX_N) |
| 893 | return buf |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 894 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 895 | end = n + self._offset |
| 896 | if end < len(self._readbuffer): |
| 897 | buf = self._readbuffer[self._offset:end] |
| 898 | self._offset = end |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 899 | return buf |
| 900 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 901 | n = end - len(self._readbuffer) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 902 | buf = self._readbuffer[self._offset:] |
| 903 | self._readbuffer = b'' |
| 904 | self._offset = 0 |
| 905 | while n > 0 and not self._eof: |
| 906 | data = self._read1(n) |
| 907 | if n < len(data): |
| 908 | self._readbuffer = data |
| 909 | self._offset = n |
| 910 | buf += data[:n] |
| 911 | break |
| 912 | buf += data |
| 913 | n -= len(data) |
| 914 | return buf |
| 915 | |
| 916 | def _update_crc(self, newdata): |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 917 | # Update the CRC using the given data. |
| 918 | if self._expected_crc is None: |
| 919 | # No need to compute the CRC if we don't have a reference value |
| 920 | return |
Martin Panter | b82032f | 2015-12-11 05:19:29 +0000 | [diff] [blame] | 921 | self._running_crc = crc32(newdata, self._running_crc) |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 922 | # 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] | 923 | if self._eof and self._running_crc != self._expected_crc: |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 924 | raise BadZipFile("Bad CRC-32 for file %r" % self.name) |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 925 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 926 | def read1(self, n): |
| 927 | """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] | 928 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 929 | if n is None or n < 0: |
| 930 | buf = self._readbuffer[self._offset:] |
| 931 | self._readbuffer = b'' |
| 932 | self._offset = 0 |
Serhiy Storchaka | d2c07a5 | 2013-09-27 22:11:57 +0300 | [diff] [blame] | 933 | while not self._eof: |
| 934 | data = self._read1(self.MAX_N) |
| 935 | if data: |
| 936 | buf += data |
| 937 | break |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 938 | return buf |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 939 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 940 | end = n + self._offset |
| 941 | if end < len(self._readbuffer): |
| 942 | buf = self._readbuffer[self._offset:end] |
| 943 | self._offset = end |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 944 | return buf |
| 945 | |
Antoine Pitrou | 78157b3 | 2012-06-23 16:44:48 +0200 | [diff] [blame] | 946 | n = end - len(self._readbuffer) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 947 | buf = self._readbuffer[self._offset:] |
| 948 | self._readbuffer = b'' |
| 949 | self._offset = 0 |
| 950 | if n > 0: |
Serhiy Storchaka | d2c07a5 | 2013-09-27 22:11:57 +0300 | [diff] [blame] | 951 | while not self._eof: |
| 952 | data = self._read1(n) |
| 953 | if n < len(data): |
| 954 | self._readbuffer = data |
| 955 | self._offset = n |
| 956 | buf += data[:n] |
| 957 | break |
| 958 | if data: |
| 959 | buf += data |
| 960 | break |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 961 | return buf |
| 962 | |
| 963 | def _read1(self, n): |
| 964 | # Read up to n compressed bytes with at most one read() system call, |
| 965 | # decrypt and decompress them. |
| 966 | if self._eof or n <= 0: |
| 967 | return b'' |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 968 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 969 | # Read from file. |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 970 | if self._compress_type == ZIP_DEFLATED: |
| 971 | ## Handle unconsumed data. |
| 972 | data = self._decompressor.unconsumed_tail |
| 973 | if n > len(data): |
| 974 | data += self._read2(n - len(data)) |
| 975 | else: |
| 976 | data = self._read2(n) |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 977 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 978 | if self._compress_type == ZIP_STORED: |
| 979 | self._eof = self._compress_left <= 0 |
| 980 | elif self._compress_type == ZIP_DEFLATED: |
| 981 | n = max(n, self.MIN_READ_SIZE) |
| 982 | data = self._decompressor.decompress(data, n) |
| 983 | self._eof = (self._decompressor.eof or |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 984 | self._compress_left <= 0 and |
| 985 | not self._decompressor.unconsumed_tail) |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 986 | if self._eof: |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 987 | data += self._decompressor.flush() |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 988 | else: |
| 989 | data = self._decompressor.decompress(data) |
| 990 | self._eof = self._decompressor.eof or self._compress_left <= 0 |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 991 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 992 | data = data[:self._left] |
| 993 | self._left -= len(data) |
| 994 | if self._left <= 0: |
| 995 | self._eof = True |
| 996 | self._update_crc(data) |
| 997 | return data |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 998 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 999 | def _read2(self, n): |
| 1000 | if self._compress_left <= 0: |
| 1001 | return b'' |
| 1002 | |
| 1003 | n = max(n, self.MIN_READ_SIZE) |
| 1004 | n = min(n, self._compress_left) |
| 1005 | |
| 1006 | data = self._fileobj.read(n) |
| 1007 | self._compress_left -= len(data) |
Serhiy Storchaka | 5ce3f10 | 2014-01-09 14:50:20 +0200 | [diff] [blame] | 1008 | if not data: |
| 1009 | raise EOFError |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 1010 | |
| 1011 | if self._decrypter is not None: |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 1012 | data = self._decrypter(data) |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1013 | return data |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1014 | |
Łukasz Langa | e94980a | 2010-11-22 23:31:26 +0000 | [diff] [blame] | 1015 | def close(self): |
| 1016 | try: |
| 1017 | if self._close_fileobj: |
| 1018 | self._fileobj.close() |
| 1019 | finally: |
| 1020 | super().close() |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1021 | |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 1022 | def seekable(self): |
| 1023 | return self._seekable |
| 1024 | |
| 1025 | def seek(self, offset, whence=0): |
| 1026 | if not self._seekable: |
| 1027 | raise io.UnsupportedOperation("underlying stream is not seekable") |
| 1028 | curr_pos = self.tell() |
| 1029 | if whence == 0: # Seek from start of file |
| 1030 | new_pos = offset |
| 1031 | elif whence == 1: # Seek from current position |
| 1032 | new_pos = curr_pos + offset |
| 1033 | elif whence == 2: # Seek from EOF |
| 1034 | new_pos = self._orig_file_size + offset |
| 1035 | else: |
| 1036 | raise ValueError("whence must be os.SEEK_SET (0), " |
| 1037 | "os.SEEK_CUR (1), or os.SEEK_END (2)") |
| 1038 | |
| 1039 | if new_pos > self._orig_file_size: |
| 1040 | new_pos = self._orig_file_size |
| 1041 | |
| 1042 | if new_pos < 0: |
| 1043 | new_pos = 0 |
| 1044 | |
| 1045 | read_offset = new_pos - curr_pos |
| 1046 | buff_offset = read_offset + self._offset |
| 1047 | |
| 1048 | if buff_offset >= 0 and buff_offset < len(self._readbuffer): |
| 1049 | # Just move the _offset index if the new position is in the _readbuffer |
| 1050 | self._offset = buff_offset |
| 1051 | read_offset = 0 |
| 1052 | elif read_offset < 0: |
| 1053 | # Position is before the current position. Reset the ZipExtFile |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 1054 | self._fileobj.seek(self._orig_compress_start) |
| 1055 | self._running_crc = self._orig_start_crc |
| 1056 | self._compress_left = self._orig_compress_size |
| 1057 | self._left = self._orig_file_size |
| 1058 | self._readbuffer = b'' |
| 1059 | self._offset = 0 |
Mickaël Schoentgen | 3f8c691 | 2018-07-29 20:26:52 +0200 | [diff] [blame] | 1060 | self._decompressor = _get_decompressor(self._compress_type) |
John Jolly | 066df4f | 2018-01-30 01:51:35 -0700 | [diff] [blame] | 1061 | self._eof = False |
| 1062 | read_offset = new_pos |
| 1063 | |
| 1064 | while read_offset > 0: |
| 1065 | read_len = min(self.MAX_SEEK_READ, read_offset) |
| 1066 | self.read(read_len) |
| 1067 | read_offset -= read_len |
| 1068 | |
| 1069 | return self.tell() |
| 1070 | |
| 1071 | def tell(self): |
| 1072 | if not self._seekable: |
| 1073 | raise io.UnsupportedOperation("underlying stream is not seekable") |
| 1074 | filepos = self._orig_file_size - self._left - len(self._readbuffer) + self._offset |
| 1075 | return filepos |
| 1076 | |
Antoine Pitrou | a32f9a2 | 2010-01-27 21:18:57 +0000 | [diff] [blame] | 1077 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1078 | class _ZipWriteFile(io.BufferedIOBase): |
| 1079 | def __init__(self, zf, zinfo, zip64): |
| 1080 | self._zinfo = zinfo |
| 1081 | self._zip64 = zip64 |
| 1082 | self._zipfile = zf |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1083 | self._compressor = _get_compressor(zinfo.compress_type, |
| 1084 | zinfo._compresslevel) |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1085 | self._file_size = 0 |
| 1086 | self._compress_size = 0 |
| 1087 | self._crc = 0 |
| 1088 | |
| 1089 | @property |
| 1090 | def _fileobj(self): |
| 1091 | return self._zipfile.fp |
| 1092 | |
| 1093 | def writable(self): |
| 1094 | return True |
| 1095 | |
| 1096 | def write(self, data): |
Serhiy Storchaka | 4c0d9ea | 2017-04-12 16:03:23 +0300 | [diff] [blame] | 1097 | if self.closed: |
| 1098 | raise ValueError('I/O operation on closed file.') |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1099 | nbytes = len(data) |
| 1100 | self._file_size += nbytes |
| 1101 | self._crc = crc32(data, self._crc) |
| 1102 | if self._compressor: |
| 1103 | data = self._compressor.compress(data) |
| 1104 | self._compress_size += len(data) |
| 1105 | self._fileobj.write(data) |
| 1106 | return nbytes |
| 1107 | |
| 1108 | def close(self): |
Serhiy Storchaka | 4c0d9ea | 2017-04-12 16:03:23 +0300 | [diff] [blame] | 1109 | if self.closed: |
| 1110 | return |
Serhiy Storchaka | 2524fde | 2019-03-30 08:25:19 +0200 | [diff] [blame] | 1111 | try: |
| 1112 | super().close() |
| 1113 | # Flush any data from the compressor, and update header info |
| 1114 | if self._compressor: |
| 1115 | buf = self._compressor.flush() |
| 1116 | self._compress_size += len(buf) |
| 1117 | self._fileobj.write(buf) |
| 1118 | self._zinfo.compress_size = self._compress_size |
| 1119 | else: |
| 1120 | self._zinfo.compress_size = self._file_size |
| 1121 | self._zinfo.CRC = self._crc |
| 1122 | self._zinfo.file_size = self._file_size |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1123 | |
Serhiy Storchaka | 2524fde | 2019-03-30 08:25:19 +0200 | [diff] [blame] | 1124 | # Write updated header info |
| 1125 | if self._zinfo.flag_bits & 0x08: |
| 1126 | # Write CRC and file sizes after the file data |
| 1127 | fmt = '<LLQQ' if self._zip64 else '<LLLL' |
| 1128 | self._fileobj.write(struct.pack(fmt, _DD_SIGNATURE, self._zinfo.CRC, |
| 1129 | self._zinfo.compress_size, self._zinfo.file_size)) |
| 1130 | self._zipfile.start_dir = self._fileobj.tell() |
| 1131 | else: |
| 1132 | if not self._zip64: |
| 1133 | if self._file_size > ZIP64_LIMIT: |
| 1134 | raise RuntimeError( |
| 1135 | 'File size unexpectedly exceeded ZIP64 limit') |
| 1136 | if self._compress_size > ZIP64_LIMIT: |
| 1137 | raise RuntimeError( |
| 1138 | 'Compressed size unexpectedly exceeded ZIP64 limit') |
| 1139 | # Seek backwards and write file header (which will now include |
| 1140 | # correct CRC and file sizes) |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1141 | |
Serhiy Storchaka | 2524fde | 2019-03-30 08:25:19 +0200 | [diff] [blame] | 1142 | # Preserve current position in file |
| 1143 | self._zipfile.start_dir = self._fileobj.tell() |
| 1144 | self._fileobj.seek(self._zinfo.header_offset) |
| 1145 | self._fileobj.write(self._zinfo.FileHeader(self._zip64)) |
| 1146 | self._fileobj.seek(self._zipfile.start_dir) |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1147 | |
Serhiy Storchaka | 2524fde | 2019-03-30 08:25:19 +0200 | [diff] [blame] | 1148 | # Successfully written: Add file to our caches |
| 1149 | self._zipfile.filelist.append(self._zinfo) |
| 1150 | self._zipfile.NameToInfo[self._zinfo.filename] = self._zinfo |
| 1151 | finally: |
| 1152 | self._zipfile._writing = False |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1153 | |
Serhiy Storchaka | 2524fde | 2019-03-30 08:25:19 +0200 | [diff] [blame] | 1154 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1155 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1156 | class ZipFile: |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1157 | """ Class with methods to open, read, write, close, list zip files. |
| 1158 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1159 | z = ZipFile(file, mode="r", compression=ZIP_STORED, allowZip64=True, |
| 1160 | compresslevel=None) |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1161 | |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1162 | file: Either the path to the file, or a file-like object. |
| 1163 | 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] | 1164 | mode: The mode can be either read 'r', write 'w', exclusive create 'x', |
| 1165 | or append 'a'. |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 1166 | compression: ZIP_STORED (no compression), ZIP_DEFLATED (requires zlib), |
| 1167 | ZIP_BZIP2 (requires bz2) or ZIP_LZMA (requires lzma). |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1168 | allowZip64: if True ZipFile will create files with ZIP64 extensions when |
| 1169 | needed, otherwise it will raise an exception when this would |
| 1170 | be necessary. |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1171 | compresslevel: None (default for the given compression type) or an integer |
| 1172 | specifying the level to pass to the compressor. |
| 1173 | When using ZIP_STORED or ZIP_LZMA this keyword has no effect. |
| 1174 | When using ZIP_DEFLATED integers 0 through 9 are accepted. |
| 1175 | When using ZIP_BZIP2 integers 1 through 9 are accepted. |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1176 | |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1177 | """ |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1178 | |
Fred Drake | 90eac28 | 2001-02-28 05:29:34 +0000 | [diff] [blame] | 1179 | fp = None # Set here since __del__ checks it |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1180 | _windows_illegal_name_trans_table = None |
Fred Drake | 90eac28 | 2001-02-28 05:29:34 +0000 | [diff] [blame] | 1181 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1182 | def __init__(self, file, mode="r", compression=ZIP_STORED, allowZip64=True, |
Marcel Plch | 77b112c | 2018-08-31 16:43:31 +0200 | [diff] [blame] | 1183 | compresslevel=None, *, strict_timestamps=True): |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1184 | """Open the ZIP file with mode read 'r', write 'w', exclusive create 'x', |
| 1185 | or append 'a'.""" |
| 1186 | if mode not in ('r', 'w', 'x', 'a'): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1187 | raise ValueError("ZipFile requires mode 'r', 'w', 'x', or 'a'") |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1188 | |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 1189 | _check_compression(compression) |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1190 | |
| 1191 | self._allowZip64 = allowZip64 |
| 1192 | self._didModify = False |
Tim Peters | e119006 | 2001-01-15 03:34:38 +0000 | [diff] [blame] | 1193 | self.debug = 0 # Level of printing: 0 through 3 |
| 1194 | self.NameToInfo = {} # Find file info given name |
| 1195 | self.filelist = [] # List of ZipInfo instances for archive |
| 1196 | self.compression = compression # Method of compression |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1197 | self.compresslevel = compresslevel |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1198 | self.mode = mode |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1199 | self.pwd = None |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1200 | self._comment = b'' |
Marcel Plch | 77b112c | 2018-08-31 16:43:31 +0200 | [diff] [blame] | 1201 | self._strict_timestamps = strict_timestamps |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1202 | |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1203 | # Check if we were passed a file-like object |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 1204 | if isinstance(file, os.PathLike): |
| 1205 | file = os.fspath(file) |
Guido van Rossum | 3172c5d | 2007-10-16 18:12:55 +0000 | [diff] [blame] | 1206 | if isinstance(file, str): |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1207 | # No, it's a filename |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1208 | self._filePassed = 0 |
| 1209 | self.filename = file |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1210 | modeDict = {'r' : 'rb', 'w': 'w+b', 'x': 'x+b', 'a' : 'r+b', |
| 1211 | 'r+b': 'w+b', 'w+b': 'wb', 'x+b': 'xb'} |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1212 | filemode = modeDict[mode] |
| 1213 | while True: |
| 1214 | try: |
| 1215 | self.fp = io.open(file, filemode) |
| 1216 | except OSError: |
| 1217 | if filemode in modeDict: |
| 1218 | filemode = modeDict[filemode] |
| 1219 | continue |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1220 | raise |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1221 | break |
Fred Drake | 3d9091e | 2001-03-26 15:49:24 +0000 | [diff] [blame] | 1222 | else: |
| 1223 | self._filePassed = 1 |
| 1224 | self.fp = file |
| 1225 | self.filename = getattr(file, 'name', None) |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1226 | self._fileRefCnt = 1 |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1227 | self._lock = threading.RLock() |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1228 | self._seekable = True |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1229 | self._writing = False |
Tim Peters | a19a168 | 2001-03-29 04:36:09 +0000 | [diff] [blame] | 1230 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1231 | try: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1232 | if mode == 'r': |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 1233 | self._RealGetContents() |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1234 | elif mode in ('w', 'x'): |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 1235 | # set the modified flag so central directory gets written |
| 1236 | # even if no files are added to the archive |
| 1237 | self._didModify = True |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1238 | try: |
Serhiy Storchaka | 34cba33 | 2017-01-01 19:00:30 +0200 | [diff] [blame] | 1239 | self.start_dir = self.fp.tell() |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1240 | except (AttributeError, OSError): |
| 1241 | self.fp = _Tellable(self.fp) |
Serhiy Storchaka | 34cba33 | 2017-01-01 19:00:30 +0200 | [diff] [blame] | 1242 | self.start_dir = 0 |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1243 | self._seekable = False |
| 1244 | else: |
| 1245 | # Some file-like objects can provide tell() but not seek() |
| 1246 | try: |
| 1247 | self.fp.seek(self.start_dir) |
| 1248 | except (AttributeError, OSError): |
| 1249 | self._seekable = False |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1250 | elif mode == 'a': |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1251 | try: |
| 1252 | # See if file is a zip file |
| 1253 | self._RealGetContents() |
| 1254 | # seek to start of directory and overwrite |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1255 | self.fp.seek(self.start_dir) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1256 | except BadZipFile: |
| 1257 | # file is not a zip file, just append |
| 1258 | self.fp.seek(0, 2) |
| 1259 | |
| 1260 | # set the modified flag so central directory gets written |
| 1261 | # even if no files are added to the archive |
| 1262 | self._didModify = True |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1263 | self.start_dir = self.fp.tell() |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1264 | else: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1265 | raise ValueError("Mode must be 'r', 'w', 'x', or 'a'") |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1266 | except: |
| 1267 | fp = self.fp |
| 1268 | self.fp = None |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1269 | self._fpclose(fp) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1270 | raise |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1271 | |
Ezio Melotti | faa6b7f | 2009-12-30 12:34:59 +0000 | [diff] [blame] | 1272 | def __enter__(self): |
| 1273 | return self |
| 1274 | |
| 1275 | def __exit__(self, type, value, traceback): |
| 1276 | self.close() |
| 1277 | |
Serhiy Storchaka | 51a4370 | 2014-10-29 22:42:06 +0200 | [diff] [blame] | 1278 | def __repr__(self): |
| 1279 | result = ['<%s.%s' % (self.__class__.__module__, |
| 1280 | self.__class__.__qualname__)] |
| 1281 | if self.fp is not None: |
| 1282 | if self._filePassed: |
| 1283 | result.append(' file=%r' % self.fp) |
| 1284 | elif self.filename is not None: |
| 1285 | result.append(' filename=%r' % self.filename) |
| 1286 | result.append(' mode=%r' % self.mode) |
| 1287 | else: |
| 1288 | result.append(' [closed]') |
| 1289 | result.append('>') |
| 1290 | return ''.join(result) |
| 1291 | |
Tim Peters | 7d3bad6 | 2001-04-04 18:56:49 +0000 | [diff] [blame] | 1292 | def _RealGetContents(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1293 | """Read in the table of contents for the ZIP file.""" |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1294 | fp = self.fp |
Georg Brandl | 268e4d4 | 2010-10-14 06:59:45 +0000 | [diff] [blame] | 1295 | try: |
| 1296 | endrec = _EndRecData(fp) |
Andrew Svetlov | f7a17b4 | 2012-12-25 16:47:37 +0200 | [diff] [blame] | 1297 | except OSError: |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 1298 | raise BadZipFile("File is not a zip file") |
Martin v. Löwis | 6f6873b | 2002-10-13 13:54:50 +0000 | [diff] [blame] | 1299 | if not endrec: |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 1300 | raise BadZipFile("File is not a zip file") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1301 | if self.debug > 1: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1302 | print(endrec) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1303 | size_cd = endrec[_ECD_SIZE] # bytes in central directory |
| 1304 | offset_cd = endrec[_ECD_OFFSET] # offset of central directory |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1305 | self._comment = endrec[_ECD_COMMENT] # archive comment |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1306 | |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1307 | # "concat" is zero, unless zip was concatenated to another file |
| 1308 | concat = endrec[_ECD_LOCATION] - size_cd - offset_cd |
Antoine Pitrou | 9e4fdf4 | 2008-09-05 23:43:02 +0000 | [diff] [blame] | 1309 | if endrec[_ECD_SIGNATURE] == stringEndArchive64: |
| 1310 | # If Zip64 extension structures are present, account for them |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1311 | concat -= (sizeEndCentDir64 + sizeEndCentDir64Locator) |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1312 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1313 | if self.debug > 2: |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1314 | inferred = concat + offset_cd |
| 1315 | print("given, inferred, offset", offset_cd, inferred, concat) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1316 | # self.start_dir: Position of start of central directory |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1317 | self.start_dir = offset_cd + concat |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1318 | fp.seek(self.start_dir, 0) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1319 | data = fp.read(size_cd) |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1320 | fp = io.BytesIO(data) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1321 | total = 0 |
| 1322 | while total < size_cd: |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1323 | centdir = fp.read(sizeCentralDir) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 1324 | if len(centdir) != sizeCentralDir: |
| 1325 | raise BadZipFile("Truncated central directory") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1326 | centdir = struct.unpack(structCentralDir, centdir) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 1327 | if centdir[_CD_SIGNATURE] != stringCentralDir: |
| 1328 | raise BadZipFile("Bad magic number for central directory") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1329 | if self.debug > 2: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1330 | print(centdir) |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 1331 | filename = fp.read(centdir[_CD_FILENAME_LENGTH]) |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 1332 | flags = centdir[5] |
| 1333 | if flags & 0x800: |
| 1334 | # UTF-8 file names extension |
| 1335 | filename = filename.decode('utf-8') |
| 1336 | else: |
| 1337 | # Historical ZIP filename encoding |
| 1338 | filename = filename.decode('cp437') |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1339 | # Create ZipInfo instance to store file information |
Martin v. Löwis | 8570f6a | 2008-05-05 17:44:38 +0000 | [diff] [blame] | 1340 | x = ZipInfo(filename) |
Fred Drake | 3e038e5 | 2001-02-28 17:56:26 +0000 | [diff] [blame] | 1341 | x.extra = fp.read(centdir[_CD_EXTRA_FIELD_LENGTH]) |
| 1342 | x.comment = fp.read(centdir[_CD_COMMENT_LENGTH]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1343 | x.header_offset = centdir[_CD_LOCAL_HEADER_OFFSET] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1344 | (x.create_version, x.create_system, x.extract_version, x.reserved, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1345 | x.flag_bits, x.compress_type, t, d, |
| 1346 | x.CRC, x.compress_size, x.file_size) = centdir[1:12] |
Martin v. Löwis | d099b56 | 2012-05-01 14:08:22 +0200 | [diff] [blame] | 1347 | if x.extract_version > MAX_EXTRACT_VERSION: |
| 1348 | raise NotImplementedError("zip file version %.1f" % |
| 1349 | (x.extract_version / 10)) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1350 | x.volume, x.internal_attr, x.external_attr = centdir[15:18] |
| 1351 | # Convert date/time code to (year, month, day, hour, min, sec) |
Christian Heimes | fdab48e | 2008-01-20 09:06:41 +0000 | [diff] [blame] | 1352 | x._raw_time = t |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1353 | x.date_time = ( (d>>9)+1980, (d>>5)&0xF, d&0x1F, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1354 | t>>11, (t>>5)&0x3F, (t&0x1F) * 2 ) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1355 | |
| 1356 | x._decodeExtra() |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1357 | x.header_offset = x.header_offset + concat |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1358 | self.filelist.append(x) |
| 1359 | self.NameToInfo[x.filename] = x |
Martin v. Löwis | b09b844 | 2008-07-03 14:13:42 +0000 | [diff] [blame] | 1360 | |
| 1361 | # update total bytes read from central directory |
| 1362 | total = (total + sizeCentralDir + centdir[_CD_FILENAME_LENGTH] |
| 1363 | + centdir[_CD_EXTRA_FIELD_LENGTH] |
| 1364 | + centdir[_CD_COMMENT_LENGTH]) |
| 1365 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1366 | if self.debug > 2: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1367 | print("total", total) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1368 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1369 | |
| 1370 | def namelist(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1371 | """Return a list of file names in the archive.""" |
Ezio Melotti | 006917e | 2012-04-16 21:34:24 -0600 | [diff] [blame] | 1372 | return [data.filename for data in self.filelist] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1373 | |
| 1374 | def infolist(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1375 | """Return a list of class ZipInfo instances for files in the |
| 1376 | archive.""" |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1377 | return self.filelist |
| 1378 | |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1379 | def printdir(self, file=None): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1380 | """Print a table of contents for the zip file.""" |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1381 | print("%-46s %19s %12s" % ("File Name", "Modified ", "Size"), |
| 1382 | file=file) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1383 | for zinfo in self.filelist: |
Guido van Rossum | 7736b5b | 2008-01-15 21:44:53 +0000 | [diff] [blame] | 1384 | date = "%d-%02d-%02d %02d:%02d:%02d" % zinfo.date_time[:6] |
Guido van Rossum | d6ca546 | 2007-05-22 01:29:33 +0000 | [diff] [blame] | 1385 | print("%-46s %s %12d" % (zinfo.filename, date, zinfo.file_size), |
| 1386 | file=file) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1387 | |
| 1388 | def testzip(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1389 | """Read all the files and check the CRC.""" |
Benjamin Peterson | 4cd6a95 | 2008-08-17 20:23:46 +0000 | [diff] [blame] | 1390 | chunk_size = 2 ** 20 |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1391 | for zinfo in self.filelist: |
| 1392 | try: |
Benjamin Peterson | 4cd6a95 | 2008-08-17 20:23:46 +0000 | [diff] [blame] | 1393 | # Read by chunks, to avoid an OverflowError or a |
| 1394 | # MemoryError with very large embedded files. |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1395 | with self.open(zinfo.filename, "r") as f: |
| 1396 | while f.read(chunk_size): # Check CRC-32 |
| 1397 | pass |
Georg Brandl | 4d54088 | 2010-10-28 06:42:33 +0000 | [diff] [blame] | 1398 | except BadZipFile: |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1399 | return zinfo.filename |
| 1400 | |
| 1401 | def getinfo(self, name): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1402 | """Return the instance of ZipInfo given 'name'.""" |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1403 | info = self.NameToInfo.get(name) |
| 1404 | if info is None: |
| 1405 | raise KeyError( |
| 1406 | 'There is no item named %r in the archive' % name) |
| 1407 | |
| 1408 | return info |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1409 | |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1410 | def setpassword(self, pwd): |
| 1411 | """Set default password for encrypted files.""" |
R. David Murray | 8d855d8 | 2010-12-21 21:53:37 +0000 | [diff] [blame] | 1412 | if pwd and not isinstance(pwd, bytes): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1413 | raise TypeError("pwd: expected bytes, got %s" % type(pwd).__name__) |
R. David Murray | 8d855d8 | 2010-12-21 21:53:37 +0000 | [diff] [blame] | 1414 | if pwd: |
| 1415 | self.pwd = pwd |
| 1416 | else: |
| 1417 | self.pwd = None |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1418 | |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1419 | @property |
| 1420 | def comment(self): |
| 1421 | """The comment text associated with the ZIP file.""" |
| 1422 | return self._comment |
| 1423 | |
| 1424 | @comment.setter |
| 1425 | def comment(self, comment): |
| 1426 | if not isinstance(comment, bytes): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1427 | raise TypeError("comment: expected bytes, got %s" % type(comment).__name__) |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1428 | # check for valid comment length |
Serhiy Storchaka | 9b7a1a1 | 2014-01-20 21:57:40 +0200 | [diff] [blame] | 1429 | if len(comment) > ZIP_MAX_COMMENT: |
| 1430 | import warnings |
| 1431 | warnings.warn('Archive comment is too long; truncating to %d bytes' |
| 1432 | % ZIP_MAX_COMMENT, stacklevel=2) |
R David Murray | f50b38a | 2012-04-12 18:44:58 -0400 | [diff] [blame] | 1433 | comment = comment[:ZIP_MAX_COMMENT] |
| 1434 | self._comment = comment |
| 1435 | self._didModify = True |
| 1436 | |
Thomas Wouters | cf297e4 | 2007-02-23 15:07:44 +0000 | [diff] [blame] | 1437 | def read(self, name, pwd=None): |
Serhiy Storchaka | 4bb186d | 2018-11-25 09:51:14 +0200 | [diff] [blame] | 1438 | """Return file bytes for name.""" |
Benjamin Peterson | d285bdb | 2010-10-31 17:57:22 +0000 | [diff] [blame] | 1439 | with self.open(name, "r", pwd) as fp: |
| 1440 | return fp.read() |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1441 | |
Serhiy Storchaka | f47fc55 | 2016-05-15 12:27:16 +0300 | [diff] [blame] | 1442 | def open(self, name, mode="r", pwd=None, *, force_zip64=False): |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1443 | """Return file-like object for 'name'. |
| 1444 | |
| 1445 | name is a string for the file name within the ZIP file, or a ZipInfo |
| 1446 | object. |
| 1447 | |
| 1448 | mode should be 'r' to read a file already in the ZIP file, or 'w' to |
| 1449 | write to a file newly added to the archive. |
| 1450 | |
| 1451 | pwd is the password to decrypt files (only used for reading). |
| 1452 | |
| 1453 | When writing, if the file size is not known in advance but may exceed |
| 1454 | 2 GiB, pass force_zip64 to use the ZIP64 format, which can handle large |
| 1455 | files. If the size is known in advance, it is best to pass a ZipInfo |
| 1456 | instance for name, with zinfo.file_size set. |
| 1457 | """ |
Serhiy Storchaka | e670be2 | 2016-06-11 19:32:44 +0300 | [diff] [blame] | 1458 | if mode not in {"r", "w"}: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1459 | raise ValueError('open() requires mode "r" or "w"') |
R. David Murray | 8d855d8 | 2010-12-21 21:53:37 +0000 | [diff] [blame] | 1460 | if pwd and not isinstance(pwd, bytes): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1461 | raise TypeError("pwd: expected bytes, got %s" % type(pwd).__name__) |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1462 | if pwd and (mode == "w"): |
| 1463 | raise ValueError("pwd is only supported for reading files") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1464 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1465 | raise ValueError( |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1466 | "Attempt to use ZIP archive that was already closed") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1467 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1468 | # Make sure we have an info object |
| 1469 | if isinstance(name, ZipInfo): |
| 1470 | # 'name' is already an info object |
| 1471 | zinfo = name |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1472 | elif mode == 'w': |
| 1473 | zinfo = ZipInfo(name) |
| 1474 | zinfo.compress_type = self.compression |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1475 | zinfo._compresslevel = self.compresslevel |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1476 | else: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1477 | # Get info object for name |
| 1478 | zinfo = self.getinfo(name) |
Guido van Rossum | d8faa36 | 2007-04-27 19:54:29 +0000 | [diff] [blame] | 1479 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1480 | if mode == 'w': |
| 1481 | return self._open_to_write(zinfo, force_zip64=force_zip64) |
| 1482 | |
| 1483 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1484 | raise ValueError("Can't read from the ZIP file while there " |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1485 | "is an open writing handle on it. " |
| 1486 | "Close the writing handle before trying to read.") |
| 1487 | |
| 1488 | # Open for reading: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1489 | self._fileRefCnt += 1 |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1490 | zef_file = _SharedFile(self.fp, zinfo.header_offset, |
| 1491 | self._fpclose, self._lock, lambda: self._writing) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1492 | try: |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1493 | # Skip the file header: |
| 1494 | fheader = zef_file.read(sizeFileHeader) |
Serhiy Storchaka | d2b1527 | 2013-01-31 15:27:07 +0200 | [diff] [blame] | 1495 | if len(fheader) != sizeFileHeader: |
| 1496 | raise BadZipFile("Truncated file header") |
| 1497 | fheader = struct.unpack(structFileHeader, fheader) |
| 1498 | if fheader[_FH_SIGNATURE] != stringFileHeader: |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1499 | raise BadZipFile("Bad magic number for file header") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1500 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1501 | fname = zef_file.read(fheader[_FH_FILENAME_LENGTH]) |
| 1502 | if fheader[_FH_EXTRA_FIELD_LENGTH]: |
| 1503 | zef_file.read(fheader[_FH_EXTRA_FIELD_LENGTH]) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1504 | |
Antoine Pitrou | 8572da5 | 2012-11-17 23:52:05 +0100 | [diff] [blame] | 1505 | if zinfo.flag_bits & 0x20: |
| 1506 | # Zip 2.7: compressed patched data |
| 1507 | raise NotImplementedError("compressed patched data (flag bit 5)") |
Martin v. Löwis | 2a2ce32 | 2012-05-01 08:44:08 +0200 | [diff] [blame] | 1508 | |
Antoine Pitrou | 8572da5 | 2012-11-17 23:52:05 +0100 | [diff] [blame] | 1509 | if zinfo.flag_bits & 0x40: |
| 1510 | # strong encryption |
| 1511 | raise NotImplementedError("strong encryption (flag bit 6)") |
Martin v. Löwis | 7fb79fc | 2012-05-13 10:06:36 +0200 | [diff] [blame] | 1512 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1513 | if zinfo.flag_bits & 0x800: |
| 1514 | # UTF-8 filename |
| 1515 | fname_str = fname.decode("utf-8") |
| 1516 | else: |
| 1517 | fname_str = fname.decode("cp437") |
Georg Brandl | 5ba11de | 2011-01-01 10:09:32 +0000 | [diff] [blame] | 1518 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1519 | if fname_str != zinfo.orig_filename: |
| 1520 | raise BadZipFile( |
| 1521 | 'File name in directory %r and header %r differ.' |
| 1522 | % (zinfo.orig_filename, fname)) |
| 1523 | |
| 1524 | # check for encrypted flag & handle password |
| 1525 | is_encrypted = zinfo.flag_bits & 0x1 |
| 1526 | zd = None |
| 1527 | if is_encrypted: |
| 1528 | if not pwd: |
| 1529 | pwd = self.pwd |
| 1530 | if not pwd: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1531 | raise RuntimeError("File %r is encrypted, password " |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1532 | "required for extraction" % name) |
| 1533 | |
| 1534 | zd = _ZipDecrypter(pwd) |
| 1535 | # The first 12 bytes in the cypher stream is an encryption header |
| 1536 | # used to strengthen the algorithm. The first 11 bytes are |
| 1537 | # completely random, while the 12th contains the MSB of the CRC, |
| 1538 | # or the MSB of the file time depending on the header type |
| 1539 | # and is used to check the correctness of the password. |
| 1540 | header = zef_file.read(12) |
Serhiy Storchaka | 06e5225 | 2017-03-30 19:09:08 +0300 | [diff] [blame] | 1541 | h = zd(header[0:12]) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1542 | if zinfo.flag_bits & 0x8: |
| 1543 | # compare against the file type from extended local headers |
| 1544 | check_byte = (zinfo._raw_time >> 8) & 0xff |
| 1545 | else: |
| 1546 | # compare against the CRC otherwise |
| 1547 | check_byte = (zinfo.CRC >> 24) & 0xff |
| 1548 | if h[11] != check_byte: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1549 | raise RuntimeError("Bad password for file %r" % name) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1550 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1551 | return ZipExtFile(zef_file, mode, zinfo, zd, True) |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1552 | except: |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1553 | zef_file.close() |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1554 | raise |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1555 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1556 | def _open_to_write(self, zinfo, force_zip64=False): |
| 1557 | if force_zip64 and not self._allowZip64: |
| 1558 | raise ValueError( |
| 1559 | "force_zip64 is True, but allowZip64 was False when opening " |
| 1560 | "the ZIP file." |
| 1561 | ) |
| 1562 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1563 | raise ValueError("Can't write to the ZIP file while there is " |
| 1564 | "another write handle open on it. " |
| 1565 | "Close the first handle before opening another.") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1566 | |
| 1567 | # Sizes and CRC are overwritten with correct data after processing the file |
| 1568 | if not hasattr(zinfo, 'file_size'): |
| 1569 | zinfo.file_size = 0 |
| 1570 | zinfo.compress_size = 0 |
| 1571 | zinfo.CRC = 0 |
| 1572 | |
| 1573 | zinfo.flag_bits = 0x00 |
| 1574 | if zinfo.compress_type == ZIP_LZMA: |
| 1575 | # Compressed data includes an end-of-stream (EOS) marker |
| 1576 | zinfo.flag_bits |= 0x02 |
| 1577 | if not self._seekable: |
| 1578 | zinfo.flag_bits |= 0x08 |
| 1579 | |
| 1580 | if not zinfo.external_attr: |
| 1581 | zinfo.external_attr = 0o600 << 16 # permissions: ?rw------- |
| 1582 | |
| 1583 | # Compressed size can be larger than uncompressed size |
| 1584 | zip64 = self._allowZip64 and \ |
| 1585 | (force_zip64 or zinfo.file_size * 1.05 > ZIP64_LIMIT) |
| 1586 | |
| 1587 | if self._seekable: |
| 1588 | self.fp.seek(self.start_dir) |
| 1589 | zinfo.header_offset = self.fp.tell() |
| 1590 | |
| 1591 | self._writecheck(zinfo) |
| 1592 | self._didModify = True |
| 1593 | |
| 1594 | self.fp.write(zinfo.FileHeader(zip64)) |
| 1595 | |
| 1596 | self._writing = True |
| 1597 | return _ZipWriteFile(self, zinfo, zip64) |
| 1598 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1599 | def extract(self, member, path=None, pwd=None): |
| 1600 | """Extract a member from the archive to the current working directory, |
| 1601 | using its full name. Its file information is extracted as accurately |
| 1602 | as possible. `member' may be a filename or a ZipInfo object. You can |
| 1603 | specify a different directory using `path'. |
| 1604 | """ |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1605 | if path is None: |
| 1606 | path = os.getcwd() |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 1607 | else: |
| 1608 | path = os.fspath(path) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1609 | |
| 1610 | return self._extract_member(member, path, pwd) |
| 1611 | |
| 1612 | def extractall(self, path=None, members=None, pwd=None): |
| 1613 | """Extract all members from the archive to the current working |
| 1614 | directory. `path' specifies a different directory to extract to. |
| 1615 | `members' is optional and must be a subset of the list returned |
| 1616 | by namelist(). |
| 1617 | """ |
| 1618 | if members is None: |
| 1619 | members = self.namelist() |
| 1620 | |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 1621 | if path is None: |
| 1622 | path = os.getcwd() |
| 1623 | else: |
| 1624 | path = os.fspath(path) |
| 1625 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1626 | for zipinfo in members: |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 1627 | self._extract_member(zipinfo, path, pwd) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1628 | |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1629 | @classmethod |
| 1630 | def _sanitize_windows_name(cls, arcname, pathsep): |
| 1631 | """Replace bad characters and remove trailing dots from parts.""" |
| 1632 | table = cls._windows_illegal_name_trans_table |
| 1633 | if not table: |
| 1634 | illegal = ':<>|"?*' |
| 1635 | table = str.maketrans(illegal, '_' * len(illegal)) |
| 1636 | cls._windows_illegal_name_trans_table = table |
| 1637 | arcname = arcname.translate(table) |
| 1638 | # remove trailing dots |
| 1639 | arcname = (x.rstrip('.') for x in arcname.split(pathsep)) |
| 1640 | # rejoin, removing empty parts. |
| 1641 | arcname = pathsep.join(x for x in arcname if x) |
| 1642 | return arcname |
| 1643 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1644 | def _extract_member(self, member, targetpath, pwd): |
| 1645 | """Extract the ZipInfo object 'member' to a physical |
| 1646 | file on the path targetpath. |
| 1647 | """ |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 1648 | if not isinstance(member, ZipInfo): |
| 1649 | member = self.getinfo(member) |
| 1650 | |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1651 | # build the destination pathname, replacing |
| 1652 | # forward slashes to platform specific separators. |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1653 | arcname = member.filename.replace('/', os.path.sep) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1654 | |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1655 | if os.path.altsep: |
| 1656 | arcname = arcname.replace(os.path.altsep, os.path.sep) |
| 1657 | # interpret absolute pathname as relative, remove drive letter or |
| 1658 | # UNC path, redundant separators, "." and ".." components. |
| 1659 | arcname = os.path.splitdrive(arcname)[1] |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1660 | invalid_path_parts = ('', os.path.curdir, os.path.pardir) |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1661 | 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] | 1662 | if x not in invalid_path_parts) |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1663 | if os.path.sep == '\\': |
Serhiy Storchaka | e5e6444 | 2013-02-02 19:50:59 +0200 | [diff] [blame] | 1664 | # filter illegal characters on Windows |
Gregory P. Smith | 09aa752 | 2013-02-03 00:36:32 -0800 | [diff] [blame] | 1665 | arcname = self._sanitize_windows_name(arcname, os.path.sep) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1666 | |
Gregory P. Smith | b47acbf | 2013-02-01 11:22:43 -0800 | [diff] [blame] | 1667 | targetpath = os.path.join(targetpath, arcname) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1668 | targetpath = os.path.normpath(targetpath) |
| 1669 | |
| 1670 | # Create all upper directories if necessary. |
| 1671 | upperdirs = os.path.dirname(targetpath) |
| 1672 | if upperdirs and not os.path.exists(upperdirs): |
| 1673 | os.makedirs(upperdirs) |
| 1674 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 1675 | if member.is_dir(): |
Martin v. Löwis | 70ccd16 | 2009-05-24 19:47:22 +0000 | [diff] [blame] | 1676 | if not os.path.isdir(targetpath): |
| 1677 | os.mkdir(targetpath) |
Martin v. Löwis | 59e4779 | 2009-01-24 14:10:07 +0000 | [diff] [blame] | 1678 | return targetpath |
| 1679 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1680 | with self.open(member, pwd=pwd) as source, \ |
| 1681 | open(targetpath, "wb") as target: |
| 1682 | shutil.copyfileobj(source, target) |
Christian Heimes | 790c823 | 2008-01-07 21:14:23 +0000 | [diff] [blame] | 1683 | |
| 1684 | return targetpath |
| 1685 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1686 | def _writecheck(self, zinfo): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1687 | """Check for errors before writing a file to the archive.""" |
Raymond Hettinger | 54f0222 | 2002-06-01 14:18:47 +0000 | [diff] [blame] | 1688 | if zinfo.filename in self.NameToInfo: |
Serhiy Storchaka | 9b7a1a1 | 2014-01-20 21:57:40 +0200 | [diff] [blame] | 1689 | import warnings |
| 1690 | warnings.warn('Duplicate name: %r' % zinfo.filename, stacklevel=3) |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1691 | if self.mode not in ('w', 'x', 'a'): |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1692 | raise ValueError("write() requires mode 'w', 'x', or 'a'") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1693 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1694 | raise ValueError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1695 | "Attempt to write ZIP archive that was already closed") |
Martin v. Löwis | f6b16a4 | 2012-05-01 07:58:44 +0200 | [diff] [blame] | 1696 | _check_compression(zinfo.compress_type) |
Serhiy Storchaka | cfbb394 | 2014-09-23 21:34:24 +0300 | [diff] [blame] | 1697 | if not self._allowZip64: |
| 1698 | requires_zip64 = None |
| 1699 | if len(self.filelist) >= ZIP_FILECOUNT_LIMIT: |
| 1700 | requires_zip64 = "Files count" |
| 1701 | elif zinfo.file_size > ZIP64_LIMIT: |
| 1702 | requires_zip64 = "Filesize" |
| 1703 | elif zinfo.header_offset > ZIP64_LIMIT: |
| 1704 | requires_zip64 = "Zipfile size" |
| 1705 | if requires_zip64: |
| 1706 | raise LargeZipFile(requires_zip64 + |
| 1707 | " would require ZIP64 extensions") |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1708 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1709 | def write(self, filename, arcname=None, |
Marcel Plch | 77b112c | 2018-08-31 16:43:31 +0200 | [diff] [blame] | 1710 | compress_type=None, compresslevel=None): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1711 | """Put the bytes from filename into the archive under the name |
| 1712 | arcname.""" |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1713 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1714 | raise ValueError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1715 | "Attempt to write to ZIP archive that was already closed") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1716 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1717 | raise ValueError( |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1718 | "Can't write to ZIP archive while an open writing handle exists" |
| 1719 | ) |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1720 | |
Marcel Plch | a2fe1e5 | 2018-08-02 15:04:52 +0200 | [diff] [blame] | 1721 | zinfo = ZipInfo.from_file(filename, arcname, |
Marcel Plch | 77b112c | 2018-08-31 16:43:31 +0200 | [diff] [blame] | 1722 | strict_timestamps=self._strict_timestamps) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1723 | |
Serhiy Storchaka | 503f908 | 2016-02-08 00:02:25 +0200 | [diff] [blame] | 1724 | if zinfo.is_dir(): |
| 1725 | zinfo.compress_size = 0 |
| 1726 | zinfo.CRC = 0 |
| 1727 | else: |
| 1728 | if compress_type is not None: |
| 1729 | zinfo.compress_type = compress_type |
| 1730 | else: |
| 1731 | zinfo.compress_type = self.compression |
| 1732 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1733 | if compresslevel is not None: |
| 1734 | zinfo._compresslevel = compresslevel |
| 1735 | else: |
| 1736 | zinfo._compresslevel = self.compresslevel |
| 1737 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1738 | if zinfo.is_dir(): |
| 1739 | with self._lock: |
| 1740 | if self._seekable: |
| 1741 | self.fp.seek(self.start_dir) |
| 1742 | zinfo.header_offset = self.fp.tell() # Start of header bytes |
| 1743 | if zinfo.compress_type == ZIP_LZMA: |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1744 | # Compressed data includes an end-of-stream (EOS) marker |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1745 | zinfo.flag_bits |= 0x02 |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1746 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1747 | self._writecheck(zinfo) |
| 1748 | self._didModify = True |
Martin v. Löwis | 59e4779 | 2009-01-24 14:10:07 +0000 | [diff] [blame] | 1749 | |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1750 | self.filelist.append(zinfo) |
| 1751 | self.NameToInfo[zinfo.filename] = zinfo |
| 1752 | self.fp.write(zinfo.FileHeader(False)) |
| 1753 | self.start_dir = self.fp.tell() |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1754 | else: |
| 1755 | with open(filename, "rb") as src, self.open(zinfo, 'w') as dest: |
| 1756 | shutil.copyfileobj(src, dest, 1024*8) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1757 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1758 | def writestr(self, zinfo_or_arcname, data, |
| 1759 | compress_type=None, compresslevel=None): |
Guido van Rossum | 85825dc | 2007-08-27 17:03:28 +0000 | [diff] [blame] | 1760 | """Write a file into the archive. The contents is 'data', which |
| 1761 | may be either a 'str' or a 'bytes' instance; if it is a 'str', |
| 1762 | it is encoded as UTF-8 first. |
| 1763 | 'zinfo_or_arcname' is either a ZipInfo instance or |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1764 | the name of the file in the archive.""" |
Guido van Rossum | 85825dc | 2007-08-27 17:03:28 +0000 | [diff] [blame] | 1765 | if isinstance(data, str): |
| 1766 | data = data.encode("utf-8") |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1767 | if not isinstance(zinfo_or_arcname, ZipInfo): |
| 1768 | zinfo = ZipInfo(filename=zinfo_or_arcname, |
Guido van Rossum | 7736b5b | 2008-01-15 21:44:53 +0000 | [diff] [blame] | 1769 | date_time=time.localtime(time.time())[:6]) |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1770 | zinfo.compress_type = self.compression |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1771 | zinfo._compresslevel = self.compresslevel |
Serhiy Storchaka | 46a3492 | 2014-09-23 22:40:23 +0300 | [diff] [blame] | 1772 | if zinfo.filename[-1] == '/': |
| 1773 | zinfo.external_attr = 0o40775 << 16 # drwxrwxr-x |
| 1774 | zinfo.external_attr |= 0x10 # MS-DOS directory flag |
| 1775 | else: |
| 1776 | zinfo.external_attr = 0o600 << 16 # ?rw------- |
Just van Rossum | b083cb3 | 2002-12-12 12:23:32 +0000 | [diff] [blame] | 1777 | else: |
| 1778 | zinfo = zinfo_or_arcname |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1779 | |
| 1780 | if not self.fp: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1781 | raise ValueError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1782 | "Attempt to write to ZIP archive that was already closed") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1783 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1784 | raise ValueError( |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1785 | "Can't write to ZIP archive while an open writing handle exists." |
| 1786 | ) |
| 1787 | |
| 1788 | if compress_type is not None: |
| 1789 | zinfo.compress_type = compress_type |
Guido van Rossum | b5a755e | 2007-07-18 18:15:48 +0000 | [diff] [blame] | 1790 | |
Bo Bayles | ce237c7 | 2018-01-29 23:54:07 -0600 | [diff] [blame] | 1791 | if compresslevel is not None: |
| 1792 | zinfo._compresslevel = compresslevel |
| 1793 | |
Guido van Rossum | 85825dc | 2007-08-27 17:03:28 +0000 | [diff] [blame] | 1794 | zinfo.file_size = len(data) # Uncompressed size |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1795 | with self._lock: |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1796 | with self.open(zinfo, mode='w') as dest: |
| 1797 | dest.write(data) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1798 | |
| 1799 | def __del__(self): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1800 | """Call the "close()" method in case the user forgot.""" |
Tim Peters | d15f8bb | 2001-11-28 23:16:40 +0000 | [diff] [blame] | 1801 | self.close() |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1802 | |
| 1803 | def close(self): |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1804 | """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] | 1805 | records.""" |
Tim Peters | d15f8bb | 2001-11-28 23:16:40 +0000 | [diff] [blame] | 1806 | if self.fp is None: |
| 1807 | return |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 1808 | |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1809 | if self._writing: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1810 | raise ValueError("Can't close the ZIP file while there is " |
| 1811 | "an open writing handle on it. " |
| 1812 | "Close the writing handle before closing the zip.") |
Serhiy Storchaka | 18ee29d | 2016-05-13 13:52:49 +0300 | [diff] [blame] | 1813 | |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1814 | try: |
Serhiy Storchaka | 764fc9b | 2015-03-25 10:09:41 +0200 | [diff] [blame] | 1815 | 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] | 1816 | with self._lock: |
Serhiy Storchaka | 77d8997 | 2015-03-23 01:09:35 +0200 | [diff] [blame] | 1817 | if self._seekable: |
Serhiy Storchaka | a14f7d2 | 2015-01-26 14:01:27 +0200 | [diff] [blame] | 1818 | self.fp.seek(self.start_dir) |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1819 | self._write_end_record() |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 1820 | finally: |
| 1821 | fp = self.fp |
| 1822 | self.fp = None |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1823 | self._fpclose(fp) |
| 1824 | |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1825 | def _write_end_record(self): |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1826 | for zinfo in self.filelist: # write central directory |
| 1827 | dt = zinfo.date_time |
| 1828 | dosdate = (dt[0] - 1980) << 9 | dt[1] << 5 | dt[2] |
| 1829 | dostime = dt[3] << 11 | dt[4] << 5 | (dt[5] // 2) |
| 1830 | extra = [] |
| 1831 | if zinfo.file_size > ZIP64_LIMIT \ |
| 1832 | or zinfo.compress_size > ZIP64_LIMIT: |
| 1833 | extra.append(zinfo.file_size) |
| 1834 | extra.append(zinfo.compress_size) |
| 1835 | file_size = 0xffffffff |
| 1836 | compress_size = 0xffffffff |
| 1837 | else: |
| 1838 | file_size = zinfo.file_size |
| 1839 | compress_size = zinfo.compress_size |
| 1840 | |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1841 | if zinfo.header_offset > ZIP64_LIMIT: |
| 1842 | extra.append(zinfo.header_offset) |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1843 | header_offset = 0xffffffff |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1844 | else: |
| 1845 | header_offset = zinfo.header_offset |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1846 | |
| 1847 | extra_data = zinfo.extra |
| 1848 | min_version = 0 |
| 1849 | if extra: |
| 1850 | # Append a ZIP64 field to the extra's |
Serhiy Storchaka | 9bdb7be | 2018-09-17 15:36:40 +0300 | [diff] [blame] | 1851 | extra_data = _strip_extra(extra_data, (1,)) |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1852 | extra_data = struct.pack( |
| 1853 | '<HH' + 'Q'*len(extra), |
| 1854 | 1, 8*len(extra), *extra) + extra_data |
| 1855 | |
| 1856 | min_version = ZIP64_VERSION |
| 1857 | |
| 1858 | if zinfo.compress_type == ZIP_BZIP2: |
| 1859 | min_version = max(BZIP2_VERSION, min_version) |
| 1860 | elif zinfo.compress_type == ZIP_LZMA: |
| 1861 | min_version = max(LZMA_VERSION, min_version) |
| 1862 | |
| 1863 | extract_version = max(min_version, zinfo.extract_version) |
| 1864 | create_version = max(min_version, zinfo.create_version) |
| 1865 | try: |
| 1866 | filename, flag_bits = zinfo._encodeFilenameFlags() |
| 1867 | centdir = struct.pack(structCentralDir, |
| 1868 | stringCentralDir, create_version, |
| 1869 | zinfo.create_system, extract_version, zinfo.reserved, |
| 1870 | flag_bits, zinfo.compress_type, dostime, dosdate, |
| 1871 | zinfo.CRC, compress_size, file_size, |
| 1872 | len(filename), len(extra_data), len(zinfo.comment), |
| 1873 | 0, zinfo.internal_attr, zinfo.external_attr, |
| 1874 | header_offset) |
| 1875 | except DeprecationWarning: |
| 1876 | print((structCentralDir, stringCentralDir, create_version, |
| 1877 | zinfo.create_system, extract_version, zinfo.reserved, |
| 1878 | zinfo.flag_bits, zinfo.compress_type, dostime, dosdate, |
| 1879 | zinfo.CRC, compress_size, file_size, |
| 1880 | len(zinfo.filename), len(extra_data), len(zinfo.comment), |
| 1881 | 0, zinfo.internal_attr, zinfo.external_attr, |
| 1882 | header_offset), file=sys.stderr) |
| 1883 | raise |
| 1884 | self.fp.write(centdir) |
| 1885 | self.fp.write(filename) |
| 1886 | self.fp.write(extra_data) |
| 1887 | self.fp.write(zinfo.comment) |
| 1888 | |
| 1889 | pos2 = self.fp.tell() |
| 1890 | # Write end-of-zip-archive record |
| 1891 | centDirCount = len(self.filelist) |
| 1892 | centDirSize = pos2 - self.start_dir |
Serhiy Storchaka | 3763ea8 | 2017-05-06 14:46:01 +0300 | [diff] [blame] | 1893 | centDirOffset = self.start_dir |
Serhiy Storchaka | f15e524 | 2015-01-26 13:53:38 +0200 | [diff] [blame] | 1894 | requires_zip64 = None |
| 1895 | if centDirCount > ZIP_FILECOUNT_LIMIT: |
| 1896 | requires_zip64 = "Files count" |
| 1897 | elif centDirOffset > ZIP64_LIMIT: |
| 1898 | requires_zip64 = "Central directory offset" |
| 1899 | elif centDirSize > ZIP64_LIMIT: |
| 1900 | requires_zip64 = "Central directory size" |
| 1901 | if requires_zip64: |
| 1902 | # Need to write the ZIP64 end-of-archive records |
| 1903 | if not self._allowZip64: |
| 1904 | raise LargeZipFile(requires_zip64 + |
| 1905 | " would require ZIP64 extensions") |
| 1906 | zip64endrec = struct.pack( |
| 1907 | structEndArchive64, stringEndArchive64, |
| 1908 | 44, 45, 45, 0, 0, centDirCount, centDirCount, |
| 1909 | centDirSize, centDirOffset) |
| 1910 | self.fp.write(zip64endrec) |
| 1911 | |
| 1912 | zip64locrec = struct.pack( |
| 1913 | structEndArchive64Locator, |
| 1914 | stringEndArchive64Locator, 0, pos2, 1) |
| 1915 | self.fp.write(zip64locrec) |
| 1916 | centDirCount = min(centDirCount, 0xFFFF) |
| 1917 | centDirSize = min(centDirSize, 0xFFFFFFFF) |
| 1918 | centDirOffset = min(centDirOffset, 0xFFFFFFFF) |
| 1919 | |
| 1920 | endrec = struct.pack(structEndArchive, stringEndArchive, |
| 1921 | 0, 0, centDirCount, centDirCount, |
| 1922 | centDirSize, centDirOffset, len(self._comment)) |
| 1923 | self.fp.write(endrec) |
| 1924 | self.fp.write(self._comment) |
| 1925 | self.fp.flush() |
| 1926 | |
Serhiy Storchaka | 1ad088f | 2014-12-03 09:11:57 +0200 | [diff] [blame] | 1927 | def _fpclose(self, fp): |
| 1928 | assert self._fileRefCnt > 0 |
| 1929 | self._fileRefCnt -= 1 |
| 1930 | if not self._fileRefCnt and not self._filePassed: |
| 1931 | fp.close() |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1932 | |
| 1933 | |
| 1934 | class PyZipFile(ZipFile): |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1935 | """Class to create ZIP archives with Python library files and packages.""" |
| 1936 | |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1937 | def __init__(self, file, mode="r", compression=ZIP_STORED, |
Serhiy Storchaka | 235c5e0 | 2013-11-23 15:55:38 +0200 | [diff] [blame] | 1938 | allowZip64=True, optimize=-1): |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 1939 | ZipFile.__init__(self, file, mode=mode, compression=compression, |
| 1940 | allowZip64=allowZip64) |
| 1941 | self._optimize = optimize |
| 1942 | |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1943 | def writepy(self, pathname, basename="", filterfunc=None): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1944 | """Add all files from "pathname" to the ZIP archive. |
| 1945 | |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1946 | If pathname is a package directory, search the directory and |
| 1947 | all package subdirectories recursively for all *.py and enter |
| 1948 | the modules into the archive. If pathname is a plain |
| 1949 | directory, listdir *.py and enter all modules. Else, pathname |
| 1950 | 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] | 1951 | archive. Added modules are always module.pyc. |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1952 | This method will compile the module.py into module.pyc if |
| 1953 | necessary. |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1954 | If filterfunc(pathname) is given, it is called with every argument. |
| 1955 | When it is False, the file or directory is skipped. |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 1956 | """ |
Serhiy Storchaka | 8606e95 | 2017-03-08 14:37:51 +0200 | [diff] [blame] | 1957 | pathname = os.fspath(pathname) |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1958 | if filterfunc and not filterfunc(pathname): |
| 1959 | if self.debug: |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1960 | label = 'path' if os.path.isdir(pathname) else 'file' |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1961 | print('%s %r skipped by filterfunc' % (label, pathname)) |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1962 | return |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1963 | dir, name = os.path.split(pathname) |
| 1964 | if os.path.isdir(pathname): |
| 1965 | initname = os.path.join(pathname, "__init__.py") |
| 1966 | if os.path.isfile(initname): |
| 1967 | # This is a package directory, add it |
| 1968 | if basename: |
| 1969 | basename = "%s/%s" % (basename, name) |
| 1970 | else: |
| 1971 | basename = name |
| 1972 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1973 | print("Adding package in", pathname, "as", basename) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1974 | fname, arcname = self._get_codename(initname[0:-3], basename) |
| 1975 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1976 | print("Adding", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1977 | self.write(fname, arcname) |
Bernhard M. Wiedemann | 8452104 | 2018-01-31 11:17:10 +0100 | [diff] [blame] | 1978 | dirlist = sorted(os.listdir(pathname)) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1979 | dirlist.remove("__init__.py") |
| 1980 | # Add all *.py files and package subdirectories |
| 1981 | for filename in dirlist: |
| 1982 | path = os.path.join(pathname, filename) |
| 1983 | root, ext = os.path.splitext(filename) |
| 1984 | if os.path.isdir(path): |
| 1985 | if os.path.isfile(os.path.join(path, "__init__.py")): |
| 1986 | # This is a package directory, add it |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1987 | self.writepy(path, basename, |
| 1988 | filterfunc=filterfunc) # Recursive call |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1989 | elif ext == ".py": |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1990 | if filterfunc and not filterfunc(path): |
| 1991 | if self.debug: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 1992 | print('file %r skipped by filterfunc' % path) |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 1993 | continue |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1994 | fname, arcname = self._get_codename(path[0:-3], |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 1995 | basename) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1996 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 1997 | print("Adding", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 1998 | self.write(fname, arcname) |
| 1999 | else: |
| 2000 | # This is NOT a package directory, add its files at top level |
| 2001 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 2002 | print("Adding files from directory", pathname) |
Bernhard M. Wiedemann | 8452104 | 2018-01-31 11:17:10 +0100 | [diff] [blame] | 2003 | for filename in sorted(os.listdir(pathname)): |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2004 | path = os.path.join(pathname, filename) |
| 2005 | root, ext = os.path.splitext(filename) |
| 2006 | if ext == ".py": |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 2007 | if filterfunc and not filterfunc(path): |
| 2008 | if self.debug: |
Serhiy Storchaka | b0d497c | 2016-09-10 21:28:07 +0300 | [diff] [blame] | 2009 | print('file %r skipped by filterfunc' % path) |
Christian Tismer | 410d931 | 2013-10-22 04:09:28 +0200 | [diff] [blame] | 2010 | continue |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2011 | fname, arcname = self._get_codename(path[0:-3], |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 2012 | basename) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2013 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 2014 | print("Adding", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2015 | self.write(fname, arcname) |
| 2016 | else: |
| 2017 | if pathname[-3:] != ".py": |
Collin Winter | ce36ad8 | 2007-08-30 01:19:48 +0000 | [diff] [blame] | 2018 | raise RuntimeError( |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 2019 | 'Files added with writepy() must end with ".py"') |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2020 | fname, arcname = self._get_codename(pathname[0:-3], basename) |
| 2021 | if self.debug: |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 2022 | print("Adding file", arcname) |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2023 | self.write(fname, arcname) |
| 2024 | |
| 2025 | def _get_codename(self, pathname, basename): |
| 2026 | """Return (filename, archivename) for the path. |
| 2027 | |
Fred Drake | 484d735 | 2000-10-02 21:14:52 +0000 | [diff] [blame] | 2028 | Given a module name path, return the correct file path and |
| 2029 | archive name, compiling if necessary. For example, given |
| 2030 | /python/lib/string, return (/python/lib/string.pyc, string). |
| 2031 | """ |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2032 | def _compile(file, optimize=-1): |
| 2033 | import py_compile |
| 2034 | if self.debug: |
| 2035 | print("Compiling", file) |
| 2036 | try: |
| 2037 | py_compile.compile(file, doraise=True, optimize=optimize) |
Serhiy Storchaka | 45c4375 | 2013-01-29 20:10:28 +0200 | [diff] [blame] | 2038 | except py_compile.PyCompileError as err: |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2039 | print(err.msg) |
| 2040 | return False |
| 2041 | return True |
| 2042 | |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2043 | file_py = pathname + ".py" |
| 2044 | file_pyc = pathname + ".pyc" |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2045 | pycache_opt0 = importlib.util.cache_from_source(file_py, optimization='') |
| 2046 | pycache_opt1 = importlib.util.cache_from_source(file_py, optimization=1) |
| 2047 | pycache_opt2 = importlib.util.cache_from_source(file_py, optimization=2) |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2048 | if self._optimize == -1: |
| 2049 | # legacy mode: use whatever file is present |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2050 | if (os.path.isfile(file_pyc) and |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2051 | os.stat(file_pyc).st_mtime >= os.stat(file_py).st_mtime): |
| 2052 | # Use .pyc file. |
| 2053 | arcname = fname = file_pyc |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2054 | elif (os.path.isfile(pycache_opt0) and |
| 2055 | os.stat(pycache_opt0).st_mtime >= os.stat(file_py).st_mtime): |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2056 | # Use the __pycache__/*.pyc file, but write it to the legacy pyc |
| 2057 | # file name in the archive. |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2058 | fname = pycache_opt0 |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2059 | arcname = file_pyc |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2060 | elif (os.path.isfile(pycache_opt1) and |
| 2061 | os.stat(pycache_opt1).st_mtime >= os.stat(file_py).st_mtime): |
| 2062 | # Use the __pycache__/*.pyc file, but write it to the legacy pyc |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2063 | # file name in the archive. |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2064 | fname = pycache_opt1 |
| 2065 | arcname = file_pyc |
| 2066 | elif (os.path.isfile(pycache_opt2) and |
| 2067 | os.stat(pycache_opt2).st_mtime >= os.stat(file_py).st_mtime): |
| 2068 | # Use the __pycache__/*.pyc file, but write it to the legacy pyc |
| 2069 | # file name in the archive. |
| 2070 | fname = pycache_opt2 |
| 2071 | arcname = file_pyc |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 2072 | else: |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2073 | # Compile py into PEP 3147 pyc file. |
| 2074 | if _compile(file_py): |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2075 | if sys.flags.optimize == 0: |
| 2076 | fname = pycache_opt0 |
| 2077 | elif sys.flags.optimize == 1: |
| 2078 | fname = pycache_opt1 |
| 2079 | else: |
| 2080 | fname = pycache_opt2 |
| 2081 | arcname = file_pyc |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2082 | else: |
| 2083 | fname = arcname = file_py |
| 2084 | else: |
| 2085 | # new mode: use given optimization level |
| 2086 | if self._optimize == 0: |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2087 | fname = pycache_opt0 |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2088 | arcname = file_pyc |
| 2089 | else: |
Brett Cannon | f299abd | 2015-04-13 14:21:02 -0400 | [diff] [blame] | 2090 | arcname = file_pyc |
| 2091 | if self._optimize == 1: |
| 2092 | fname = pycache_opt1 |
| 2093 | elif self._optimize == 2: |
| 2094 | fname = pycache_opt2 |
| 2095 | else: |
| 2096 | msg = "invalid value for 'optimize': {!r}".format(self._optimize) |
| 2097 | raise ValueError(msg) |
Georg Brandl | 8334fd9 | 2010-12-04 10:26:46 +0000 | [diff] [blame] | 2098 | if not (os.path.isfile(fname) and |
| 2099 | os.stat(fname).st_mtime >= os.stat(file_py).st_mtime): |
| 2100 | if not _compile(file_py, optimize=self._optimize): |
| 2101 | fname = arcname = file_py |
Barry Warsaw | 28a691b | 2010-04-17 00:19:56 +0000 | [diff] [blame] | 2102 | archivename = os.path.split(arcname)[1] |
Guido van Rossum | 32abe6f | 2000-03-31 17:30:02 +0000 | [diff] [blame] | 2103 | if basename: |
| 2104 | archivename = "%s/%s" % (basename, archivename) |
| 2105 | return (fname, archivename) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2106 | |
| 2107 | |
Miss Islington (bot) | c410f38 | 2019-08-24 09:03:52 -0700 | [diff] [blame^] | 2108 | def _unique_everseen(iterable, key=None): |
| 2109 | "List unique elements, preserving order. Remember all elements ever seen." |
| 2110 | # unique_everseen('AAAABBBCCDAABBB') --> A B C D |
| 2111 | # unique_everseen('ABBCcAD', str.lower) --> A B C D |
| 2112 | seen = set() |
| 2113 | seen_add = seen.add |
| 2114 | if key is None: |
| 2115 | for element in itertools.filterfalse(seen.__contains__, iterable): |
| 2116 | seen_add(element) |
| 2117 | yield element |
| 2118 | else: |
| 2119 | for element in iterable: |
| 2120 | k = key(element) |
| 2121 | if k not in seen: |
| 2122 | seen_add(k) |
| 2123 | yield element |
| 2124 | |
| 2125 | |
| 2126 | def _parents(path): |
| 2127 | """ |
| 2128 | Given a path with elements separated by |
| 2129 | posixpath.sep, generate all parents of that path. |
| 2130 | |
| 2131 | >>> list(_parents('b/d')) |
| 2132 | ['b'] |
| 2133 | >>> list(_parents('/b/d/')) |
| 2134 | ['/b'] |
| 2135 | >>> list(_parents('b/d/f/')) |
| 2136 | ['b/d', 'b'] |
| 2137 | >>> list(_parents('b')) |
| 2138 | [] |
| 2139 | >>> list(_parents('')) |
| 2140 | [] |
| 2141 | """ |
| 2142 | return itertools.islice(_ancestry(path), 1, None) |
| 2143 | |
| 2144 | |
| 2145 | def _ancestry(path): |
| 2146 | """ |
| 2147 | Given a path with elements separated by |
| 2148 | posixpath.sep, generate all elements of that path |
| 2149 | |
| 2150 | >>> list(_ancestry('b/d')) |
| 2151 | ['b/d', 'b'] |
| 2152 | >>> list(_ancestry('/b/d/')) |
| 2153 | ['/b/d', '/b'] |
| 2154 | >>> list(_ancestry('b/d/f/')) |
| 2155 | ['b/d/f', 'b/d', 'b'] |
| 2156 | >>> list(_ancestry('b')) |
| 2157 | ['b'] |
| 2158 | >>> list(_ancestry('')) |
| 2159 | [] |
| 2160 | """ |
| 2161 | path = path.rstrip(posixpath.sep) |
| 2162 | while path and path != posixpath.sep: |
| 2163 | yield path |
| 2164 | path, tail = posixpath.split(path) |
| 2165 | |
| 2166 | |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 2167 | class Path: |
| 2168 | """ |
| 2169 | A pathlib-compatible interface for zip files. |
| 2170 | |
| 2171 | Consider a zip file with this structure:: |
| 2172 | |
| 2173 | . |
| 2174 | ├── a.txt |
| 2175 | └── b |
| 2176 | ├── c.txt |
| 2177 | └── d |
| 2178 | └── e.txt |
| 2179 | |
| 2180 | >>> data = io.BytesIO() |
| 2181 | >>> zf = ZipFile(data, 'w') |
| 2182 | >>> zf.writestr('a.txt', 'content of a') |
| 2183 | >>> zf.writestr('b/c.txt', 'content of c') |
| 2184 | >>> zf.writestr('b/d/e.txt', 'content of e') |
| 2185 | >>> zf.filename = 'abcde.zip' |
| 2186 | |
| 2187 | Path accepts the zipfile object itself or a filename |
| 2188 | |
| 2189 | >>> root = Path(zf) |
| 2190 | |
| 2191 | From there, several path operations are available. |
| 2192 | |
| 2193 | Directory iteration (including the zip file itself): |
| 2194 | |
| 2195 | >>> a, b = root.iterdir() |
| 2196 | >>> a |
| 2197 | Path('abcde.zip', 'a.txt') |
| 2198 | >>> b |
| 2199 | Path('abcde.zip', 'b/') |
| 2200 | |
| 2201 | name property: |
| 2202 | |
| 2203 | >>> b.name |
| 2204 | 'b' |
| 2205 | |
| 2206 | join with divide operator: |
| 2207 | |
| 2208 | >>> c = b / 'c.txt' |
| 2209 | >>> c |
| 2210 | Path('abcde.zip', 'b/c.txt') |
| 2211 | >>> c.name |
| 2212 | 'c.txt' |
| 2213 | |
| 2214 | Read text: |
| 2215 | |
| 2216 | >>> c.read_text() |
| 2217 | 'content of c' |
| 2218 | |
| 2219 | existence: |
| 2220 | |
| 2221 | >>> c.exists() |
| 2222 | True |
| 2223 | >>> (b / 'missing.txt').exists() |
| 2224 | False |
| 2225 | |
Xtreak | 0d70227 | 2019-06-03 04:42:33 +0530 | [diff] [blame] | 2226 | Coercion to string: |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 2227 | |
| 2228 | >>> str(c) |
| 2229 | 'abcde.zip/b/c.txt' |
| 2230 | """ |
| 2231 | |
| 2232 | __repr = "{self.__class__.__name__}({self.root.filename!r}, {self.at!r})" |
| 2233 | |
| 2234 | def __init__(self, root, at=""): |
| 2235 | self.root = root if isinstance(root, ZipFile) else ZipFile(root) |
| 2236 | self.at = at |
| 2237 | |
| 2238 | @property |
| 2239 | def open(self): |
| 2240 | return functools.partial(self.root.open, self.at) |
| 2241 | |
| 2242 | @property |
| 2243 | def name(self): |
| 2244 | return posixpath.basename(self.at.rstrip("/")) |
| 2245 | |
| 2246 | def read_text(self, *args, **kwargs): |
| 2247 | with self.open() as strm: |
| 2248 | return io.TextIOWrapper(strm, *args, **kwargs).read() |
| 2249 | |
| 2250 | def read_bytes(self): |
| 2251 | with self.open() as strm: |
| 2252 | return strm.read() |
| 2253 | |
| 2254 | def _is_child(self, path): |
| 2255 | return posixpath.dirname(path.at.rstrip("/")) == self.at.rstrip("/") |
| 2256 | |
| 2257 | def _next(self, at): |
| 2258 | return Path(self.root, at) |
| 2259 | |
| 2260 | def is_dir(self): |
| 2261 | return not self.at or self.at.endswith("/") |
| 2262 | |
| 2263 | def is_file(self): |
| 2264 | return not self.is_dir() |
| 2265 | |
| 2266 | def exists(self): |
| 2267 | return self.at in self._names() |
| 2268 | |
| 2269 | def iterdir(self): |
| 2270 | if not self.is_dir(): |
| 2271 | raise ValueError("Can't listdir a file") |
| 2272 | subs = map(self._next, self._names()) |
| 2273 | return filter(self._is_child, subs) |
| 2274 | |
| 2275 | def __str__(self): |
| 2276 | return posixpath.join(self.root.filename, self.at) |
| 2277 | |
| 2278 | def __repr__(self): |
| 2279 | return self.__repr.format(self=self) |
| 2280 | |
Jason R. Coombs | 33e067d | 2019-05-09 11:34:36 -0400 | [diff] [blame] | 2281 | def joinpath(self, add): |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 2282 | next = posixpath.join(self.at, add) |
| 2283 | next_dir = posixpath.join(self.at, add, "") |
| 2284 | names = self._names() |
| 2285 | return self._next(next_dir if next not in names and next_dir in names else next) |
| 2286 | |
Jason R. Coombs | 33e067d | 2019-05-09 11:34:36 -0400 | [diff] [blame] | 2287 | __truediv__ = joinpath |
| 2288 | |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 2289 | @staticmethod |
Miss Islington (bot) | c410f38 | 2019-08-24 09:03:52 -0700 | [diff] [blame^] | 2290 | def _implied_dirs(names): |
| 2291 | return _unique_everseen( |
| 2292 | parent + "/" |
| 2293 | for name in names |
| 2294 | for parent in _parents(name) |
| 2295 | if parent + "/" not in names |
| 2296 | ) |
| 2297 | |
| 2298 | @classmethod |
| 2299 | def _add_implied_dirs(cls, names): |
| 2300 | return names + list(cls._implied_dirs(names)) |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 2301 | |
Jason R. Coombs | 33e067d | 2019-05-09 11:34:36 -0400 | [diff] [blame] | 2302 | @property |
| 2303 | def parent(self): |
Miss Islington (bot) | 66905d1 | 2019-07-07 15:05:53 -0700 | [diff] [blame] | 2304 | parent_at = posixpath.dirname(self.at.rstrip('/')) |
Jason R. Coombs | 33e067d | 2019-05-09 11:34:36 -0400 | [diff] [blame] | 2305 | if parent_at: |
| 2306 | parent_at += '/' |
| 2307 | return self._next(parent_at) |
| 2308 | |
Jason R. Coombs | b2758ff | 2019-05-08 09:45:06 -0400 | [diff] [blame] | 2309 | def _names(self): |
| 2310 | return self._add_implied_dirs(self.root.namelist()) |
| 2311 | |
| 2312 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2313 | def main(args=None): |
| 2314 | import argparse |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2315 | |
Serhiy Storchaka | 150cd19 | 2017-04-07 18:56:12 +0300 | [diff] [blame] | 2316 | description = 'A simple command-line interface for zipfile module.' |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2317 | parser = argparse.ArgumentParser(description=description) |
Serhiy Storchaka | 150cd19 | 2017-04-07 18:56:12 +0300 | [diff] [blame] | 2318 | group = parser.add_mutually_exclusive_group(required=True) |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2319 | group.add_argument('-l', '--list', metavar='<zipfile>', |
| 2320 | help='Show listing of a zipfile') |
| 2321 | group.add_argument('-e', '--extract', nargs=2, |
| 2322 | metavar=('<zipfile>', '<output_dir>'), |
| 2323 | help='Extract zipfile into target dir') |
| 2324 | group.add_argument('-c', '--create', nargs='+', |
| 2325 | metavar=('<name>', '<file>'), |
| 2326 | help='Create zipfile from sources') |
| 2327 | group.add_argument('-t', '--test', metavar='<zipfile>', |
| 2328 | help='Test if a zipfile is valid') |
| 2329 | args = parser.parse_args(args) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2330 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2331 | if args.test is not None: |
| 2332 | src = args.test |
| 2333 | with ZipFile(src, 'r') as zf: |
Antoine Pitrou | 17babc5 | 2012-11-17 23:50:08 +0100 | [diff] [blame] | 2334 | badfile = zf.testzip() |
Antoine Pitrou | 7c8bcb6 | 2010-08-12 15:11:50 +0000 | [diff] [blame] | 2335 | if badfile: |
| 2336 | print("The following enclosed file is corrupted: {!r}".format(badfile)) |
Guido van Rossum | be19ed7 | 2007-02-09 05:37:30 +0000 | [diff] [blame] | 2337 | print("Done testing") |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2338 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2339 | elif args.list is not None: |
| 2340 | src = args.list |
| 2341 | with ZipFile(src, 'r') as zf: |
| 2342 | zf.printdir() |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2343 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2344 | elif args.extract is not None: |
| 2345 | src, curdir = args.extract |
| 2346 | with ZipFile(src, 'r') as zf: |
| 2347 | zf.extractall(curdir) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2348 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2349 | elif args.create is not None: |
| 2350 | zip_name = args.create.pop(0) |
| 2351 | files = args.create |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2352 | |
| 2353 | def addToZip(zf, path, zippath): |
| 2354 | if os.path.isfile(path): |
| 2355 | zf.write(path, zippath, ZIP_DEFLATED) |
| 2356 | elif os.path.isdir(path): |
Serhiy Storchaka | 518e71b | 2014-10-04 13:39:34 +0300 | [diff] [blame] | 2357 | if zippath: |
| 2358 | zf.write(path, zippath) |
Bernhard M. Wiedemann | 8452104 | 2018-01-31 11:17:10 +0100 | [diff] [blame] | 2359 | for nm in sorted(os.listdir(path)): |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2360 | addToZip(zf, |
Christian Tismer | 59202e5 | 2013-10-21 03:59:23 +0200 | [diff] [blame] | 2361 | os.path.join(path, nm), os.path.join(zippath, nm)) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2362 | # else: ignore |
| 2363 | |
Serhiy Storchaka | 8c93310 | 2016-10-23 13:32:12 +0300 | [diff] [blame] | 2364 | with ZipFile(zip_name, 'w') as zf: |
| 2365 | for path in files: |
Serhiy Storchaka | 518e71b | 2014-10-04 13:39:34 +0300 | [diff] [blame] | 2366 | zippath = os.path.basename(path) |
| 2367 | if not zippath: |
| 2368 | zippath = os.path.basename(os.path.dirname(path)) |
| 2369 | if zippath in ('', os.curdir, os.pardir): |
| 2370 | zippath = '' |
| 2371 | addToZip(zf, path, zippath) |
Thomas Wouters | 0e3f591 | 2006-08-11 14:57:12 +0000 | [diff] [blame] | 2372 | |
| 2373 | if __name__ == "__main__": |
| 2374 | main() |