Merged revisions 83959-83960 via svnmerge from

svn+ssh://pythondev@svn.python.org/python/branches/py3k

........
  r83959 | antoine.pitrou | 2010-08-12 17:11:50 +0200 (jeu., 12 août 2010) | 5 lines

  Issue #7467: when a file from a ZIP archive, its CRC is checked and a
  BadZipfile error is raised if it doesn't match (as used to be the
  case in Python 2.5 and earlier).
........
  r83960 | antoine.pitrou | 2010-08-12 17:15:01 +0200 (jeu., 12 août 2010) | 3 lines

  Typo.
........

Lib/test/test_zipfile.py

@@ -5,6 +5,7 @@
     zlib = None
 
 import os
+import io
 import sys
 import time
 import shutil
@@ -653,6 +654,27 @@ def test_write_non_pyfile(self):
 
 
 class OtherTests(unittest.TestCase):
+    zips_with_bad_crc = {
+        zipfile.ZIP_STORED: (
+            b'PK\003\004\024\0\0\0\0\0 \213\212;:r'
+            b'\253\377\f\0\0\0\f\0\0\0\005\0\0\000af'
+            b'ilehello,AworldP'
+            b'K\001\002\024\003\024\0\0\0\0\0 \213\212;:'
+            b'r\253\377\f\0\0\0\f\0\0\0\005\0\0\0\0'
+            b'\0\0\0\0\0\0\0\200\001\0\0\0\000afi'
+            b'lePK\005\006\0\0\0\0\001\0\001\0003\000'
+            b'\0\0/\0\0\0\0\0'),
+        zipfile.ZIP_DEFLATED: (
+            b'PK\x03\x04\x14\x00\x00\x00\x08\x00n}\x0c=FA'
+            b'KE\x10\x00\x00\x00n\x00\x00\x00\x05\x00\x00\x00af'
+            b'ile\xcbH\xcd\xc9\xc9W(\xcf/\xcaI\xc9\xa0'
+            b'=\x13\x00PK\x01\x02\x14\x03\x14\x00\x00\x00\x08\x00n'
+            b'}\x0c=FAKE\x10\x00\x00\x00n\x00\x00\x00\x05'
+            b'\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x80\x01\x00\x00\x00'
+            b'\x00afilePK\x05\x06\x00\x00\x00\x00\x01\x00'
+            b'\x01\x003\x00\x00\x003\x00\x00\x00\x00\x00'),
+    }
+
     def test_unicode_filenames(self):
         with zipfile.ZipFile(TESTFN, "w") as zf:
             zf.writestr(u"foo.txt", "Test for unicode filename")
@@ -864,6 +886,49 @@ def test_comments(self):
         with zipfile.ZipFile(TESTFN, mode="r") as zipf:
             self.assertEqual(zipf.comment, comment2)
 
+    def check_testzip_with_bad_crc(self, compression):
+        """Tests that files with bad CRCs return their name from testzip."""
+        zipdata = self.zips_with_bad_crc[compression]
+
+        with zipfile.ZipFile(io.BytesIO(zipdata), mode="r") as zipf:
+            # testzip returns the name of the first corrupt file, or None
+            self.assertEqual('afile', zipf.testzip())
+
+    def test_testzip_with_bad_crc_stored(self):
+        self.check_testzip_with_bad_crc(zipfile.ZIP_STORED)
+
+    @skipUnless(zlib, "requires zlib")
+    def test_testzip_with_bad_crc_deflated(self):
+        self.check_testzip_with_bad_crc(zipfile.ZIP_DEFLATED)
+
+    def check_read_with_bad_crc(self, compression):
+        """Tests that files with bad CRCs raise a BadZipfile exception when read."""
+        zipdata = self.zips_with_bad_crc[compression]
+
+        # Using ZipFile.read()
+        with zipfile.ZipFile(io.BytesIO(zipdata), mode="r") as zipf:
+            self.assertRaises(zipfile.BadZipfile, zipf.read, 'afile')
+
+        # Using ZipExtFile.read()
+        with zipfile.ZipFile(io.BytesIO(zipdata), mode="r") as zipf:
+            with zipf.open('afile', 'r') as corrupt_file:
+                self.assertRaises(zipfile.BadZipfile, corrupt_file.read)
+
+        # Same with small reads (in order to exercise the buffering logic)
+        with zipfile.ZipFile(io.BytesIO(zipdata), mode="r") as zipf:
+            with zipf.open('afile', 'r') as corrupt_file:
+                corrupt_file.MIN_READ_SIZE = 2
+                with self.assertRaises(zipfile.BadZipfile):
+                    while corrupt_file.read(2):
+                        pass
+
+    def test_read_with_bad_crc_stored(self):
+        self.check_read_with_bad_crc(zipfile.ZIP_STORED)
+
+    @skipUnless(zlib, "requires zlib")
+    def test_read_with_bad_crc_deflated(self):
+        self.check_read_with_bad_crc(zipfile.ZIP_DEFLATED)
+
     def tearDown(self):
         unlink(TESTFN)
         unlink(TESTFN2)
@@ -963,6 +1028,11 @@ def test_stored(self):
         for f in (TESTFN2, TemporaryFile(), StringIO()):
             self.zip_test(f, zipfile.ZIP_STORED)
 
+    @skipUnless(zlib, "requires zlib")
+    def test_deflated(self):
+        for f in (TESTFN2, TemporaryFile(), io.BytesIO()):
+            self.zip_test(f, zipfile.ZIP_DEFLATED)
+
     def zip_open_test(self, f, compression):
         self.make_test_archive(f, compression)
 
@@ -996,6 +1066,11 @@ def test_open_stored(self):
         for f in (TESTFN2, TemporaryFile(), StringIO()):
             self.zip_open_test(f, zipfile.ZIP_STORED)
 
+    @skipUnless(zlib, "requires zlib")
+    def test_open_deflated(self):
+        for f in (TESTFN2, TemporaryFile(), io.BytesIO()):
+            self.zip_open_test(f, zipfile.ZIP_DEFLATED)
+
     def zip_random_open_test(self, f, compression):
         self.make_test_archive(f, compression)
 
@@ -1017,6 +1092,11 @@ def test_random_open_stored(self):
         for f in (TESTFN2, TemporaryFile(), StringIO()):
             self.zip_random_open_test(f, zipfile.ZIP_STORED)
 
+    @skipUnless(zlib, "requires zlib")
+    def test_random_open_deflated(self):
+        for f in (TESTFN2, TemporaryFile(), io.BytesIO()):
+            self.zip_random_open_test(f, zipfile.ZIP_DEFLATED)
+
 
 @skipUnless(zlib, "requires zlib")
 class TestsWithMultipleOpens(unittest.TestCase):

Lib/zipfile.py

@@ -493,6 +493,12 @@ def __init__(self, fileobj, mode, zipinfo, decrypter=None):
         self.mode = mode
         self.name = zipinfo.filename
 
+        if hasattr(zipinfo, 'CRC'):
+            self._expected_crc = zipinfo.CRC
+            self._running_crc = crc32(b'') & 0xffffffff
+        else:
+            self._expected_crc = None
+
     def readline(self, limit=-1):
         """Read and return a line from the stream.
 
@@ -570,6 +576,16 @@ def read(self, n=-1):
 
         return buf
 
+    def _update_crc(self, newdata, eof):
+        # Update the CRC using the given data.
+        if self._expected_crc is None:
+            # No need to compute the CRC if we don't have a reference value
+            return
+        self._running_crc = crc32(newdata, self._running_crc) & 0xffffffff
+        # Check the CRC if we're at the end of the file
+        if eof and self._running_crc != self._expected_crc:
+            raise BadZipfile("Bad CRC-32 for file %r" % self.name)
+
     def read1(self, n):
         """Read up to n bytes with at most one read() system call."""
 
@@ -593,6 +609,7 @@ def read1(self, n):
                 data = ''.join(map(self._decrypter, data))
 
             if self._compress_type == ZIP_STORED:
+                self._update_crc(data, eof=(self._compress_left==0))
                 self._readbuffer = self._readbuffer[self._offset:] + data
                 self._offset = 0
             else:
@@ -608,9 +625,11 @@ def read1(self, n):
             )
 
             self._unconsumed = self._decompressor.unconsumed_tail
-            if len(self._unconsumed) == 0 and self._compress_left == 0:
+            eof = len(self._unconsumed) == 0 and self._compress_left == 0
+            if eof:
                 data += self._decompressor.flush()
 
+            self._update_crc(data, eof=eof)
             self._readbuffer = self._readbuffer[self._offset:] + data
             self._offset = 0
 
@@ -1349,7 +1368,9 @@ def main(args = None):
             print USAGE
             sys.exit(1)
         zf = ZipFile(args[1], 'r')
-        zf.testzip()
+        badfile = zf.testzip()
+        if badfile:
+            print("The following enclosed file is corrupted: {!r}".format(badfile))
         print "Done testing"
 
     elif args[0] == '-e':

Misc/NEWS

@@ -29,6 +29,10 @@ Core and Builtins
 Library
 -------
 
+- Issue #7467: when reading a file from a ZIP archive, its CRC is checked
+  and a BadZipfile error is raised if it doesn't match (as used to be the
+  case in Python 2.5 and earlier).
+
 - Issue #9550: a BufferedReader could issue an additional read when the
   original read request had been satisfied, which could block indefinitely
   when the underlying raw IO channel was e.g. a socket.  Report and original