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

@@ -582,6 +582,27 @@ class PyZipFileTests(unittest.TestCase):
 
 
 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 testUnicodeFilenames(self):
         zf = zipfile.ZipFile(TESTFN, "w")
         zf.writestr("foo.txt", "Test for unicode filename")
@@ -809,6 +830,55 @@ class OtherTests(unittest.TestCase):
         self.assertEqual(zipfr.comment, comment2)
         zipfr.close()
 
+    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]
+
+        zipf = zipfile.ZipFile(io.BytesIO(zipdata), mode="r")
+        # testzip returns the name of the first corrupt file, or None
+        self.assertEqual('afile', zipf.testzip())
+        zipf.close()
+
+    def test_testzip_with_bad_crc_stored(self):
+        self.check_testzip_with_bad_crc(zipfile.ZIP_STORED)
+
+    if 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()
+        zipf = zipfile.ZipFile(io.BytesIO(zipdata), mode="r")
+        self.assertRaises(zipfile.BadZipfile, zipf.read, 'afile')
+        zipf.close()
+
+        # Using ZipExtFile.read()
+        zipf = zipfile.ZipFile(io.BytesIO(zipdata), mode="r")
+        corrupt_file = zipf.open('afile', 'r')
+        self.assertRaises(zipfile.BadZipfile, corrupt_file.read)
+        corrupt_file.close()
+        zipf.close()
+
+        # Same with small reads (in order to exercise the buffering logic)
+        zipf = zipfile.ZipFile(io.BytesIO(zipdata), mode="r")
+        corrupt_file = zipf.open('afile', 'r')
+        corrupt_file.MIN_READ_SIZE = 2
+        with self.assertRaises(zipfile.BadZipfile):
+            while corrupt_file.read(2):
+                pass
+        corrupt_file.close()
+        zipf.close()
+
+    def test_read_with_bad_crc_stored(self):
+        self.check_read_with_bad_crc(zipfile.ZIP_STORED)
+
+    if zlib:
+        def test_read_with_bad_crc_deflated(self):
+            self.check_read_with_bad_crc(zipfile.ZIP_DEFLATED)
+
     def tearDown(self):
         support.unlink(TESTFN)
         support.unlink(TESTFN2)

Lib/zipfile.py

@@ -472,6 +472,12 @@ class ZipExtFile:
         if self.compress_type == ZIP_DEFLATED:
             self.dc = zlib.decompressobj(-15)
 
+        if hasattr(zipinfo, 'CRC'):
+            self._expected_crc = zipinfo.CRC
+            self._running_crc = crc32(b'') & 0xffffffff
+        else:
+            self._expected_crc = None
+
     def set_univ_newlines(self, univ_newlines):
         self.univ_newlines = univ_newlines
 
@@ -565,6 +571,16 @@ class ZipExtFile:
             result.append(line)
         return result
 
+    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 read(self, size = None):
         # act like file obj and return empty string if size is 0
         if size == 0:
@@ -628,9 +644,11 @@ class ZipExtFile:
                         # prevent decompressor from being used again
                         self.dc = None
 
+                self._update_crc(newdata, eof=(
+                    self.compress_size == self.bytes_read and
+                    len(self.rawbuffer) == 0))
                 self.readbuffer += newdata
 
-
         # return what the user asked for
         if size is None or len(self.readbuffer) <= size:
             data = self.readbuffer
@@ -1382,7 +1400,9 @@ 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

@@ -93,6 +93,10 @@ C-API
 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