[Python-checkins] cpython (2.7): Issue #21560: An attempt to write a data of wrong type no longer cause

[serhiy.storchaka]

https://hg.python.org/cpython/rev/4dfe0634d11a
changeset:   95141:4dfe0634d11a
branch:      2.7
parent:      95137:000814bbc899
user:        Serhiy Storchaka <storchaka at gmail.com>
date:        Mon Mar 23 15:25:18 2015 +0200
summary:
  Issue #21560: An attempt to write a data of wrong type no longer cause
GzipFile corruption.  Original patch by Wolfgang Maier.

files:
  Lib/gzip.py           |   4 ++--
  Lib/test/test_gzip.py |  27 +++++++++++++++++++++++++++
  Misc/NEWS             |   3 +++
  3 files changed, 32 insertions(+), 2 deletions(-)


diff --git a/Lib/gzip.py b/Lib/gzip.py
--- a/Lib/gzip.py
+++ b/Lib/gzip.py
@@ -238,9 +238,9 @@
             data = data.tobytes()
 
         if len(data) > 0:
-            self.size = self.size + len(data)
+            self.fileobj.write(self.compress.compress(data))
+            self.size += len(data)
             self.crc = zlib.crc32(data, self.crc) & 0xffffffffL
-            self.fileobj.write( self.compress.compress(data) )
             self.offset += len(data)
 
         return len(data)
diff --git a/Lib/test/test_gzip.py b/Lib/test/test_gzip.py
--- a/Lib/test/test_gzip.py
+++ b/Lib/test/test_gzip.py
@@ -30,6 +30,14 @@
     def tearDown(self):
         test_support.unlink(self.filename)
 
+    def write_and_read_back(self, data, mode='b'):
+        b_data = memoryview(data).tobytes()
+        with gzip.GzipFile(self.filename, 'w'+mode) as f:
+            l = f.write(data)
+        self.assertEqual(l, len(b_data))
+        with gzip.GzipFile(self.filename, 'r'+mode) as f:
+            self.assertEqual(f.read(), b_data)
+
     @test_support.requires_unicode
     def test_unicode_filename(self):
         unicode_filename = test_support.TESTFN_UNICODE
@@ -60,6 +68,25 @@
         # Test multiple close() calls.
         f.close()
 
+    # The following test_write_xy methods test that write accepts
+    # the corresponding bytes-like object type as input
+    # and that the data written equals bytes(xy) in all cases.
+    def test_write_memoryview(self):
+        self.write_and_read_back(memoryview(data1 * 50))
+
+    def test_write_incompatible_type(self):
+        # Test that non-bytes-like types raise TypeError.
+        # Issue #21560: attempts to write incompatible types
+        # should not affect the state of the fileobject
+        with gzip.GzipFile(self.filename, 'wb') as f:
+            with self.assertRaises(UnicodeEncodeError):
+                f.write(u'\xff')
+            with self.assertRaises(TypeError):
+                f.write([1])
+            f.write(data1)
+        with gzip.GzipFile(self.filename, 'rb') as f:
+            self.assertEqual(f.read(), data1)
+
     def test_read(self):
         self.test_write()
         # Try reading.
diff --git a/Misc/NEWS b/Misc/NEWS
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -21,6 +21,9 @@
 Library
 -------
 
+- Issue #21560: An attempt to write a data of wrong type no longer cause
+  GzipFile corruption.  Original patch by Wolfgang Maier.
+
 - Issue #23647: Increase impalib's MAXLINE to accommodate modern mailbox sizes.
 
 - Issue #23539: If body is None, http.client.HTTPConnection.request now sets

-- 
Repository URL: https://hg.python.org/cpython