Close #17666: Fix reading gzip files with an extra field.

serhiy-storchaka · serhiy-storchaka · commit ffcd339aac93 · 2013-04-08T22:37:15.000+03:00
diff --git a/Lib/gzip.py b/Lib/gzip.py
@@ -302,7 +302,8 @@ def _read_gzip_header(self):
 
         if flag & FEXTRA:
             # Read & discard the extra field, if present
-            self._read_exact(struct.unpack("<H", self._read_exact(2)))
+            extra_len, = struct.unpack("<H", self._read_exact(2))
+            self._read_exact(extra_len)
         if flag & FNAME:
             # Read and discard a null-terminated string containing the filename
             while True:
diff --git a/Lib/test/test_gzip.py b/Lib/test/test_gzip.py
@@ -403,6 +403,13 @@ def test_read_truncated(self):
             with gzip.GzipFile(fileobj=io.BytesIO(truncated[:i])) as f:
                 self.assertRaises(EOFError, f.read, 1)
 
+    def test_read_with_extra(self):
+        # Gzip data with an extra field
+        gzdata = (b'\x1f\x8b\x08\x04\xb2\x17cQ\x02\xff'
+                  b'\x05\x00Extra'
+                  b'\x0bI-.\x01\x002\xd1Mx\x04\x00\x00\x00')
+        with gzip.GzipFile(fileobj=io.BytesIO(gzdata)) as f:
+            self.assertEqual(f.read(), b'Test')
 
 class TestOpen(BaseTest):
     def test_binary_modes(self):
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -30,6 +30,8 @@ Core and Builtins
 Library
 -------
 
+- Issue #17666: Fix reading gzip files with an extra field.
+
 - Issue #16475: Support object instancing, recursion and interned strings
   in marshal
 
