Issue #1159051: GzipFile now raises EOFError when reading a corrupted file

with truncated header or footer.
Added tests for reading truncated gzip, bzip2, and lzma files.

3 files changed, 41 insertions, 0 deletions

diff --git a/Lib/test/test_bz2.py b/Lib/test/test_bz2.py
index 8703df7..7090cd6 100644
--- a/Lib/test/test_bz2.py
+++ b/Lib/test/test_bz2.py
@@ -569,6 +569,19 @@ class BZ2FileTest(BaseTest):
                 bz2f.seek(-150, 1)
                 self.assertEqual(bz2f.read(), self.TEXT[500-150:])
 
+    def test_read_truncated(self):
+        # Drop the eos_magic field (6 bytes) and CRC (4 bytes).
+        truncated = self.DATA[:-10]
+        with BZ2File(BytesIO(truncated)) as f:
+            self.assertRaises(EOFError, f.read)
+        with BZ2File(BytesIO(truncated)) as f:
+            self.assertEqual(f.read(len(self.TEXT)), self.TEXT)
+            self.assertRaises(EOFError, f.read, 1)
+        # Incomplete 4-byte file header, and block header of at least 146 bits.
+        for i in range(22):
+            with BZ2File(BytesIO(truncated[:i])) as f:
+                self.assertRaises(EOFError, f.read, 1)
+
 
 class BZ2CompressorTest(BaseTest):
     def testCompress(self):
diff --git a/Lib/test/test_gzip.py b/Lib/test/test_gzip.py
index af73953..ebd4c43 100644
--- a/Lib/test/test_gzip.py
+++ b/Lib/test/test_gzip.py
@@ -389,6 +389,20 @@ class TestGzip(BaseTest):
             datac = gzip.compress(data)
             self.assertEqual(gzip.decompress(datac), data)
 
+    def test_read_truncated(self):
+        data = data1*50
+        # Drop the CRC (4 bytes) and file size (4 bytes).
+        truncated = gzip.compress(data)[:-8]
+        with gzip.GzipFile(fileobj=io.BytesIO(truncated)) as f:
+            self.assertRaises(EOFError, f.read)
+        with gzip.GzipFile(fileobj=io.BytesIO(truncated)) as f:
+            self.assertEqual(f.read(len(data)), data)
+            self.assertRaises(EOFError, f.read, 1)
+        # Incomplete 10-byte header.
+        for i in range(2, 10):
+            with gzip.GzipFile(fileobj=io.BytesIO(truncated[:i])) as f:
+                self.assertRaises(EOFError, f.read, 1)
+
 
 class TestOpen(BaseTest):
     def test_binary_modes(self):
diff --git a/Lib/test/test_lzma.py b/Lib/test/test_lzma.py
index a13cf3b..4669ee2 100644
--- a/Lib/test/test_lzma.py
+++ b/Lib/test/test_lzma.py
@@ -669,6 +669,20 @@ class FileTestCase(unittest.TestCase):
         with LZMAFile(BytesIO(COMPRESSED_XZ[:128])) as f:
             self.assertRaises(EOFError, f.read)
 
+    def test_read_truncated(self):
+        # Drop stream footer: CRC (4 bytes), index size (4 bytes),
+        # flags (2 bytes) and magic number (2 bytes).
+        truncated = COMPRESSED_XZ[:-12]
+        with LZMAFile(BytesIO(truncated)) as f:
+            self.assertRaises(EOFError, f.read)
+        with LZMAFile(BytesIO(truncated)) as f:
+            self.assertEqual(f.read(len(INPUT)), INPUT)
+            self.assertRaises(EOFError, f.read, 1)
+        # Incomplete 12-byte header.
+        for i in range(12):
+            with LZMAFile(BytesIO(truncated[:i])) as f:
+                self.assertRaises(EOFError, f.read, 1)
+
     def test_read_bad_args(self):
         f = LZMAFile(BytesIO(COMPRESSED_XZ))
         f.close()