From 61dce89fbd7759a9d2c19c46930030fd4467374f Mon Sep 17 00:00:00 2001
From: jvoisin <julien.voisin@dustri.org>
Date: Mon, 23 Nov 2020 19:50:46 +0100
Subject: [PATCH] Raise a ValueError explicitly

---
 libmat2/parser_factory.py     | 13 ++++++-------
 mat2                          | 12 ++++++++++--
 tests/test_corrupted_files.py | 26 ++++++++++++++------------
 3 files changed, 30 insertions(+), 21 deletions(-)

diff --git a/libmat2/parser_factory.py b/libmat2/parser_factory.py
index 23529db..842f0d7 100644
--- a/libmat2/parser_factory.py
+++ b/libmat2/parser_factory.py
@@ -40,7 +40,10 @@ def _get_parsers() -> List[T]:
 
 
 def get_parser(filename: str) -> Tuple[Optional[T], Optional[str]]:
-    """ Return the appropriate parser for a given filename. """
+    """ Return the appropriate parser for a given filename.
+
+        :raises ValueError: Raised if the instantiation of the parser went wrong.
+    """
     mtype, _ = mimetypes.guess_type(filename)
 
     _, extension = os.path.splitext(filename)
@@ -53,10 +56,6 @@ def get_parser(filename: str) -> Tuple[Optional[T], Optional[str]]:
 
     for parser_class in _get_parsers():  # type: ignore
         if mtype in parser_class.mimetypes:
-            try:
-                return parser_class(filename), mtype
-            except ValueError as e:
-                logging.info("Got an exception when trying to instantiate "
-                             "%s for %s: %s", parser_class, filename, e)
-                return None, mtype
+            # This instantiation might raise a ValueError on malformed files
+            return parser_class(filename), mtype
     return None, mtype
diff --git a/mat2 b/mat2
index 5134e77..f0d0539 100755
--- a/mat2
+++ b/mat2
@@ -85,7 +85,11 @@ def show_meta(filename: str, sandbox: bool):
     if not __check_file(filename):
         return
 
-    p, mtype = parser_factory.get_parser(filename)  # type: ignore
+    try:
+        p, mtype = parser_factory.get_parser(filename)  # type: ignore
+    except ValueError as e:
+        print("[-] something went wrong when processing %s: %s" % (filename, e))
+        return
     if p is None:
         print("[-] %s's format (%s) is not supported" % (filename, mtype))
         return
@@ -126,7 +130,11 @@ def clean_meta(filename: str, is_lightweight: bool, inplace: bool, sandbox: bool
     if not __check_file(filename, mode):
         return False
 
-    p, mtype = parser_factory.get_parser(filename)  # type: ignore
+    try:
+        p, mtype = parser_factory.get_parser(filename)  # type: ignore
+    except ValueError as e:
+        print("[-] something went wrong when cleaning %s: %s" % (filename, e))
+        return False
     if p is None:
         print("[-] %s's format (%s) is not supported" % (filename, mtype))
         return False
diff --git a/tests/test_corrupted_files.py b/tests/test_corrupted_files.py
index 8a8cffe..2adf42e 100644
--- a/tests/test_corrupted_files.py
+++ b/tests/test_corrupted_files.py
@@ -65,8 +65,10 @@ class TestCorruptedEmbedded(unittest.TestCase):
     def test_docx(self):
         shutil.copy('./tests/data/embedded_corrupted.docx', './tests/data/clean.docx')
         parser, _ = parser_factory.get_parser('./tests/data/clean.docx')
-        self.assertFalse(parser.remove_all())
-        self.assertIsNotNone(parser.get_meta())
+        with self.assertRaises(ValueError):
+            parser.remove_all()
+        with self.assertRaises(ValueError):
+            self.assertIsNotNone(parser.get_meta())
         os.remove('./tests/data/clean.docx')
 
     def test_odt(self):
@@ -120,8 +122,8 @@ class TestCorruptedFiles(unittest.TestCase):
 
     def test_png2(self):
         shutil.copy('./tests/test_libmat2.py', './tests/clean.png')
-        parser, _ = parser_factory.get_parser('./tests/clean.png')
-        self.assertIsNone(parser)
+        with self.assertRaises(ValueError):
+            parser_factory.get_parser('./tests/clean.png')
         os.remove('./tests/clean.png')
 
     def test_torrent(self):
@@ -237,10 +239,10 @@ class TestCorruptedFiles(unittest.TestCase):
             zout.write('./tests/data/embedded_corrupted.docx')
         p, mimetype = parser_factory.get_parser('./tests/data/clean.zip')
         self.assertEqual(mimetype, 'application/zip')
-        meta = p.get_meta()
-        self.assertEqual(meta['tests/data/dirty.flac']['comments'], 'Thank you for using MAT !')
-        self.assertEqual(meta['tests/data/dirty.docx']['word/media/image1.png']['Comment'], 'This is a comment, be careful!')
-        self.assertFalse(p.remove_all())
+        with self.assertRaises(ValueError):
+            p.get_meta()
+        with self.assertRaises(ValueError):
+            self.assertFalse(p.remove_all())
         os.remove('./tests/data/clean.zip')
 
     def test_html(self):
@@ -315,10 +317,10 @@ class TestCorruptedFiles(unittest.TestCase):
                 zout.addfile(tarinfo, f)
         p, mimetype = parser_factory.get_parser('./tests/data/clean.tar')
         self.assertEqual(mimetype, 'application/x-tar')
-        meta = p.get_meta()
-        self.assertEqual(meta['./tests/data/dirty.flac']['comments'], 'Thank you for using MAT !')
-        self.assertEqual(meta['./tests/data/dirty.docx']['word/media/image1.png']['Comment'], 'This is a comment, be careful!')
-        self.assertFalse(p.remove_all())
+        with self.assertRaises(ValueError):
+            p.get_meta()
+        with self.assertRaises(ValueError):
+            self.assertFalse(p.remove_all())
         os.remove('./tests/data/clean.tar')
 
         shutil.copy('./tests/data/dirty.png', './tests/data/clean.tar')