(original) (raw)
changeset: 81740:513b728695fc parent: 81736:2c0197c95ec6 parent: 81739:580d536cc910 user: Serhiy Storchaka storchaka@gmail.com date: Fri Jan 25 23:33:22 2013 +0200 files: Lib/test/test_codecs.py Misc/NEWS Objects/bytesobject.c description: Issue #16975: Fix error handling bug in the escape-decode bytes decoder. diff -r 2c0197c95ec6 -r 513b728695fc Lib/test/test_codecs.py --- a/Lib/test/test_codecs.py Fri Jan 25 13:57:16 2013 -0500 +++ b/Lib/test/test_codecs.py Fri Jan 25 23:33:22 2013 +0200 @@ -927,6 +927,50 @@ def test_empty(self): self.assertEqual(codecs.escape_decode(""), (b"", 0)) + def test_raw(self): + for b in range(256): + if b != b'\\'[0]: + self.assertEqual(codecs.escape_decode(bytes([b]) + b'0'), + (bytes([b]) + b'0', 2)) + + def test_escape(self): + self.assertEqual(codecs.escape_decode(b"[\\\n]"), (b"[]", 4)) + self.assertEqual(codecs.escape_decode(br'[\"]'), (b'["]', 4)) + self.assertEqual(codecs.escape_decode(br"[\']"), (b"[']", 4)) + self.assertEqual(codecs.escape_decode(br"[\\]"), (br"[\]", 4)) + self.assertEqual(codecs.escape_decode(br"[\a]"), (b"[\x07]", 4)) + self.assertEqual(codecs.escape_decode(br"[\b]"), (b"[\x08]", 4)) + self.assertEqual(codecs.escape_decode(br"[\t]"), (b"[\x09]", 4)) + self.assertEqual(codecs.escape_decode(br"[\n]"), (b"[\x0a]", 4)) + self.assertEqual(codecs.escape_decode(br"[\v]"), (b"[\x0b]", 4)) + self.assertEqual(codecs.escape_decode(br"[\f]"), (b"[\x0c]", 4)) + self.assertEqual(codecs.escape_decode(br"[\r]"), (b"[\x0d]", 4)) + self.assertEqual(codecs.escape_decode(br"[\7]"), (b"[\x07]", 4)) + self.assertEqual(codecs.escape_decode(br"[\8]"), (br"[\8]", 4)) + self.assertEqual(codecs.escape_decode(br"[\78]"), (b"[\x078]", 5)) + self.assertEqual(codecs.escape_decode(br"[\41]"), (b"[!]", 5)) + self.assertEqual(codecs.escape_decode(br"[\418]"), (b"[!8]", 6)) + self.assertEqual(codecs.escape_decode(br"[\101]"), (b"[A]", 6)) + self.assertEqual(codecs.escape_decode(br"[\1010]"), (b"[A0]", 7)) + self.assertEqual(codecs.escape_decode(br"[\501]"), (b"[A]", 6)) + self.assertEqual(codecs.escape_decode(br"[\x41]"), (b"[A]", 6)) + self.assertEqual(codecs.escape_decode(br"[\X41]"), (br"[\X41]", 6)) + self.assertEqual(codecs.escape_decode(br"[\x410]"), (b"[A0]", 7)) + for b in range(256): + if b not in b'\n"\'\\abtnvfr01234567x': + self.assertEqual(codecs.escape_decode(b'\\' + bytes([b])), + (b'\\' + bytes([b]), 2)) + + def test_errors(self): + self.assertRaises(ValueError, codecs.escape_decode, br"\x") + self.assertRaises(ValueError, codecs.escape_decode, br"[\x]") + self.assertEqual(codecs.escape_decode(br"[\x]\x", "ignore"), (b"[]", 6)) + self.assertEqual(codecs.escape_decode(br"[\x]\x", "replace"), (b"[?]?", 6)) + self.assertRaises(ValueError, codecs.escape_decode, br"\x0") + self.assertRaises(ValueError, codecs.escape_decode, br"[\x0]") + self.assertEqual(codecs.escape_decode(br"[\x0]\x0", "ignore"), (b"[]", 8)) + self.assertEqual(codecs.escape_decode(br"[\x0]\x0", "replace"), (b"[?]?", 8)) + class RecodingTest(unittest.TestCase): def test_recoding(self): f = io.BytesIO() diff -r 2c0197c95ec6 -r 513b728695fc Misc/NEWS --- a/Misc/NEWS Fri Jan 25 13:57:16 2013 -0500 +++ b/Misc/NEWS Fri Jan 25 23:33:22 2013 +0200 @@ -10,6 +10,8 @@ Core and Builtins ----------------- +- Issue #16975: Fix error handling bug in the escape-decode bytes decoder. + - Issue #14850: Now a charmap decoder treats U+FFFE as "undefined mapping" in any mapping, not only in a string. diff -r 2c0197c95ec6 -r 513b728695fc Objects/bytesobject.c --- a/Objects/bytesobject.c Fri Jan 25 13:57:16 2013 -0500 +++ b/Objects/bytesobject.c Fri Jan 25 23:33:22 2013 +0200 @@ -489,6 +489,10 @@ errors); goto failed; } + /* skip \x */ + if (s < end && Py_ISXDIGIT(s[0])) + s++; /* and a hexdigit */ + break; default: *p++ = '\\'; s--; /storchaka@gmail.com