diff -r 668aba845fb2 Lib/test/test_coding.py --- a/Lib/test/test_coding.py Mon Jun 10 09:24:01 2013 -0700 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,65 +0,0 @@ -import test.support, unittest -from test.support import TESTFN, unlink, unload -import importlib, os, sys - -class CodingTest(unittest.TestCase): - def test_bad_coding(self): - module_name = 'bad_coding' - self.verify_bad_module(module_name) - - def test_bad_coding2(self): - module_name = 'bad_coding2' - self.verify_bad_module(module_name) - - def verify_bad_module(self, module_name): - self.assertRaises(SyntaxError, __import__, 'test.' + module_name) - - path = os.path.dirname(__file__) - filename = os.path.join(path, module_name + '.py') - with open(filename, "rb") as fp: - bytes = fp.read() - self.assertRaises(SyntaxError, compile, bytes, filename, 'exec') - - def test_exec_valid_coding(self): - d = {} - exec('# coding: cp949\na = 5\n', d) - self.assertEqual(d['a'], 5) - - def test_file_parse(self): - # issue1134: all encodings outside latin-1 and utf-8 fail on - # multiline strings and long lines (>512 columns) - unload(TESTFN) - sys.path.insert(0, os.curdir) - filename = TESTFN + ".py" - f = open(filename, "w") - try: - f.write("# -*- coding: cp1252 -*-\n") - f.write("'''A short string\n") - f.write("'''\n") - f.write("'A very long string %s'\n" % ("X" * 1000)) - f.close() - - importlib.invalidate_caches() - __import__(TESTFN) - finally: - f.close() - unlink(filename) - unlink(filename + "c") - unload(TESTFN) - del sys.path[0] - - def test_error_from_string(self): - # See http://bugs.python.org/issue6289 - input = "# coding: ascii\n\N{SNOWMAN}".encode('utf-8') - with self.assertRaises(SyntaxError) as c: - compile(input, "", "exec") - expected = "'ascii' codec can't decode byte 0xe2 in position 16: " \ - "ordinal not in range(128)" - self.assertTrue(c.exception.args[0].startswith(expected)) - - -def test_main(): - test.support.run_unittest(CodingTest) - -if __name__ == "__main__": - test_main() diff -r 668aba845fb2 Lib/test/test_pep263.py --- a/Lib/test/test_pep263.py Mon Jun 10 09:24:01 2013 -0700 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,80 +0,0 @@ -# -*- coding: koi8-r -*- - -import unittest -from test import support - -class PEP263Test(unittest.TestCase): - - def test_pep263(self): - self.assertEqual( - "ðÉÔÏÎ".encode("utf-8"), - b'\xd0\x9f\xd0\xb8\xd1\x82\xd0\xbe\xd0\xbd' - ) - self.assertEqual( - "\ð".encode("utf-8"), - b'\\\xd0\x9f' - ) - - def test_compilestring(self): - # see #1882 - c = compile(b"\n# coding: utf-8\nu = '\xc3\xb3'\n", "dummy", "exec") - d = {} - exec(c, d) - self.assertEqual(d['u'], '\xf3') - - def test_issue2301(self): - try: - compile(b"# coding: cp932\nprint '\x94\x4e'", "dummy", "exec") - except SyntaxError as v: - self.assertEqual(v.text, "print '\u5e74'\n") - else: - self.fail() - - def test_issue4626(self): - c = compile("# coding=latin-1\n\u00c6 = '\u00c6'", "dummy", "exec") - d = {} - exec(c, d) - self.assertEqual(d['\xc6'], '\xc6') - - def test_issue3297(self): - c = compile("a, b = '\U0001010F', '\\U0001010F'", "dummy", "exec") - d = {} - exec(c, d) - self.assertEqual(d['a'], d['b']) - self.assertEqual(len(d['a']), len(d['b'])) - self.assertEqual(ascii(d['a']), ascii(d['b'])) - - def test_issue7820(self): - # Ensure that check_bom() restores all bytes in the right order if - # check_bom() fails in pydebug mode: a buffer starts with the first - # byte of a valid BOM, but next bytes are different - - # one byte in common with the UTF-16-LE BOM - self.assertRaises(SyntaxError, eval, b'\xff\x20') - - # two bytes in common with the UTF-8 BOM - self.assertRaises(SyntaxError, eval, b'\xef\xbb\x20') - - def test_error_message(self): - compile(b'# -*- coding: iso-8859-15 -*-\n', 'dummy', 'exec') - compile(b'\xef\xbb\xbf\n', 'dummy', 'exec') - compile(b'\xef\xbb\xbf# -*- coding: utf-8 -*-\n', 'dummy', 'exec') - with self.assertRaisesRegexp(SyntaxError, 'fake'): - compile(b'# -*- coding: fake -*-\n', 'dummy', 'exec') - with self.assertRaisesRegexp(SyntaxError, 'iso-8859-15'): - compile(b'\xef\xbb\xbf# -*- coding: iso-8859-15 -*-\n', - 'dummy', 'exec') - with self.assertRaisesRegexp(SyntaxError, 'BOM'): - compile(b'\xef\xbb\xbf# -*- coding: iso-8859-15 -*-\n', - 'dummy', 'exec') - with self.assertRaisesRegexp(SyntaxError, 'fake'): - compile(b'\xef\xbb\xbf# -*- coding: fake -*-\n', 'dummy', 'exec') - with self.assertRaisesRegexp(SyntaxError, 'BOM'): - compile(b'\xef\xbb\xbf# -*- coding: fake -*-\n', 'dummy', 'exec') - - -def test_main(): - support.run_unittest(PEP263Test) - -if __name__=="__main__": - test_main() diff -r 668aba845fb2 Lib/test/test_source_encoding.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/Lib/test/test_source_encoding.py Mon Jun 10 19:52:53 2013 -0700 @@ -0,0 +1,135 @@ +# -*- coding: koi8-r -*- + +import unittest, test.support +from test.support import TESTFN, unlink, unload +import importlib, os, sys + +class testSourceEncoding(unittest.TestCase): + + def test_pep263(self): + self.assertEqual( + "ðÉÔÏÎ".encode("utf-8"), + b'\xd0\x9f\xd0\xb8\xd1\x82\xd0\xbe\xd0\xbd' + ) + self.assertEqual( + "\ð".encode("utf-8"), + b'\\\xd0\x9f' + ) + + def test_compilestring(self): + # see #1882 + c = compile(b"\n# coding: utf-8\nu = '\xc3\xb3'\n", "dummy", "exec") + d = {} + exec(c, d) + self.assertEqual(d['u'], '\xf3') + + def test_issue2301(self): + try: + compile(b"# coding: cp932\nprint '\x94\x4e'", "dummy", "exec") + except SyntaxError as v: + self.assertEqual(v.text, "print '\u5e74'\n") + else: + self.fail() + + def test_issue4626(self): + c = compile("# coding=latin-1\n\u00c6 = '\u00c6'", "dummy", "exec") + d = {} + exec(c, d) + self.assertEqual(d['\xc6'], '\xc6') + + def test_issue3297(self): + c = compile("a, b = '\U0001010F', '\\U0001010F'", "dummy", "exec") + d = {} + exec(c, d) + self.assertEqual(d['a'], d['b']) + self.assertEqual(len(d['a']), len(d['b'])) + self.assertEqual(ascii(d['a']), ascii(d['b'])) + + def test_issue7820(self): + # Ensure that check_bom() restores all bytes in the right order if + # check_bom() fails in pydebug mode: a buffer starts with the first + # byte of a valid BOM, but next bytes are different + + # one byte in common with the UTF-16-LE BOM + self.assertRaises(SyntaxError, eval, b'\xff\x20') + + # two bytes in common with the UTF-8 BOM + self.assertRaises(SyntaxError, eval, b'\xef\xbb\x20') + + def test_error_message(self): + compile(b'# -*- coding: iso-8859-15 -*-\n', 'dummy', 'exec') + compile(b'\xef\xbb\xbf\n', 'dummy', 'exec') + compile(b'\xef\xbb\xbf# -*- coding: utf-8 -*-\n', 'dummy', 'exec') + with self.assertRaisesRegexp(SyntaxError, 'fake'): + compile(b'# -*- coding: fake -*-\n', 'dummy', 'exec') + with self.assertRaisesRegexp(SyntaxError, 'iso-8859-15'): + compile(b'\xef\xbb\xbf# -*- coding: iso-8859-15 -*-\n', + 'dummy', 'exec') + with self.assertRaisesRegexp(SyntaxError, 'BOM'): + compile(b'\xef\xbb\xbf# -*- coding: iso-8859-15 -*-\n', + 'dummy', 'exec') + with self.assertRaisesRegexp(SyntaxError, 'fake'): + compile(b'\xef\xbb\xbf# -*- coding: fake -*-\n', 'dummy', 'exec') + with self.assertRaisesRegexp(SyntaxError, 'BOM'): + compile(b'\xef\xbb\xbf# -*- coding: fake -*-\n', 'dummy', 'exec') + + def test_bad_coding(self): + module_name = 'bad_coding' + self.verify_bad_module(module_name) + + def test_bad_coding2(self): + module_name = 'bad_coding2' + self.verify_bad_module(module_name) + + def verify_bad_module(self, module_name): + self.assertRaises(SyntaxError, __import__, 'test.' + module_name) + + path = os.path.dirname(__file__) + filename = os.path.join(path, module_name + '.py') + with open(filename, "rb") as fp: + bytes = fp.read() + self.assertRaises(SyntaxError, compile, bytes, filename, 'exec') + + def test_exec_valid_coding(self): + d = {} + exec('# coding: cp949\na = 5\n', d) + self.assertEqual(d['a'], 5) + + def test_file_parse(self): + # issue1134: all encodings outside latin-1 and utf-8 fail on + # multiline strings and long lines (>512 columns) + unload(TESTFN) + sys.path.insert(0, os.curdir) + filename = TESTFN + ".py" + f = open(filename, "w") + try: + f.write("# -*- coding: cp1252 -*-\n") + f.write("'''A short string\n") + f.write("'''\n") + f.write("'A very long string %s'\n" % ("X" * 1000)) + f.close() + + importlib.invalidate_caches() + __import__(TESTFN) + finally: + f.close() + unlink(filename) + unlink(filename + "c") + unload(TESTFN) + del sys.path[0] + + def test_error_from_string(self): + # See http://bugs.python.org/issue6289 + input = "# coding: ascii\n\N{SNOWMAN}".encode('utf-8') + with self.assertRaises(SyntaxError) as c: + compile(input, "", "exec") + expected = "'ascii' codec can't decode byte 0xe2 in position 16: " \ + "ordinal not in range(128)" + self.assertTrue(c.exception.args[0].startswith(expected)) + + +def test_main(): + test.support.run_unittest(testSourceEncoding) + +if __name__=="__main__": + test_main()