blob: d5a5b3e35e68d7dbe071647720cc26529da795a6 [file] [log] [blame]
# -*- 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.assertEquals(v.text, "print '\u5e74'")
else:
self.fail()
def test_issue4626(self):
c = compile("# coding=latin-1\n\u00c6 = '\u00c6'", "dummy", "exec")
d = {}
exec(c, d)
self.assertEquals(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_main():
support.run_unittest(PEP263Test)
if __name__=="__main__":
test_main()