1# Copyright (C) 2002-2006 Python Software Foundation
2# Contact: email-sig@python.org
3# email package unit tests for (optional) Asian codecs
4
5import unittest
6
7from test.test_email import TestEmailBase
8from email.charset import Charset
9from email.header import Header, decode_header
10from email.message import Message
11
12# We're compatible with Python 2.3, but it doesn't have the built-in Asian
13# codecs, so we have to skip all these tests.
14try:
15    str(b'foo', 'euc-jp')
16except LookupError:
17    raise unittest.SkipTest
18
19
20
21class TestEmailAsianCodecs(TestEmailBase):
22    def test_japanese_codecs(self):
23        eq = self.ndiffAssertEqual
24        jcode = "euc-jp"
25        gcode = "iso-8859-1"
26        j = Charset(jcode)
27        g = Charset(gcode)
28        h = Header("Hello World!")
29        jhello = str(b'\xa5\xcf\xa5\xed\xa1\xbc\xa5\xef\xa1\xbc'
30                     b'\xa5\xeb\xa5\xc9\xa1\xaa', jcode)
31        ghello = str(b'Gr\xfc\xdf Gott!', gcode)
32        h.append(jhello, j)
33        h.append(ghello, g)
34        # BAW: This used to -- and maybe should -- fold the two iso-8859-1
35        # chunks into a single encoded word.  However it doesn't violate the
36        # standard to have them as two encoded chunks and maybe it's
37        # reasonable <wink> for each .append() call to result in a separate
38        # encoded word.
39        eq(h.encode(), """\
40Hello World! =?iso-2022-jp?b?GyRCJU8lbSE8JW8hPCVrJUkhKhsoQg==?=
41 =?iso-8859-1?q?Gr=FC=DF_Gott!?=""")
42        eq(decode_header(h.encode()),
43           [(b'Hello World! ', None),
44            (b'\x1b$B%O%m!<%o!<%k%I!*\x1b(B', 'iso-2022-jp'),
45            (b'Gr\xfc\xdf Gott!', gcode)])
46        subject_bytes = (b'test-ja \xa4\xd8\xc5\xea\xb9\xc6\xa4\xb5'
47            b'\xa4\xec\xa4\xbf\xa5\xe1\xa1\xbc\xa5\xeb\xa4\xcf\xbb\xca\xb2'
48            b'\xf1\xbc\xd4\xa4\xce\xbe\xb5\xc7\xa7\xa4\xf2\xc2\xd4\xa4\xc3'
49            b'\xa4\xc6\xa4\xa4\xa4\xde\xa4\xb9')
50        subject = str(subject_bytes, jcode)
51        h = Header(subject, j, header_name="Subject")
52        # test a very long header
53        enc = h.encode()
54        # TK: splitting point may differ by codec design and/or Header encoding
55        eq(enc , """\
56=?iso-2022-jp?b?dGVzdC1qYSAbJEIkWEVqOUYkNSRsJD8lYSE8JWskTztKGyhC?=
57 =?iso-2022-jp?b?GyRCMnE8VCROPjVHJyRyQlQkQyRGJCQkXiQ5GyhC?=""")
58        # TK: full decode comparison
59        eq(str(h).encode(jcode), subject_bytes)
60
61    def test_payload_encoding_utf8(self):
62        jhello = str(b'\xa5\xcf\xa5\xed\xa1\xbc\xa5\xef\xa1\xbc'
63                     b'\xa5\xeb\xa5\xc9\xa1\xaa', 'euc-jp')
64        msg = Message()
65        msg.set_payload(jhello, 'utf-8')
66        ustr = msg.get_payload(decode=True).decode(msg.get_content_charset())
67        self.assertEqual(jhello, ustr)
68
69    def test_payload_encoding(self):
70        jcode  = 'euc-jp'
71        jhello = str(b'\xa5\xcf\xa5\xed\xa1\xbc\xa5\xef\xa1\xbc'
72                     b'\xa5\xeb\xa5\xc9\xa1\xaa', jcode)
73        msg = Message()
74        msg.set_payload(jhello, jcode)
75        ustr = msg.get_payload(decode=True).decode(msg.get_content_charset())
76        self.assertEqual(jhello, ustr)
77
78
79
80if __name__ == '__main__':
81    unittest.main()
82