test_email.py

来自「mallet是自然语言处理、机器学习领域的一个开源项目。」· Python 代码 · 共 1,770 行 · 第 1/5 页

        else:
            tzsecs = time.altzone
        if tzsecs > 0:
            sign = '-'
        else:
            sign = '+'
        tzoffset = ' %s%04d' % (sign, tzsecs / 36)
        container['Date'] = time.strftime(
            '%a, %d %b %Y %H:%M:%S',
            time.localtime(now)) + tzoffset
        self._msg = container
        self._im = image
        self._txt = intro

    def test_hierarchy(self):
        # convenience
        eq = self.assertEqual
        unless = self.failUnless
        raises = self.assertRaises
        # tests
        m = self._msg
        unless(m.is_multipart())
        eq(m.get_type(), 'multipart/mixed')
        eq(len(m.get_payload()), 2)
        raises(IndexError, m.get_payload, 2)
        m0 = m.get_payload(0)
        m1 = m.get_payload(1)
        unless(m0 is self._txt)
        unless(m1 is self._im)
        eq(m.get_payload(), [m0, m1])
        unless(not m0.is_multipart())
        unless(not m1.is_multipart())

    def test_no_parts_in_a_multipart(self):
        outer = MIMEBase('multipart', 'mixed')
        outer['Subject'] = 'A subject'
        outer['To'] = '[e-mail]'
        outer['From'] = '[e-mail]'
        outer.preamble = ''
        outer.epilogue = ''
        outer.set_boundary('BOUNDARY')
        msg = MIMEText('hello world')
        self.assertEqual(outer.as_string(), '''\
Content-Type: multipart/mixed; boundary="BOUNDARY"
MIME-Version: 1.0
Subject: A subject
To: [e-mail]
From: [e-mail]

--BOUNDARY


--BOUNDARY--
''')

    def test_one_part_in_a_multipart(self):
        eq = self.ndiffAssertEqual
        outer = MIMEBase('multipart', 'mixed')
        outer['Subject'] = 'A subject'
        outer['To'] = '[e-mail]'
        outer['From'] = '[e-mail]'
        outer.preamble = ''
        outer.epilogue = ''
        outer.set_boundary('BOUNDARY')
        msg = MIMEText('hello world')
        outer.attach(msg)
        eq(outer.as_string(), '''\
Content-Type: multipart/mixed; boundary="BOUNDARY"
MIME-Version: 1.0
Subject: A subject
To: [e-mail]
From: [e-mail]

--BOUNDARY
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit

hello world
--BOUNDARY--
''')

    def test_seq_parts_in_a_multipart(self):
        eq = self.ndiffAssertEqual
        outer = MIMEBase('multipart', 'mixed')
        outer['Subject'] = 'A subject'
        outer['To'] = '[e-mail]'
        outer['From'] = '[e-mail]'
        outer.preamble = ''
        outer.epilogue = ''
        msg = MIMEText('hello world')
        outer.attach(msg)
        outer.set_boundary('BOUNDARY')
        eq(outer.as_string(), '''\
Content-Type: multipart/mixed; boundary="BOUNDARY"
MIME-Version: 1.0
Subject: A subject
To: [e-mail]
From: [e-mail]

--BOUNDARY
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit

hello world
--BOUNDARY--
''')



# Test some badly formatted messages
class TestNonConformant(TestEmailBase):
    def test_parse_missing_minor_type(self):
        eq = self.assertEqual
        msg = self._msgobj('msg_14.txt')
        eq(msg.get_type(), 'text')
        eq(msg.get_main_type(), None)
        eq(msg.get_subtype(), None)

    def test_bogus_boundary(self):
        fp = openfile(findfile('msg_15.txt'))
        try:
            data = fp.read()
        finally:
            fp.close()
        p = Parser(strict=True)
        # Note, under a future non-strict parsing mode, this would parse the
        # message into the intended message tree.
        self.assertRaises(Errors.BoundaryError, p.parsestr, data)

    def test_multipart_no_boundary(self):
        fp = openfile(findfile('msg_25.txt'))
        try:
            self.assertRaises(Errors.BoundaryError,
                              email.message_from_file, fp)
        finally:
            fp.close()

    def test_invalid_content_type(self):
        eq = self.assertEqual
        neq = self.ndiffAssertEqual
        msg = Message()
        # RFC 2045, $5.2 says invalid yields text/plain
        msg['Content-Type'] = 'text'
        eq(msg.get_content_maintype(), 'text')
        eq(msg.get_content_subtype(), 'plain')
        eq(msg.get_content_type(), 'text/plain')
        # Clear the old value and try something /really/ invalid
        del msg['content-type']
        msg['Content-Type'] = 'foo'
        eq(msg.get_content_maintype(), 'text')
        eq(msg.get_content_subtype(), 'plain')
        eq(msg.get_content_type(), 'text/plain')
        # Still, make sure that the message is idempotently generated
        s = StringIO()
        g = Generator(s)
        g.flatten(msg)
        neq(s.getvalue(), 'Content-Type: foo\n\n')

    def test_no_start_boundary(self):
        eq = self.ndiffAssertEqual
        msg = self._msgobj('msg_31.txt')
        eq(msg.get_payload(), """\
--BOUNDARY
Content-Type: text/plain

message 1

--BOUNDARY
Content-Type: text/plain

message 2

--BOUNDARY--
""")

    def test_no_separating_blank_line(self):
        eq = self.ndiffAssertEqual
        msg = self._msgobj('msg_35.txt')
        eq(msg.as_string(), """\
From: [e-mail]
To: [e-mail]
Subject: here's something interesting

counter to RFC 2822, there's no separating newline here
""")
        # strict=True should raise an exception
        self.assertRaises(Errors.HeaderParseError,
                          self._msgobj, 'msg_35.txt', True)



# Test RFC 2047 header encoding and decoding
class TestRFC2047(unittest.TestCase):
    def test_iso_8859_1(self):
        eq = self.assertEqual
        s = '=?iso-8859-1?q?this=20is=20some=20text?='
        eq(Utils.decode(s), 'this is some text')
        s = '=?ISO-8859-1?Q?Keld_J=F8rn_Simonsen?='
        eq(Utils.decode(s), u'Keld J\xf8rn Simonsen')
        s = '=?ISO-8859-1?B?SWYgeW91IGNhbiByZWFkIHRoaXMgeW8=?=' \
            '=?ISO-8859-2?B?dSB1bmRlcnN0YW5kIHRoZSBleGFtcGxlLg==?='
        eq(Utils.decode(s), 'If you can read this you understand the example.')
        s = '=?iso-8859-8?b?7eXs+SDv4SDp7Oj08A==?='
        eq(Utils.decode(s),
           u'\u05dd\u05d5\u05dc\u05e9 \u05df\u05d1 \u05d9\u05dc\u05d8\u05e4\u05e0')
        s = '=?iso-8859-1?q?this=20is?= =?iso-8859-1?q?some=20text?='
        eq(Utils.decode(s), u'this issome text')
        s = '=?iso-8859-1?q?this=20is_?= =?iso-8859-1?q?some=20text?='
        eq(Utils.decode(s), u'this is some text')

    def test_encode_header(self):
        eq = self.assertEqual
        s = 'this is some text'
        eq(Utils.encode(s), '=?iso-8859-1?q?this=20is=20some=20text?=')
        s = 'Keld_J\xf8rn_Simonsen'
        eq(Utils.encode(s), '=?iso-8859-1?q?Keld_J=F8rn_Simonsen?=')
        s1 = 'If you can read this yo'
        s2 = 'u understand the example.'
        eq(Utils.encode(s1, encoding='b'),
           '=?iso-8859-1?b?SWYgeW91IGNhbiByZWFkIHRoaXMgeW8=?=')
        eq(Utils.encode(s2, charset='iso-8859-2', encoding='b'),
           '=?iso-8859-2?b?dSB1bmRlcnN0YW5kIHRoZSBleGFtcGxlLg==?=')

    def test_rfc2047_multiline(self):
        eq = self.assertEqual
        s = """Re: =?mac-iceland?q?r=8Aksm=9Arg=8Cs?= baz
 foo bar =?mac-iceland?q?r=8Aksm=9Arg=8Cs?="""
        dh = decode_header(s)
        eq(dh, [
            ('Re:', None),
            ('r\x8aksm\x9arg\x8cs', 'mac-iceland'),
            ('baz foo bar', None),
            ('r\x8aksm\x9arg\x8cs', 'mac-iceland')])
        eq(str(make_header(dh)),
           """Re: =?mac-iceland?q?r=8Aksm=9Arg=8Cs?= baz foo bar
 =?mac-iceland?q?r=8Aksm=9Arg=8Cs?=""")

    def test_whitespace_eater_unicode(self):
        eq = self.assertEqual
        s = '=?ISO-8859-1?Q?Andr=E9?= Pirard <[e-mail]>'
        dh = decode_header(s)
        eq(dh, [('Andr\xe9', 'iso-8859-1'), ('Pirard <[e-mail]>', None)])
        # Python 2.1's unicode() builtin doesn't call the object's
        # __unicode__() method.  Use the following alternative instead.
        #hu = unicode(make_header(dh)).encode('latin-1')
        hu = make_header(dh).__unicode__().encode('latin-1')
        eq(hu, 'Andr\xe9 Pirard <[e-mail]>')



# Test the MIMEMessage class
class TestMIMEMessage(TestEmailBase):
    def setUp(self):
        fp = openfile('msg_11.txt')
        try:
            self._text = fp.read()
        finally:
            fp.close()

    def test_type_error(self):
        self.assertRaises(TypeError, MIMEMessage, 'a plain string')

    def test_valid_argument(self):
        eq = self.assertEqual
        unless = self.failUnless
        subject = 'A sub-message'
        m = Message()
        m['Subject'] = subject
        r = MIMEMessage(m)
        eq(r.get_type(), 'message/rfc822')
        payload = r.get_payload()
        unless(type(payload), ListType)
        eq(len(payload), 1)
        subpart = payload[0]
        unless(subpart is m)
        eq(subpart['subject'], subject)

    def test_bad_multipart(self):
        eq = self.assertEqual
        msg1 = Message()
        msg1['Subject'] = 'subpart 1'
        msg2 = Message()
        msg2['Subject'] = 'subpart 2'
        r = MIMEMessage(msg1)
        self.assertRaises(Errors.MultipartConversionError, r.attach, msg2)

    def test_generate(self):
        # First craft the message to be encapsulated
        m = Message()
        m['Subject'] = 'An enclosed message'
        m.set_payload('Here is the body of the message.\n')
        r = MIMEMessage(m)
        r['Subject'] = 'The enclosing message'
        s = StringIO()
        g = Generator(s)
        g.flatten(r)
        self.assertEqual(s.getvalue(), """\
Content-Type: message/rfc822
MIME-Version: 1.0
Subject: The enclosing message

Subject: An enclosed message

Here is the body of the message.
""")

    def test_parse_message_rfc822(self):
        eq = self.assertEqual
        unless = self.failUnless
        msg = self._msgobj('msg_11.txt')
        eq(msg.get_type(), 'message/rfc822')
        payload = msg.get_payload()
        unless(isinstance(payload, ListType))
        eq(len(payload), 1)
        submsg = payload[0]
        self.failUnless(isinstance(submsg, Message))
        eq(submsg['subject'], 'An enclosed message')
        eq(submsg.get_payload(), 'Here is the body of the message.\n')

    def test_dsn(self):
        eq = self.assertEqual
        unless = self.failUnless
        # msg 16 is a Delivery Status Notification, see RFC 1894
        msg = self._msgobj('msg_16.txt')
        eq(msg.get_type(), 'multipart/report')
        unless(msg.is_multipart())
        eq(len(msg.get_payload()), 3)
        # Subpart 1 is a text/plain, human readable section
        subpart = msg.get_payload(0)
        eq(subpart.get_type(), 'text/plain')
        eq(subpart.get_payload(), """\
This report relates to a message you sent with the following header fields:

  Message-id: <002001c144a6$8752e060$[e-mail]>
  Date: Sun, 23 Sep 2001 20:10:55 -0700
  From: "Ian T. Henry" <[e-mail]>
  To: SoCal Raves <[e-mail]>
  Subject: [scr] yeah for Ians!!

Your message cannot be delivered to the following recipients:

  Recipient address: [e-mail]
  Reason: recipient reached disk quota

""")
        # Subpart 2 contains the machine parsable DSN information.  It
        # consists of two blocks of headers, represented by two nested Message
        # objects.
        subpart = msg.get_payload(1)
        eq(subpart.get_type(), 'message/delivery-status')
        eq(len(subpart.get_payload()), 2)
        # message/delivery-status should treat each block as a bunch of