#14380: Have MIMEText defaults to utf-8 when passed non-ASCII unicode
Previously it would just accept the unicode, which would wind up as unicode in
the transfer-encoded message object, which is just wrong.
Patch by Jeff Knupp.
diff --git a/Lib/email/mime/text.py b/Lib/email/mime/text.py
index 5747db5..5a5e39b 100644
--- a/Lib/email/mime/text.py
+++ b/Lib/email/mime/text.py
@@ -27,4 +27,14 @@
"""
MIMENonMultipart.__init__(self, 'text', _subtype,
**{'charset': _charset})
+
+ # If _charset was defualted, check to see see if there are non-ascii
+ # characters present. Default to utf-8 if there are.
+ # XXX: This can be removed once #7304 is fixed.
+ if _charset =='us-ascii':
+ try:
+ _text.encode(_charset)
+ except UnicodeEncodeError:
+ _charset = 'utf-8'
+
self.set_payload(_text, _charset)
diff --git a/Lib/test/test_email/test_email.py b/Lib/test/test_email/test_email.py
index e771c20..86aa60c 100644
--- a/Lib/test/test_email/test_email.py
+++ b/Lib/test/test_email/test_email.py
@@ -617,6 +617,19 @@
abc
"""))
+ def test_unicode_body_defaults_to_utf8_encoding(self):
+ # Issue 14291
+ m = MIMEText('É testabc\n')
+ self.assertEqual(str(m),textwrap.dedent("""\
+ MIME-Version: 1.0
+ Content-Type: text/plain; charset="utf-8"
+ Content-Transfer-Encoding: base64
+
+ w4kgdGVzdGFiYwo=
+ """))
+
+
+
# Test the email.encoders module
class TestEncoders(unittest.TestCase):
@@ -642,7 +655,7 @@
eq(msg['content-transfer-encoding'], '7bit')
# Similar, but with 8bit data
msg = MIMEText('hello \xf8 world')
- eq(msg['content-transfer-encoding'], '8bit')
+ eq(msg['content-transfer-encoding'], 'base64')
# And now with a different charset
msg = MIMEText('hello \xf8 world', _charset='iso-8859-1')
eq(msg['content-transfer-encoding'], 'quoted-printable')