[1.1.X] Fixed #6918, #12791: If an email message has an encoding, actually use that encoding to encode body and headers. Thanks for patch with tests oyvind.

kmtracey · kmtracey · commit daec7345845c · 2010-03-06T16:07:05.000Z
Backport of r12683 and r12688 from trunk. git-svn-id: http://code.djangoproject.com/svn/django/branches/releases/1.1.X@12689 bcc190cf-cafb-0310-a4f2-bffc1f526a37
diff --git a/django/core/mail.py b/django/core/mail.py
@@ -69,8 +69,9 @@ def make_msgid(idstring=None):
 class BadHeaderError(ValueError):
     pass
 
-def forbid_multi_line_headers(name, val):
+def forbid_multi_line_headers(name, val, encoding):
     """Forbids multi-line headers, to prevent header injection."""
+    encoding = encoding or settings.DEFAULT_CHARSET
     val = force_unicode(val)
     if '\n' in val or '\r' in val:
         raise BadHeaderError("Header values can't contain newlines (got %r for header %r)" % (val, name))
@@ -80,24 +81,34 @@ def forbid_multi_line_headers(name, val):
         if name.lower() in ('to', 'from', 'cc'):
             result = []
             for nm, addr in getaddresses((val,)):
-                nm = str(Header(nm, settings.DEFAULT_CHARSET))
+                nm = str(Header(nm.encode(encoding), encoding))
                 result.append(formataddr((nm, str(addr))))
             val = ', '.join(result)
         else:
-            val = Header(val, settings.DEFAULT_CHARSET)
+            val = Header(val.encode(encoding), encoding)
     else:
         if name.lower() == 'subject':
             val = Header(val)
     return name, val
 
 class SafeMIMEText(MIMEText):
+
+    def __init__(self, text, subtype, charset):
+        self.encoding = charset
+        MIMEText.__init__(self, text, subtype, charset)
+
     def __setitem__(self, name, val):
-        name, val = forbid_multi_line_headers(name, val)
+        name, val = forbid_multi_line_headers(name, val, self.encoding)
         MIMEText.__setitem__(self, name, val)
 
 class SafeMIMEMultipart(MIMEMultipart):
+
+    def __init__(self, _subtype='mixed', boundary=None, _subparts=None, encoding=None, **_params):
+        self.encoding = encoding
+        MIMEMultipart.__init__(self, _subtype, boundary, _subparts, **_params)
+
     def __setitem__(self, name, val):
-        name, val = forbid_multi_line_headers(name, val)
+        name, val = forbid_multi_line_headers(name, val, self.encoding)
         MIMEMultipart.__setitem__(self, name, val)
 
 class SMTPConnection(object):
@@ -234,7 +245,7 @@ def get_connection(self, fail_silently=False):
 
     def message(self):
         encoding = self.encoding or settings.DEFAULT_CHARSET
-        msg = SafeMIMEText(smart_str(self.body, settings.DEFAULT_CHARSET),
+        msg = SafeMIMEText(smart_str(self.body, encoding),
                            self.content_subtype, encoding)
         msg = self._create_message(msg)
         msg['Subject'] = self.subject
@@ -293,8 +304,9 @@ def _create_message(self, msg):
 
     def _create_attachments(self, msg):
         if self.attachments:
+            encoding = self.encoding or settings.DEFAULT_CHARSET
             body_msg = msg
-            msg = SafeMIMEMultipart(_subtype=self.mixed_subtype)
+            msg = SafeMIMEMultipart(_subtype=self.mixed_subtype, encoding=encoding)
             if self.body:
                 msg.attach(body_msg)
             for attachment in self.attachments:
@@ -310,8 +322,8 @@ def _create_mime_attachment(self, content, mimetype):
         """
         basetype, subtype = mimetype.split('/', 1)
         if basetype == 'text':
-            attachment = SafeMIMEText(smart_str(content,
-                settings.DEFAULT_CHARSET), subtype, settings.DEFAULT_CHARSET)
+            encoding = self.encoding or settings.DEFAULT_CHARSET
+            attachment = SafeMIMEText(smart_str(content, encoding), subtype, encoding)
         else:
             # Encode non-text attachments with base64.
             attachment = MIMEBase(basetype, subtype)
@@ -365,9 +377,10 @@ def _create_message(self, msg):
         return self._create_attachments(self._create_alternatives(msg))
 
     def _create_alternatives(self, msg):
+        encoding = self.encoding or settings.DEFAULT_CHARSET
         if self.alternatives:
             body_msg = msg
-            msg = SafeMIMEMultipart(_subtype=self.alternative_subtype)
+            msg = SafeMIMEMultipart(_subtype=self.alternative_subtype, encoding=encoding)
             if self.body:
                 msg.attach(body_msg)
             for alternative in self.alternatives:
diff --git a/tests/regressiontests/mail/tests.py b/tests/regressiontests/mail/tests.py
@@ -106,10 +106,49 @@
 >>> email.message()['To']
 '=?utf-8?q?S=C3=BCrname=2C_Firstname?= <to@example.com>, other@example.com'
 
+# Regression for #6918 - When a header contains unicode,
+# make sure headers can be set with a different encoding than utf-8
+>>> email = EmailMessage('Message from Firstname Sürname', 'Content', 'from@example.com', ['"Sürname, Firstname" <to@example.com>','other@example.com'])
+>>> email.encoding = 'iso-8859-1'
+>>> email.message()['To']
+'=?iso-8859-1?q?S=FCrname=2C_Firstname?= <to@example.com>, other@example.com'
+>>> email.message()['Subject'].encode()
+u'=?iso-8859-1?q?Message_from_Firstname_S=FCrname?='
+
+# Make sure headers can be set with a different encoding than utf-8 in SafeMIMEMultipart as well
+>>> headers = {"Date": "Fri, 09 Nov 2001 01:08:47 -0000", "Message-ID": "foo"}
+>>> subject, from_email, to = 'hello', 'from@example.com', '"Sürname, Firstname" <to@example.com>'
+>>> text_content = 'This is an important message.'
+>>> html_content = '<p>This is an <strong>important</strong> message.</p>'
+>>> msg = EmailMultiAlternatives('Message from Firstname Sürname', text_content, from_email, [to], headers=headers)
+>>> msg.attach_alternative(html_content, "text/html")
+>>> msg.encoding = 'iso-8859-1'
+>>> msg.message()['To']
+'=?iso-8859-1?q?S=FCrname=2C_Firstname?= <to@example.com>'
+>>> msg.message()['Subject'].encode()
+u'=?iso-8859-1?q?Message_from_Firstname_S=FCrname?='
+
+# Regression for #12791  - Encode body correctly with other encodings than utf-8
+>>> email = EmailMessage('Subject', 'Firstname Sürname is a great guy.', 'from@example.com', ['other@example.com'])
+>>> email.encoding = 'iso-8859-1'
+>>> message = email.message()
+>>> message.as_string()
+'Content-Type: text/plain; charset="iso-8859-1"\nMIME-Version: 1.0\nContent-Transfer-Encoding: quoted-printable\nSubject: Subject\nFrom: from@example.com\nTo: other@example.com\nDate: ...\nMessage-ID: <...>\n\nFirstname S=FCrname is a great guy.'
+
+# Make sure MIME attachments also works correctly with other encodings than utf-8
+>>> text_content = 'Firstname Sürname is a great guy.'
+>>> html_content = '<p>Firstname Sürname is a <strong>great</strong> guy.</p>'
+>>> msg = EmailMultiAlternatives('Subject', text_content, 'from@example.com', ['to@example.com'])
+>>> msg.encoding = 'iso-8859-1'
+>>> msg.attach_alternative(html_content, "text/html")
+>>> msg.message().get_payload(0).as_string()
+'Content-Type: text/plain; charset="iso-8859-1"\nMIME-Version: 1.0\nContent-Transfer-Encoding: quoted-printable\n\nFirstname S=FCrname is a great guy.'
+>>> msg.message().get_payload(1).as_string()
+'Content-Type: text/html; charset="iso-8859-1"\nMIME-Version: 1.0\nContent-Transfer-Encoding: quoted-printable\n\n<p>Firstname S=FCrname is a <strong>great</strong> guy.</p>'
+
 # Handle attachments within an multipart/alternative mail correctly (#9367)
 # (test is not as precise/clear as it could be w.r.t. email tree structure,
 #  but it's good enough.)
-
 >>> headers = {"Date": "Fri, 09 Nov 2001 01:08:47 -0000", "Message-ID": "foo"}
 >>> subject, from_email, to = 'hello', 'from@example.com', 'to@example.com'
 >>> text_content = 'This is an important message.'
