Fixed #27007 -- Handled non-UTF-8 bytes objects for text/* attachments.

The fallback logic which allows non-UTF-8 encoded files to be passed to
attach_file() even when a `text/*` mime type has been specified is
moved to attach(). Both functions now fall back to a content type of
`application/octet-stream`.

A side effect is that a file's content is decoded in memory instead of
opening it in text mode and reading it into a string.

Some mimetype-related logic in _create_attachment() has become
obsolete as the code moved from attach_file() to attach() already
handles this.

1 files changed, 29 insertions, 28 deletions

diff --git a/django/core/mail/message.py b/django/core/mail/message.py
index 29012201ef..44178d2447 100644
--- a/django/core/mail/message.py
+++ b/django/core/mail/message.py
@@ -356,6 +356,11 @@ class EmailMessage(object):
 
         If the first parameter is a MIMEBase subclass it is inserted directly
         into the resulting message attachments.
+
+        For a text/* mimetype (guessed or specified), when a bytes object is
+        specified as content, it will be decoded as UTF-8. If that fails,
+        the mimetype will be set to DEFAULT_ATTACHMENT_MIME_TYPE and the
+        content is not decoded.
         """
         if isinstance(filename, MIMEBase):
             assert content is None
@@ -363,6 +368,22 @@ class EmailMessage(object):
             self.attachments.append(filename)
         else:
             assert content is not None
+
+            if not mimetype:
+                mimetype, _ = mimetypes.guess_type(filename)
+                if not mimetype:
+                    mimetype = DEFAULT_ATTACHMENT_MIME_TYPE
+            basetype, subtype = mimetype.split('/', 1)
+
+            if basetype == 'text':
+                if isinstance(content, six.binary_type):
+                    try:
+                        content = content.decode('utf-8')
+                    except UnicodeDecodeError:
+                        # If mimetype suggests the file is text but it's actually
+                        # binary, read() will raise a UnicodeDecodeError on Python 3.
+                        mimetype = DEFAULT_ATTACHMENT_MIME_TYPE
+
             self.attachments.append((filename, content, mimetype))
 
     def attach_file(self, path, mimetype=None):
@@ -370,33 +391,17 @@ class EmailMessage(object):
         Attaches a file from the filesystem.
 
         The mimetype will be set to the DEFAULT_ATTACHMENT_MIME_TYPE if it is
-        not specified and cannot be guessed or (PY3 only) if it suggests
-        text/* for a binary file.
+        not specified and cannot be guessed.
+
+        For a text/* mimetype (guessed or specified), the file's content
+        will be decoded as UTF-8. If that fails, the mimetype will be set to
+        DEFAULT_ATTACHMENT_MIME_TYPE and the content is not decoded.
         """
         filename = os.path.basename(path)
-        if not mimetype:
-            mimetype, _ = mimetypes.guess_type(filename)
-            if not mimetype:
-                mimetype = DEFAULT_ATTACHMENT_MIME_TYPE
-        basetype, subtype = mimetype.split('/', 1)
-        read_mode = 'r' if basetype == 'text' else 'rb'
-        content = None
-
-        with open(path, read_mode) as f:
-            try:
-                content = f.read()
-            except UnicodeDecodeError:
-                # If mimetype suggests the file is text but it's actually
-                # binary, read() will raise a UnicodeDecodeError on Python 3.
-                pass
-
-        # If the previous read in text mode failed, try binary mode.
-        if content is None:
-            with open(path, 'rb') as f:
-                content = f.read()
-                mimetype = DEFAULT_ATTACHMENT_MIME_TYPE
 
-        self.attach(filename, content, mimetype)
+        with open(path, 'rb') as file:
+            content = file.read()
+            self.attach(filename, content, mimetype)
 
     def _create_message(self, msg):
         return self._create_attachments(msg)
@@ -450,10 +455,6 @@ class EmailMessage(object):
         Converts the filename, content, mimetype triple into a MIME attachment
         object.
         """
-        if mimetype is None:
-            mimetype, _ = mimetypes.guess_type(filename)
-            if mimetype is None:
-                mimetype = DEFAULT_ATTACHMENT_MIME_TYPE
         attachment = self._create_mime_attachment(content, mimetype)
         if filename:
             try: