diff --git a/Lib/email/header.py b/Lib/email/header.py
index 7e235df..0a66df5 100644
--- a/Lib/email/header.py
+++ b/Lib/email/header.py
@@ -26,6 +26,7 @@
 SPACE8 = ' ' * 8
 EMPTYSTRING = ''
 MAXLINELEN = 78
+FWS = ' \t'
 
 USASCII = Charset('us-ascii')
 UTF8 = Charset('utf-8')
@@ -299,9 +300,15 @@
         name was specified at Header construction time.  The default value for
         maxlinelen is determined at header construction time.
 
-        Optional splitchars is a string containing characters to split long
-        ASCII lines on, in rough support of RFC 2822's `highest level
-        syntactic breaks'.  This doesn't affect RFC 2047 encoded lines.
+        Optional splitchars is a string containing characters which should be
+        given extra weight by the splitting algorithm during normal header
+        wrapping.  This is in very rough support of RFC 2822's `higher level
+        syntactic breaks':  split points preceded by a splitchar are preferred
+        during line splitting, with the characters preferred in the order in
+        which they appear in the string.  Space and tab may be included in the
+        string to indicate whether preference should be given to one over the
+        other as a split point when other split chars do not appear in the line
+        being split.  Splitchars does not affect RFC 2047 encoded lines.
 
         Optional linesep is a string to be used to separate the lines of
         the value.  The default value is the most useful for typical
@@ -320,13 +327,19 @@
                                     self._continuation_ws, splitchars)
         for string, charset in self._chunks:
             lines = string.splitlines()
-            formatter.feed(lines[0] if lines else '', charset)
+            if lines:
+                formatter.feed('', lines[0], charset)
+            else:
+                formatter.feed('', '', charset)
             for line in lines[1:]:
                 formatter.newline()
                 if charset.header_encoding is not None:
-                    formatter.feed(self._continuation_ws, USASCII)
-                    line = ' ' + line.lstrip()
-                formatter.feed(line, charset)
+                    formatter.feed(self._continuation_ws, ' ' + line.lstrip(),
+                                   charset)
+                else:
+                    sline = line.lstrip()
+                    fws = line[:len(line)-len(sline)]
+                    formatter.feed(fws, sline, charset)
             if len(lines) > 1:
                 formatter.newline()
             formatter.add_transition()
@@ -360,7 +373,7 @@
     def __init__(self, headerlen, maxlen, continuation_ws, splitchars):
         self._maxlen = maxlen
         self._continuation_ws = continuation_ws
-        self._continuation_ws_len = len(continuation_ws.replace('\t', SPACE8))
+        self._continuation_ws_len = len(continuation_ws)
         self._splitchars = splitchars
         self._lines = []
         self._current_line = _Accumulator(headerlen)
@@ -374,43 +387,26 @@
 
     def newline(self):
         end_of_line = self._current_line.pop()
-        if end_of_line is not None:
-            self._current_line.push(end_of_line)
+        if end_of_line != (' ', ''):
+            self._current_line.push(*end_of_line)
         if len(self._current_line) > 0:
-            self._lines.append(str(self._current_line))
+            if self._current_line.is_onlyws():
+                self._lines[-1] += str(self._current_line)
+            else:
+                self._lines.append(str(self._current_line))
         self._current_line.reset()
 
     def add_transition(self):
-        self._current_line.push(None)
+        self._current_line.push(' ', '')
 
-    def feed(self, string, charset):
-        # If the string itself fits on the current line in its encoded format,
-        # then add it now and be done with it.
-        encoded_string = charset.header_encode(string)
-        if len(encoded_string) + len(self._current_line) <= self._maxlen:
-            self._current_line.push(encoded_string)
-            return
+    def feed(self, fws, string, charset):
         # If the charset has no header encoding (i.e. it is an ASCII encoding)
         # then we must split the header at the "highest level syntactic break"
         # possible. Note that we don't have a lot of smarts about field
         # syntax; we just try to break on semi-colons, then commas, then
         # whitespace.  Eventually, this should be pluggable.
         if charset.header_encoding is None:
-            for ch in self._splitchars:
-                if ch in string:
-                    break
-            else:
-                ch = None
-            # If there's no available split character then regardless of
-            # whether the string fits on the line, we have to put it on a line
-            # by itself.
-            if ch is None:
-                if not self._current_line.is_onlyws():
-                    self._lines.append(str(self._current_line))
-                    self._current_line.reset(self._continuation_ws)
-                self._current_line.push(encoded_string)
-            else:
-                self._ascii_split(string, ch)
+            self._ascii_split(fws, string, self._splitchars)
             return
         # Otherwise, we're doing either a Base64 or a quoted-printable
         # encoding which means we don't need to split the line on syntactic
@@ -428,15 +424,14 @@
             # There are no encoded lines, so we're done.
             return
         if first_line is not None:
-            self._current_line.push(first_line)
-        self._lines.append(str(self._current_line))
-        self._current_line.reset(self._continuation_ws)
+            self._append_chunk(fws, first_line)
         try:
             last_line = encoded_lines.pop()
         except IndexError:
             # There was only one line.
             return
-        self._current_line.push(last_line)
+        self.newline()
+        self._current_line.push(self._continuation_ws, last_line)
         # Everything else are full lines in themselves.
         for line in encoded_lines:
             self._lines.append(self._continuation_ws + line)
@@ -447,162 +442,96 @@
         while True:
             yield self._maxlen - self._continuation_ws_len
 
-    def _ascii_split(self, string, ch):
-        holding = _Accumulator()
-        # Split the line on the split character, preserving it.  If the split
-        # character is whitespace RFC 2822 $2.2.3 requires us to fold on the
-        # whitespace, so that the line leads with the original whitespace we
-        # split on.  However, if a higher syntactic break is used instead
-        # (e.g. comma or semicolon), the folding should happen after the split
-        # character.  But then in that case, we need to add our own
-        # continuation whitespace -- although won't that break unfolding?
-        for part, splitpart, nextpart in _spliterator(ch, string):
-            if not splitpart:
-                # No splitpart means this is the last chunk.  Put this part
-                # either on the current line or the next line depending on
-                # whether it fits.
-                holding.push(part)
-                if len(holding) + len(self._current_line) <= self._maxlen:
-                    # It fits, but we're done.
-                    self._current_line.push(str(holding))
+    def _ascii_split(self, fws, string, splitchars):
+        # The RFC 2822 header folding algorithm is simple in principle but
+        # complex in practice.  Lines may be folded any place where "folding
+        # white space" appears by inserting a linesep character in front of the
+        # FWS.  The complication is that not all spaces or tabs qualify as FWS,
+        # and we are also supposed to prefer to break at "higher level
+        # syntactic breaks".  We can't do either of these without intimate
+        # knowledge of the structure of structured headers, which we don't have
+        # here.  So the best we can do here is prefer to break at the specified
+        # splitchars, and hope that we don't choose any spaces or tabs that
+        # aren't legal FWS.  (This is at least better than the old algorithm,
+        # where we would sometimes *introduce* FWS after a splitchar, or the
+        # algorithm before that, where we would turn all white space runs into
+        # single spaces or tabs.)
+        parts = re.split("(["+FWS+"]+)", fws+string)
+        if parts[0]:
+            parts[:0] = ['']
+        else:
+            parts.pop(0)
+        for fws, part in zip(*[iter(parts)]*2):
+            self._append_chunk(fws, part)
+
+    def _append_chunk(self, fws, string):
+        self._current_line.push(fws, string)
+        if len(self._current_line) > self._maxlen:
+            # Find the best split point, working backward from the end.
+            # There might be none, on a long first line.
+            for ch in self._splitchars:
+                for i in range(self._current_line.part_count()-1, 0, -1):
+                    if ch.isspace():
+                        fws = self._current_line[i][0]
+                        if fws and fws[0]==ch:
+                            break
+                    prevpart = self._current_line[i-1][1]
+                    if prevpart and prevpart[-1]==ch:
+                        break
                 else:
-                    # It doesn't fit, but we're done.  Before pushing a new
-                    # line, watch out for the current line containing only
-                    # whitespace.
-                    holding.pop()
-                    if self._current_line.is_onlyws() and holding.is_onlyws():
-                        # Don't start a new line.
-                        holding.push(part)
-                        part = None
-                    self._current_line.push(str(holding))
-                    self._lines.append(str(self._current_line))
-                    if part is None:
-                        self._current_line.reset()
-                    else:
-                        holding.reset(part)
-                        self._current_line.reset(str(holding))
-                return
-            elif not nextpart:
-                # There must be some trailing or duplicated split characters
-                # because we
-                # found a split character but no next part.  In this case we
-                # must treat the thing to fit as the part + splitpart because
-                # if splitpart is whitespace it's not allowed to be the only
-                # thing on the line, and if it's not whitespace we must split
-                # after the syntactic break.
-                holding_prelen = len(holding)
-                holding.push(part + splitpart)
-                if len(holding) + len(self._current_line) <= self._maxlen:
-                    self._current_line.push(str(holding))
-                elif holding_prelen == 0:
-                    # This is the only chunk left so it has to go on the
-                    # current line.
-                    self._current_line.push(str(holding))
-                else:
-                    save_part = holding.pop()
-                    self._current_line.push(str(holding))
-                    self._lines.append(str(self._current_line))
-                    holding.reset(save_part)
-                    self._current_line.reset(str(holding))
-                holding.reset()
-            elif not part:
-                # We're leading with a split character.  See if the splitpart
-                # and nextpart fits on the current line.
-                holding.push(splitpart + nextpart)
-                holding_len = len(holding)
-                # We know we're not leaving the nextpart on the stack.
-                holding.pop()
-                if holding_len + len(self._current_line) <= self._maxlen:
-                    holding.push(splitpart)
-                else:
-                    # It doesn't fit.  Since there's no current part really
-                    # the best we can do is start a new line and push the
-                    # split part onto it.
-                    self._current_line.push(str(holding))
-                    holding.reset()
-                    if len(self._current_line) > 0 and self._lines:
-                        self._lines.append(str(self._current_line))
-                        self._current_line.reset()
-                    holding.push(splitpart)
+                    continue
+                break
             else:
-                # All three parts are present.  First let's see if all three
-                # parts will fit on the current line.  If so, we don't need to
-                # split it.
-                holding.push(part + splitpart + nextpart)
-                holding_len = len(holding)
-                # Pop the part because we'll push nextpart on the next
-                # iteration through the loop.
-                holding.pop()
-                if holding_len + len(self._current_line) <= self._maxlen:
-                    holding.push(part + splitpart)
-                else:
-                    # The entire thing doesn't fit.  See if we need to split
-                    # before or after the split characters.
-                    if splitpart.isspace():
-                        # Split before whitespace.  Remember that the
-                        # whitespace becomes the continuation whitespace of
-                        # the next line so it goes to current_line not holding.
-                        holding.push(part)
-                        self._current_line.push(str(holding))
-                        holding.reset()
-                        self._lines.append(str(self._current_line))
-                        self._current_line.reset(splitpart)
-                    else:
-                        # Split after non-whitespace.  The continuation
-                        # whitespace comes from the instance variable.
-                        holding.push(part + splitpart)
-                        self._current_line.push(str(holding))
-                        holding.reset()
-                        self._lines.append(str(self._current_line))
-                        if nextpart[0].isspace():
-                            self._current_line.reset()
-                        else:
-                            self._current_line.reset(self._continuation_ws)
-        # Get the last of the holding part
-        self._current_line.push(str(holding))
+                fws, part = self._current_line.pop()
+                if self._current_line._initial_size > 0:
+                    # There will be a header, so leave it on a line by itself.
+                    self.newline()
+                    if not fws:
+                        # We don't use continuation_ws here because the whitespace
+                        # after a header should always be a space.
+                        fws = ' '
+                self._current_line.push(fws, part)
+                return
+            remainder = self._current_line.pop_from(i)
+            self._lines.append(str(self._current_line))
+            self._current_line.reset(remainder)
 
 
-
-def _spliterator(character, string):
-    parts = list(reversed(re.split('(%s)' % character, string)))
-    while parts:
-        part = parts.pop()
-        splitparts = (parts.pop() if parts else None)
-        nextpart = (parts.pop() if parts else None)
-        yield (part, splitparts, nextpart)
-        if nextpart is not None:
-            parts.append(nextpart)
+class _Accumulator(list):
 
-
-class _Accumulator:
     def __init__(self, initial_size=0):
         self._initial_size = initial_size
-        self._current = []
+        super().__init__()
 
-    def push(self, string):
-        self._current.append(string)
+    def push(self, fws, string):
+        self.append((fws, string))
+
+    def pop_from(self, i=0):
+        popped = self[i:]
+        self[i:] = []
+        return popped
 
     def pop(self):
-        if not self._current:
-            return None
-        return self._current.pop()
+        if self.part_count()==0:
+            return ('', '')
+        return super().pop()
 
     def __len__(self):
-        return sum(((1 if string is None else len(string))
-                    for string in self._current),
+        return sum((len(fws)+len(part) for fws, part in self),
                    self._initial_size)
 
     def __str__(self):
-        if self._current and self._current[-1] is None:
-            self._current.pop()
-        return EMPTYSTRING.join((' ' if string is None else string)
-                                for string in self._current)
+        return EMPTYSTRING.join((EMPTYSTRING.join((fws, part))
+                                for fws, part in self))
 
-    def reset(self, string=None):
-        self._current = []
+    def reset(self, startval=None):
+        if startval is None:
+            startval = []
+        self[:] = startval
         self._initial_size = 0
-        if string is not None:
-            self.push(string)
 
     def is_onlyws(self):
-        return len(self) == 0 or str(self).isspace()
+        return self._initial_size==0 and (not self or str(self).isspace())
+
+    def part_count(self):
+        return super().__len__()
diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
index 51fb229..281a65a 100644
--- a/Lib/email/test/test_email.py
+++ b/Lib/email/test/test_email.py
@@ -660,6 +660,9 @@
 
 # Test long header wrapping
 class TestLongHeaders(TestEmailBase):
+
+    maxDiff = None
+
     def test_split_long_continuation(self):
         eq = self.ndiffAssertEqual
         msg = email.message_from_string("""\
@@ -868,14 +871,12 @@
         eq = self.ndiffAssertEqual
         h = Header('; '
             'this_part_does_not_fit_within_maxlinelen_and_thus_should_'
-            'be_on_a_line_all_by_itself;')
+            'be_on_a_line_all_by_itself; ')
         eq(h.encode(), """\
 ;
- this_part_does_not_fit_within_maxlinelen_and_thus_should_be_on_a_line_all_by_itself;""")
+ this_part_does_not_fit_within_maxlinelen_and_thus_should_be_on_a_line_all_by_itself; """)
 
     def test_long_header_with_multiple_sequential_split_chars(self):
-        # Issue 11492
-
         eq = self.ndiffAssertEqual
         h = Header('This is a long line that has two whitespaces  in a row.  '
             'This used to cause truncation of the header when folded')
@@ -883,6 +884,105 @@
 This is a long line that has two whitespaces  in a row.  This used to cause
  truncation of the header when folded""")
 
+    def test_splitter_split_on_punctuation_only_if_fws(self):
+        eq = self.ndiffAssertEqual
+        h = Header('thisverylongheaderhas;semicolons;and,commas,but'
+            'they;arenotlegal;fold,points')
+        eq(h.encode(), "thisverylongheaderhas;semicolons;and,commas,butthey;"
+                        "arenotlegal;fold,points")
+
+    def test_leading_splittable_in_the_middle_just_before_overlong_last_part(self):
+        eq = self.ndiffAssertEqual
+        h = Header('this is a  test where we need to have more than one line '
+            'before; our final line that is just too big to fit;; '
+            'this_part_does_not_fit_within_maxlinelen_and_thus_should_'
+            'be_on_a_line_all_by_itself;')
+        eq(h.encode(), """\
+this is a  test where we need to have more than one line before;
+ our final line that is just too big to fit;;
+ this_part_does_not_fit_within_maxlinelen_and_thus_should_be_on_a_line_all_by_itself;""")
+
+    def test_overlong_last_part_followed_by_split_point(self):
+        eq = self.ndiffAssertEqual
+        h = Header('this_part_does_not_fit_within_maxlinelen_and_thus_should_'
+            'be_on_a_line_all_by_itself ')
+        eq(h.encode(), "this_part_does_not_fit_within_maxlinelen_and_thus_"
+                        "should_be_on_a_line_all_by_itself ")
+
+    def test_multiline_with_overlong_parts_separated_by_two_split_points(self):
+        eq = self.ndiffAssertEqual
+        h = Header('this_is_a__test_where_we_need_to_have_more_than_one_line_'
+            'before_our_final_line_; ; '
+            'this_part_does_not_fit_within_maxlinelen_and_thus_should_'
+            'be_on_a_line_all_by_itself; ')
+        eq(h.encode(), """\
+this_is_a__test_where_we_need_to_have_more_than_one_line_before_our_final_line_;
+ ;
+ this_part_does_not_fit_within_maxlinelen_and_thus_should_be_on_a_line_all_by_itself; """)
+
+    def test_multiline_with_overlong_last_part_followed_by_split_point(self):
+        eq = self.ndiffAssertEqual
+        h = Header('this is a test where we need to have more than one line '
+            'before our final line; ; '
+            'this_part_does_not_fit_within_maxlinelen_and_thus_should_'
+            'be_on_a_line_all_by_itself; ')
+        eq(h.encode(), """\
+this is a test where we need to have more than one line before our final line;
+ ;
+ this_part_does_not_fit_within_maxlinelen_and_thus_should_be_on_a_line_all_by_itself; """)
+
+    def test_long_header_with_whitespace_runs(self):
+        eq = self.ndiffAssertEqual
+        msg = Message()
+        msg['From'] = 'test@dom.ain'
+        msg['References'] = SPACE.join(['<foo@dom.ain>  '] * 10)
+        msg.set_payload('Test')
+        sfp = StringIO()
+        g = Generator(sfp)
+        g.flatten(msg)
+        eq(sfp.getvalue(), """\
+From: test@dom.ain
+References: <foo@dom.ain>   <foo@dom.ain>   <foo@dom.ain>   <foo@dom.ain>
+   <foo@dom.ain>   <foo@dom.ain>   <foo@dom.ain>   <foo@dom.ain>
+   <foo@dom.ain>   <foo@dom.ain>\x20\x20
+
+Test""")
+
+    def test_long_run_with_semi_header_splitter(self):
+        eq = self.ndiffAssertEqual
+        msg = Message()
+        msg['From'] = 'test@dom.ain'
+        msg['References'] = SPACE.join(['<foo@dom.ain>'] * 10) + '; abc'
+        msg.set_payload('Test')
+        sfp = StringIO()
+        g = Generator(sfp)
+        g.flatten(msg)
+        eq(sfp.getvalue(), """\
+From: test@dom.ain
+References: <foo@dom.ain> <foo@dom.ain> <foo@dom.ain> <foo@dom.ain>
+ <foo@dom.ain> <foo@dom.ain> <foo@dom.ain> <foo@dom.ain> <foo@dom.ain>
+ <foo@dom.ain>; abc
+
+Test""")
+
+    def test_splitter_split_on_punctuation_only_if_fws(self):
+        eq = self.ndiffAssertEqual
+        msg = Message()
+        msg['From'] = 'test@dom.ain'
+        msg['References'] = ('thisverylongheaderhas;semicolons;and,commas,but'
+            'they;arenotlegal;fold,points')
+        msg.set_payload('Test')
+        sfp = StringIO()
+        g = Generator(sfp)
+        g.flatten(msg)
+        # XXX the space after the header should not be there.
+        eq(sfp.getvalue(), """\
+From: test@dom.ain
+References:\x20
+ thisverylongheaderhas;semicolons;and,commas,butthey;arenotlegal;fold,points
+
+Test""")
+
     def test_no_split_long_header(self):
         eq = self.ndiffAssertEqual
         hstr = 'References: ' + 'x' * 80
@@ -973,7 +1073,7 @@
     def test_long_to_header(self):
         eq = self.ndiffAssertEqual
         to = ('"Someone Test #A" <someone@eecs.umich.edu>,'
-              '<someone@eecs.umich.edu>,'
+              '<someone@eecs.umich.edu>, '
               '"Someone Test #B" <someone@umich.edu>, '
               '"Someone Test #C" <someone@eecs.umich.edu>, '
               '"Someone Test #D" <someone@eecs.umich.edu>')
@@ -1028,9 +1128,11 @@
         msg['Received-2'] = h
         # This should be splitting on spaces not semicolons.
         self.ndiffAssertEqual(msg.as_string(maxheaderlen=78), """\
-Received-1: from FOO.TLD (vizworld.acl.foo.tld [123.452.678.9]) by hrothgar.la.mastaler.com (tmda-ofmipd) with ESMTP;
+Received-1: from FOO.TLD (vizworld.acl.foo.tld [123.452.678.9]) by
+ hrothgar.la.mastaler.com (tmda-ofmipd) with ESMTP;
  Wed, 05 Mar 2003 18:10:18 -0700
-Received-2: from FOO.TLD (vizworld.acl.foo.tld [123.452.678.9]) by hrothgar.la.mastaler.com (tmda-ofmipd) with ESMTP;
+Received-2: from FOO.TLD (vizworld.acl.foo.tld [123.452.678.9]) by
+ hrothgar.la.mastaler.com (tmda-ofmipd) with ESMTP;
  Wed, 05 Mar 2003 18:10:18 -0700
 
 """)
@@ -1043,12 +1145,14 @@
         msg['Received-1'] = Header(h, header_name='Received-1',
                                    continuation_ws='\t')
         msg['Received-2'] = h
-        # XXX This should be splitting on spaces not commas.
+        # XXX The space after the ':' should not be there.
         self.ndiffAssertEqual(msg.as_string(maxheaderlen=78), """\
-Received-1: <15975.17901.207240.414604@sgigritzmann1.mathematik.tu-muenchen.de> (David Bremner's message of \"Thu,
- 6 Mar 2003 13:58:21 +0100\")
-Received-2: <15975.17901.207240.414604@sgigritzmann1.mathematik.tu-muenchen.de> (David Bremner's message of \"Thu,
- 6 Mar 2003 13:58:21 +0100\")
+Received-1:\x20
+ <15975.17901.207240.414604@sgigritzmann1.mathematik.tu-muenchen.de> (David
+ Bremner's message of \"Thu, 6 Mar 2003 13:58:21 +0100\")
+Received-2:\x20
+ <15975.17901.207240.414604@sgigritzmann1.mathematik.tu-muenchen.de> (David
+ Bremner's message of \"Thu, 6 Mar 2003 13:58:21 +0100\")
 
 """)
 
@@ -1060,8 +1164,9 @@
  locQDQ4zJykFBAXJfWDjAAACYUlEQVR4nF2TQY/jIAyFc6lydlG5x8Nyp1Y69wj1PN2I5gzp"""
         msg['Face-1'] = t
         msg['Face-2'] = Header(t, header_name='Face-2')
+        msg['Face-3'] = ' ' + t
         # XXX This splitting is all wrong.  It the first value line should be
-        # snug against the field name.
+        # snug against the field name or the space after the header not there.
         eq(msg.as_string(maxheaderlen=78), """\
 Face-1:\x20
  iVBORw0KGgoAAAANSUhEUgAAADAAAAAwBAMAAAClLOS0AAAAGFBMVEUAAAAkHiJeRUIcGBi9
@@ -1069,6 +1174,9 @@
 Face-2:\x20
  iVBORw0KGgoAAAANSUhEUgAAADAAAAAwBAMAAAClLOS0AAAAGFBMVEUAAAAkHiJeRUIcGBi9
  locQDQ4zJykFBAXJfWDjAAACYUlEQVR4nF2TQY/jIAyFc6lydlG5x8Nyp1Y69wj1PN2I5gzp
+Face-3:\x20
+ iVBORw0KGgoAAAANSUhEUgAAADAAAAAwBAMAAAClLOS0AAAAGFBMVEUAAAAkHiJeRUIcGBi9
+ locQDQ4zJykFBAXJfWDjAAACYUlEQVR4nF2TQY/jIAyFc6lydlG5x8Nyp1Y69wj1PN2I5gzp
 
 """)
 
@@ -1080,8 +1188,8 @@
              'Wed, 16 Oct 2002 07:41:11 -0700')
         msg = email.message_from_string(m)
         eq(msg.as_string(maxheaderlen=78), '''\
-Received: from siimage.com ([172.25.1.3]) by zima.siliconimage.com with Microsoft SMTPSVC(5.0.2195.4905);
- Wed, 16 Oct 2002 07:41:11 -0700
+Received: from siimage.com ([172.25.1.3]) by zima.siliconimage.com with
+ Microsoft SMTPSVC(5.0.2195.4905); Wed, 16 Oct 2002 07:41:11 -0700
 
 ''')
 
@@ -1095,9 +1203,11 @@
         msg['List'] = h
         msg['List'] = Header(h, header_name='List')
         eq(msg.as_string(maxheaderlen=78), """\
-List: List-Unsubscribe: <http://lists.sourceforge.net/lists/listinfo/spamassassin-talk>,
+List: List-Unsubscribe:
+ <http://lists.sourceforge.net/lists/listinfo/spamassassin-talk>,
         <mailto:spamassassin-talk-request@lists.sourceforge.net?subject=unsubscribe>
-List: List-Unsubscribe: <http://lists.sourceforge.net/lists/listinfo/spamassassin-talk>,
+List: List-Unsubscribe:
+ <http://lists.sourceforge.net/lists/listinfo/spamassassin-talk>,
         <mailto:spamassassin-talk-request@lists.sourceforge.net?subject=unsubscribe>
 
 """)
@@ -4113,6 +4223,11 @@
         msg = email.message_from_string("EmptyHeader:")
         self.assertEqual(str(msg), "EmptyHeader: \n\n")
 
+    def test_encode_preserves_leading_ws_on_value(self):
+        msg = Message()
+        msg['SomeHeader'] = '   value with leading ws'
+        self.assertEqual(str(msg), "SomeHeader:    value with leading ws\n\n")
+
 
 
 # Test RFC 2231 header parameters (en/de)coding