RFC 2822 describes the characters allowed in a header field name. Conform to

warsaw · warsaw · commit f4c7c402d494 · 2004-11-29T03:46:54.000Z
this, and add test cases.
diff --git a/Lib/email/FeedParser.py b/Lib/email/FeedParser.py
@@ -27,7 +27,9 @@
 NLCRE_bol = re.compile('(\r\n|\r|\n)')
 NLCRE_eol = re.compile('(\r\n|\r|\n)$')
 NLCRE_crack = re.compile('(\r\n|\r|\n)')
-headerRE = re.compile(r'^(From |[-\w]{2,}:|[\t ])')
+# RFC 2822 $3.6.8 Optional fields.  ftext is %d33-57 / %d59-126, Any character
+# except controls, SP, and ":".
+headerRE = re.compile(r'^(From |[\041-\071\073-\176]{2,}:|[\t ])')
 EMPTYSTRING = ''
 NL = '\n'
 
diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
@@ -2402,6 +2402,22 @@ def test_strip_line_feed_and_carriage_return_in_headers(self):
         eq(msg.get('Header'), value1)
         eq(msg.get('Next-Header'), value2)
 
+    def test_rfc2822_header_syntax(self):
+        eq = self.assertEqual
+        m = '>From: foo\nFrom: bar\n!"#QUX;~: zoo\n\nbody'
+        msg = email.message_from_string(m)
+        eq(len(msg.keys()), 3)
+        keys = msg.keys()
+        keys.sort()
+        eq(keys, ['!"#QUX;~', '>From', 'From'])
+        eq(msg.get_payload(), 'body')
+
+    def test_rfc2822_space_not_allowed_in_header(self):
+        eq = self.assertEqual
+        m = '>From foo@example.com 11:25:53\nFrom: bar\n!"#QUX;~: zoo\n\nbody'
+        msg = email.message_from_string(m)
+        eq(len(msg.keys()), 0)
+
 
 
 class TestBase64(unittest.TestCase):
