#1194222: make parsedate always return RFC2822 four character years.

bitdancer · bitdancer · commit 219d1c8ae320 · 2010-08-25T00:45:55.000Z
Two character years are now converted to four character years using
the Posix standard rule (&lt;68 == 2000, &gt;=68==1900).  This makes the
parsed date RFC2822 compliant even if the input is not.

Patch and test by Jeffrey Finkelstein.
diff --git a/Lib/email/_parseaddr.py b/Lib/email/_parseaddr.py
@@ -107,6 +107,18 @@ def parsedate_tz(data):
         tss = int(tss)
     except ValueError:
         return None
+    # Check for a yy specified in two-digit format, then convert it to the
+    # appropriate four-digit format, according to the POSIX standard. RFC 822
+    # calls for a two-digit yy, but RFC 2822 (which obsoletes RFC 822)
+    # mandates a 4-digit yy. For more information, see the documentation for
+    # the time module.
+    if yy < 100:
+        # The year is between 1969 and 1999 (inclusive).
+        if yy > 68:
+            yy += 1900
+        # The year is between 2000 and 2068 (inclusive).
+        else:
+            yy += 2000
     tzoffset = None
     tz = tz.upper()
     if tz in _timezones:
diff --git a/Lib/email/test/test_email.py b/Lib/email/test/test_email.py
@@ -2234,6 +2234,19 @@ def test_parsedate_acceptable_to_time_functions(self):
         eq(time.localtime(t)[:6], timetup[:6])
         eq(int(time.strftime('%Y', timetup[:9])), 2003)
 
+    def test_parsedate_y2k(self):
+        """Test for parsing a date with a two-digit year.
+
+        Parsing a date with a two-digit year should return the correct
+        four-digit year. RFC822 allows two-digit years, but RFC2822 (which
+        obsoletes RFC822) requires four-digit years.
+
+        """
+        self.assertEqual(utils.parsedate_tz('25 Feb 03 13:47:26 -0800'),
+                         utils.parsedate_tz('25 Feb 2003 13:47:26 -0800'))
+        self.assertEqual(utils.parsedate_tz('25 Feb 71 13:47:26 -0800'),
+                         utils.parsedate_tz('25 Feb 1971 13:47:26 -0800'))
+
     def test_parseaddr_empty(self):
         self.assertEqual(utils.parseaddr('<>'), ('', ''))
         self.assertEqual(utils.formataddr(utils.parseaddr('<>')), '')
diff --git a/Misc/ACKS b/Misc/ACKS
@@ -260,6 +260,7 @@ Niels Ferguson
 Sebastian Fernandez
 Vincent Fiack
 Tomer Filiba
+Jeffrey Finkelstein
 Russell Finn
 Nils Fischbeck
 Frederik Fix
diff --git a/Misc/NEWS b/Misc/NEWS
@@ -126,6 +126,9 @@ Extensions
 Library
 -------
 
+- Issue #1194222: email.utils.parsedate now returns RFC2822 compliant four
+  character years even if the message contains RFC822 two character years.
+
 - Issue #8750: Fixed MutableSet's methods to correctly handle reflexive
   operations on its self, namely x -= x and x ^= x.
 
