Two changes:

- All constructors grow an optional argument `factory' which is a
  callable used when new message instances are created by the next()
  methods.  Defaults to the rfc822.Message class.

- A new subclass of UnixMailbox is added, called PortableUnixMailbox.
  It's identical to UnixMailbox, but uses a more portable test for
  From_ delimiter lines.  With PortableUnixMailbox, any line that
  starts with "From " is considered a delimiter (this should really
  check for two newlines before the F, but it doesn't.

1 files changed, 43 insertions, 12 deletions

diff --git a/Lib/mailbox.py b/Lib/mailbox.py
index a565948..2f96106 100755
--- a/Lib/mailbox.py
+++ b/Lib/mailbox.py
@@ -9,9 +9,10 @@ import os
 __all__ = ["UnixMailbox","MmdfMailbox","MHMailbox","Maildir","BabylMailbox"]
 
 class _Mailbox:
-    def __init__(self, fp):
+    def __init__(self, fp, factory=rfc822.Message):
         self.fp = fp
         self.seekp = 0
+        self.factory = factory
 
     def seek(self, pos, whence=0):
         if whence==1:           # Relative to current position
@@ -34,7 +35,7 @@ class _Mailbox:
             self.seekp = stop = self.fp.tell()
             if start != stop:
                 break
-        return rfc822.Message(_Subfile(self.fp, start, stop))
+        return self.factory(_Subfile(self.fp, start, stop))
 
 
 class _Subfile:
@@ -117,22 +118,50 @@ class UnixMailbox(_Mailbox):
                 self.fp.seek(pos)
                 return
 
-    # An overridable mechanism to test for From-line-ness.
-    # You can either specify a different regular expression
-    # or define a whole new _isrealfromline() method.
-    # Note that this only gets called for lines starting with
-    # the 5 characters "From ".
+    # An overridable mechanism to test for From-line-ness.  You can either
+    # specify a different regular expression or define a whole new
+    # _isrealfromline() method.  Note that this only gets called for lines
+    # starting with the 5 characters "From ".
+    #
+    # BAW: According to
+    #http://home.netscape.com/eng/mozilla/2.0/relnotes/demo/content-length.html
+    # the only portable, reliable way to find message delimiters in a BSD (i.e
+    # Unix mailbox) style folder is to search for "\n\nFrom .*\n", or at the
+    # beginning of the file, "^From .*\n".  While _fromlinepattern below seems
+    # like a good idea, in practice, there are too many variations for more
+    # strict parsing of the line to be completely accurate.
+    #
+    # _strict_isrealfromline() is the old version which tries to do stricter
+    # parsing of the From_ line.  _portable_isrealfromline() simply returns
+    # true, since it's never called if the line doesn't already start with
+    # "From ".
+    #
+    # This algorithm, and the way it interacts with _search_start() and
+    # _search_end() may not be completely correct, because it doesn't check
+    # that the two characters preceding "From " are \n\n or the beginning of
+    # the file.  Fixing this would require a more extensive rewrite than is
+    # necessary.  For convenience, we've added a StrictUnixMailbox class which
+    # uses the older, more strict _fromlinepattern regular expression.
 
     _fromlinepattern = r"From \s*[^\s]+\s+\w\w\w\s+\w\w\w\s+\d?\d\s+" \
                        r"\d?\d:\d\d(:\d\d)?(\s+[^\s]+)?\s+\d\d\d\d\s*$"
     _regexp = None
 
-    def _isrealfromline(self, line):
+    def _strict_isrealfromline(self, line):
         if not self._regexp:
             import re
             self._regexp = re.compile(self._fromlinepattern)
         return self._regexp.match(line)
 
+    def _portable_isrealfromline(self, line):
+        return 1
+
+    _isrealfromline = _strict_isrealfromline
+
+
+class PortableUnixMailbox(UnixMailbox):
+    _isrealfromline = UnixMailbox._portable_isrealfromline
+
 
 class MmdfMailbox(_Mailbox):
     def _search_start(self):
@@ -155,7 +184,7 @@ class MmdfMailbox(_Mailbox):
 
 
 class MHMailbox:
-    def __init__(self, dirname):
+    def __init__(self, dirname, factory=rfc822.Message):
         import re
         pat = re.compile('^[1-9][0-9]*$')
         self.dirname = dirname
@@ -168,6 +197,7 @@ class MHMailbox:
         # This only works in Python 1.6 or later;
         # before that str() added 'L':
         self.boxes = map(str, list)
+        self.factory = factory
 
     def next(self):
         if not self.boxes:
@@ -175,14 +205,15 @@ class MHMailbox:
         fn = self.boxes[0]
         del self.boxes[0]
         fp = open(os.path.join(self.dirname, fn))
-        return rfc822.Message(fp)
+        return self.factory(fp)
 
 
 class Maildir:
     # Qmail directory mailbox
 
-    def __init__(self, dirname):
+    def __init__(self, dirname, factory=rfc822.Message):
         self.dirname = dirname
+        self.factory = factory
 
         # check for new mail
         newdir = os.path.join(self.dirname, 'new')
@@ -202,7 +233,7 @@ class Maildir:
         fn = self.boxes[0]
         del self.boxes[0]
         fp = open(fn)
-        return rfc822.Message(fp)
+        return self.factory(fp)
 
 
 class BabylMailbox(_Mailbox):