Index: Lib/mailbox.py =================================================================== --- Lib/mailbox.py (revision 63023) +++ Lib/mailbox.py (working copy) @@ -1925,185 +1925,7 @@ _sync_flush(f) f.close() -## Start: classes from the original module (for backward compatibility). -# Note that the Maildir class, whose name is unchanged, itself offers a next() -# method for backward compatibility. - -class _Mailbox: - - def __init__(self, fp, factory=rfc822.Message): - self.fp = fp - self.seekp = 0 - self.factory = factory - - def __iter__(self): - return iter(self.next, None) - - def next(self): - while 1: - self.fp.seek(self.seekp) - try: - self._search_start() - except EOFError: - self.seekp = self.fp.tell() - return None - start = self.fp.tell() - self._search_end() - self.seekp = stop = self.fp.tell() - if start != stop: - break - return self.factory(_PartialFile(self.fp, start, stop)) - -# Recommended to use PortableUnixMailbox instead! -class UnixMailbox(_Mailbox): - - def _search_start(self): - while 1: - pos = self.fp.tell() - line = self.fp.readline() - if not line: - raise EOFError - if line[:5] == 'From ' and self._isrealfromline(line): - self.fp.seek(pos) - return - - def _search_end(self): - self.fp.readline() # Throw away header line - while 1: - pos = self.fp.tell() - line = self.fp.readline() - if not line: - return - if line[:5] == 'From ' and self._isrealfromline(line): - 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 ". - # - # 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 PortableUnixMailbox class - # which does no checking of the format of the 'From' line. - - _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*" - r"[^\s]*\s*" - "$") - _regexp = None - - def _strict_isrealfromline(self, line): - if not self._regexp: - import re - self._regexp = re.compile(self._fromlinepattern) - return self._regexp.match(line) - - def _portable_isrealfromline(self, line): - return True - - _isrealfromline = _strict_isrealfromline - - -class PortableUnixMailbox(UnixMailbox): - _isrealfromline = UnixMailbox._portable_isrealfromline - - -class MmdfMailbox(_Mailbox): - - def _search_start(self): - while 1: - line = self.fp.readline() - if not line: - raise EOFError - if line[:5] == '\001\001\001\001\n': - return - - def _search_end(self): - while 1: - pos = self.fp.tell() - line = self.fp.readline() - if not line: - return - if line == '\001\001\001\001\n': - self.fp.seek(pos) - return - - -class MHMailbox: - - def __init__(self, dirname, factory=rfc822.Message): - import re - pat = re.compile('^[1-9][0-9]*$') - self.dirname = dirname - # the three following lines could be combined into: - # list = map(long, filter(pat.match, os.listdir(self.dirname))) - list = os.listdir(self.dirname) - list = filter(pat.match, list) - list = map(int, list) - list.sort() - # This only works in Python 1.6 or later; - # before that str() added 'L': - self.boxes = map(str, list) - self.boxes.reverse() - self.factory = factory - - def __iter__(self): - return iter(self.next, None) - - def next(self): - if not self.boxes: - return None - fn = self.boxes.pop() - fp = open(os.path.join(self.dirname, fn), newline='') - msg = self.factory(fp) - try: - msg._mh_msgno = fn - except (AttributeError, TypeError): - pass - return msg - - -class BabylMailbox(_Mailbox): - - def _search_start(self): - while 1: - line = self.fp.readline() - if not line: - raise EOFError - if line == '*** EOOH ***\n': - return - - def _search_end(self): - while 1: - pos = self.fp.tell() - line = self.fp.readline() - if not line: - return - if line == '\037\014\n' or line == '\037': - self.fp.seek(pos) - return - -## End: classes from the original module (for backward compatibility). - - class Error(Exception): """Raised for module-specific errors.""" Index: Lib/test/test_mailbox.py =================================================================== --- Lib/test/test_mailbox.py (revision 63023) +++ Lib/test/test_mailbox.py (working copy) @@ -1768,22 +1768,9 @@ self.assert_(self.mbox.next() is None) self.assert_(self.mbox.next() is None) - def test_unix_mbox(self): - ### should be better! - import email.parser - fname = self.createMessage("cur", True) - n = 0 - for msg in mailbox.PortableUnixMailbox(open(fname), - email.parser.Parser().parse): - n += 1 - self.assertEqual(msg["subject"], "Simple Test") - # XXX Disabled until we figure out how to fix this - ##self.assertEqual(len(str(msg)), len(FROM_)+len(DUMMY_MESSAGE)) - self.assertEqual(n, 1) +## End: tests from the original module (for backward compatibility). -## End: classes from the original module (for backward compatibility). - _sample_message = """\ Return-Path: X-Original-To: gkj+person@localhost