Fixed incorrect header comparisons in compareMessages. It iterated over every header in the new message and didn't detect when the new message was missing a header that was in the reference message. I have fixed that by retrieving all header names from the new and reference (old) messages, lower casing the list and comparing them (with a fixup for the x-roundup-version header). If the list of headers isn't identical, I throw an error. Also the Content-Type mime values are now being compared as well to make sure the new message being generates matches the reference mime content type. Of course this broke some tests. A number of them were using compareMessages when they should have been using assertEqual. That was an easy fix. Also some messages when retrieved from the content property are missing trailing newlines. I think this is part of the signature removal algorithm. This may be a bug or intended. In any case I am fixing the tests to remove trailing newlines. However I have a handful of tests that are not that obvious. Two of are test_mailgw.py:testMultipartCharsetLatin1AttachFile test_mailgw.py:testMultipartCharsetUTF8AttachFile I removed a: Content-Transfer-Encoding: quoted-printable message header from the reference message in both cases. The message content-type was multipart/mixed and it had no content that was outside of a mime part. Sending a message like this using mailx and nmh resulted in an email being delivered that was missing a Content-Transfer-Encoding, so I think I am ok here. The last broken test also started as a missing Content-Transfer-Encoding header. But there was a twist: test-mailgw.py:testMultipartRFC822 had a reference copy with a mime type of text/plain which requires a Content-Transfer-Encoding. However I never looked at the mime type of the mail being generated. The generated mail was of type multipart/mixed. So I changed the reference copy to remove the Content-Transfer-Encoding and changed the reference mime type to multipart/mixed. Now all the mailgw tests are passing. With luck I haven't encoded an invalid test. Also did one minor spelling correction.

# Copyright (c) 2002 ekit.com Inc (http://www.ekit-inc.com/)
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
#   The above copyright notice and this permission notice shall be included in
#   all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.

import os, unittest, tempfile

from roundup.backends.locking import acquire_lock, release_lock

class LockingTest(unittest.TestCase):
    def setUp(self):
        self.path = tempfile.mktemp()
        open(self.path, 'w').write('hi\n')

    # XXX test disabled because it simply doesn't work on many platforms
    # (Solaris and Irix are known to fail, but Linux works)
    def xtest_basics(self):
        f = acquire_lock(self.path)
        try:
            acquire_lock(self.path, block=0)
        except:
            pass
        else:
            raise AssertionError, 'no exception'
        release_lock(f)
        f = acquire_lock(self.path)
        release_lock(f)

    def tearDown(self):
        os.remove(self.path)

# vim: set filetype=python ts=4 sw=4 et si