| Commit message (Collapse) | Author | Age | Files | Lines |
| ... | |
| |
|
|
|
| |
Fix a buglet where if maybe_send_digest_now() was called before a digest has
been prepared, a FileNotFoundError would be raised.
|
| |
|
|
|
|
|
|
|
| |
their own module inside the mailman.app package. With the latter, remove the
"all lists" functionality and require the mlist argument. We'll handle the
"all lists" use case higher up the stack.
Also, rename the send-digests handler digests since we'll next add the bump
functionality.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Add a `mailman send-digests` subcommand which replaces the functionality of
the MM2.1 senddigests.py cronjob.
* Use mlist.data_path where appropriate instead of crafting it from
config.LIST_DATA_DIR. This makes it more consistent to switch to using the
list-id as the data subdirectory.
* Refactor the to_digest handler so that we can implement
maybe_send_digest_now() for the internal API.
* Fix some typos in subcommand --help summaries.
|
| |
|
|
|
|
| |
This reverts commit 8c471e067c4b31064c2f69c0afd048a7e756a429.
The fix wasn't correct. Now we have a test for the expected behavior.
|
| |
|
|
| |
list is set to confirm-then-moderate. (Closes #114)
|
| | |
|
| |
|
|
|
|
| |
3.0 except that UUIDs are represented as hex strings instead of 128-bit
integers, since the latter are not compatible with all versions of
JavaScript.
|
| |
|
|
|
|
| |
`X-Message-ID-Hash` although the latter is still included for backward
compatibility. Also be sure that all places which add the header use the
same algorithm.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
* Make sure Registrar.discard() removces any workflow state manager state
associated with the token, and that this is flushed to SA.
* Adjust the email commands to the new IRegistrar API.
* Update the IRegistrar interface.
* Add IWorkflowStateManager.discard() and make `count` an attribute/property.
* Mark two tests as expected failures due to LP: #1444184.
|
| | |
|
| |\ |
|
| | | |
|
| | |\ |
|
| | | | |
|
| |\ \ \
| |/ /
|/| |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Backward Incompatible REST API Changes:
* The JSON representation for subscription holds now no longer contains the
`password` key. Also, the `address` key has been renamed to `email` for
consistency with established terminology and other usage.
Other Internal API Changes:
* IUserManager has grown a `make_user()` method.
Refactorings:
* Most uses in the test suite of add_member() have been replaced with a new
version of the subscribe() helper. This reduces the surface area of this
ancient internal API. Eventually add_member() will have to go away or
significantly change with the subscription policy workflow.
* hold_subscription() as well as the remaining instances of add_member() now
use a namedtuple in their arguments, to keep the signatures manageable.
|
| | | | |
|
| | |/
| |
| |
| |
| |
| | |
* RequestRecord contains no password key so these are not part of the held
requests database any more.
* Pending record contains `email` now instead of `address`.
|
| |\ \ |
|
| |/ / |
|
| |\ \
| |/
|/| |
|
| |/| |
|
| | | |
|
| |\ \
| | |
| | |
| | | |
by Manish Gill. (LP: #1166911)
|
| | | | |
|
| |\ \ \
| |_|/
|/| | |
|
| |/ / |
|
| | |
| |
| |
| | |
is created by ``mailman start``. (LP: #1411435)
|
| | | |
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
file's ``[paths.*]`` sections. A new ``[paths.here]`` section is added,
which puts the ``var_dir`` in ``$cwd``. It is made the default layout.
* You can now view the contents of, inject messages into, and delete messages
from the various queue directories via the ``<api>/queues`` resource.
Also:
inject_message() and inject_text() now return the filebase of the file
injected into the queue directory.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| | |
`mailman` foreground command, through the `master` command and to the `runner`
command. Specifically, to ensure that all agree on $VAR_DIR, we ahve to set
MAILMAN_VAR_DIR in the environment that `runner` gets.
Also, so that all agree on the same configuration, always pass -C pointing
explicitly to the cfg file that the `mailman` command sees.
|
| | | |
|
| |\ \ |
|
| | |\ \
| | | |
| | | |
| | | | |
Aurélien Bompard. (LP: #1060951)
|
| | | | | |
|
| | |/ /
| | |
| | |
| | | |
Fixes: bug #1060951
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| | | | |
|
| |/ /
| |
| |
| |
| |
| |
| |
| |
| |
| | |
By using `six` I think I have most of the imports squared away. There's
probably still uses of `unicode` built-ins that need fixing.
The idea is to first get the test suite running (which it doesn't yet), and
then to fix tests.
There's a bug in lazr.config which requires us to patch it for now.
|
| | | |
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
To do this, we have to handle random test ordering, since tox explicitly sets
PYTHONHASHSEED. That's a good thing for the future Python 3 port.
Removed `mailman conf -t/--sort`; now the output is always sorted.
RFC 2369 headers are now sorted before being added.
etag repr dicts are sorted using pprint.pformat().
|
| | | |
|
| | | |
|