summaryrefslogtreecommitdiff
path: root/src/mailman/queue
Commit message (Collapse)AuthorAgeFilesLines
* Reimplement wrap() using modern techniques, and finally get rid of Utils.py!Barry Warsaw2011-03-174-6/+5
|
* Utils.maketext() and Utils.findtext() are gone.Barry Warsaw2011-03-161-10/+10
|
* nntpsplit() is replaced by lazr.config.as_host_port().Barry Warsaw2011-02-251-2/+3
|
* Move oneline() out of Utils.py.Barry Warsaw2011-02-251-1/+2
|
* Remove some unused/untested stuff regarding password creation from Utils.py.Barry Warsaw2011-01-041-0/+1
| | | | Also, remove a obsolete test.
* Split member and nonmember moderation.Barry Warsaw2011-01-021-19/+18
| | | | | | | | | | | | | | | | * member-moderation happens at the same place in the built-in chain that the previously named moderation rule happens. nonmember-moderation happens after all the other normal moderation rules. * Handle unsubscribed nonmember posts. Other changes: * Message.senders now filters out Nones and empty strings. * Various test cleanups and simplifications. * More `address` -> `email` fixes. * Give Link class a useful repr. * Fix a potential UnboundLocalError. * Various other small changes.
* Happy New Year.Barry Warsaw2011-01-0115-15/+15
|
* Because it was just to damn confusing, rename IAddress.address toBarry Warsaw2011-01-012-4/+4
| | | | | | IAddress.email and IAddress.original_address to IAddress.original_email. From now on we'll use "address" to talk about the IAddress object and "email" to talk about the textual email address.
* * Simplify the membership.txt doctest.Barry Warsaw2010-12-313-13/+85
| | | | | | | * Add test to show that the incoming runner adds all sender addresses to the global user manager. * New doctest helper: dump_list() * Other random cleanups.
* Fairly significant change to the way member and nonmember moderation occurs.Barry Warsaw2010-12-291-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | Now, nonmembers are represented by a separate roster of IMembers, the latter which has grown a `moderation_action` enum. When that action is `defer`, then the normal processing rules apply. Anything else and the `moderation` chain is jumped to for a shortcut to moderation (which may include immediate acceptance). TODO: handle unregistered nonmembers. Details: * The member-moderation rule is renamed to just moderation, and handles both members and nonmembers (though the latter must currently be registered). * The moderation rule is moved up in the builtin chain. It is now checked after `approved`, `emergency`, and `loop`, but before the normal moderation checks. This means that nonmember postings will be (by default) held much earlier. * IMember.is_moderated is removed. * IMember.moderation_action is added. * IMailingList.default_member_moderation is removed. * IMailingList.default_member_action and IMailingList.default_nonmember_action are added. * MemberRole.nonmember is added.
* This is part 1 of the merge of Jimmy Bergman's branchBarry Warsaw2010-12-281-70/+67
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | lp:~jimmy-sigint/mailman/restapi_additional_attributes Ostensibly, this adds support for a few additional attributes through the REST API: * default_member_moderation * generic_nonmember_action * member_moderation_action * reply_goes_to_list * send_welcome_msg * welcome_msg However, I had never previously fleshed out the conversion of default_member_moderation and member_moderation_action into the MM3 way of things. That is now done. Non-member moderation still needs to be done. Specific changes: * mailman.chains.base.Chain no longer self registers * The built-in chain gets a new link for checking 'member-moderation'. If this rule matches, it jumps to the 'member-moderation' chain, which checks member_moderation_action and returns a link that jumps to the appropriate terminal chain. * Chain initialization is done by the same auto-detection as rules, handlers, etc. The one tricky thing is that abstract base classes such as Chain and TerminalChainBase can't be instantiated. For now, there's an ugly special case to skip these. * default_member_moderation is now exposed in the IMailingList interface. * Member.is_moderated gets set in the constructor from the mailing list's default_member_moderation. * The 'moderation' rule is renamed 'member-moderation'. TODO: * Work out non-member moderation * Add member_moderation_action to IMailingList * Double check tests for reply_goes_to_list, send_welcome_msg, and welcome_msg
* | Insulate the test environment from the running environment. Sadly, the testBarry Warsaw2010-12-221-2/+2
| | | | | | | | | | | | ports are still hard coded. Also, 'bin/mailman info' dumps the REST root url and credentials.
* | Upgrade from using the locknix package to using flufl.lock.Barry Warsaw2010-11-271-1/+1
|/
* Fix the obvious errors identified by pyflakes. Skip false positives (mostlyBarry Warsaw2010-10-088-13/+8
| | | | | due to _() interpolation) and a few other problems that did not have immediately obvious fixes.
* More documentation cleanups.Barry Warsaw2010-09-148-15/+59
|
* More documentation updatesBarry Warsaw2010-09-102-15/+24
|
* More documentation updatesBarry Warsaw2010-09-101-22/+23
|
* Fix an importBarry Warsaw2010-08-081-1/+1
|
* Refactor all bounce detectors.Barry Warsaw2010-08-081-3/+3
|
* checkpointingBarry Warsaw2010-08-071-5/+6
|
* No longer touch the Sender or Errors-To headers. We can no longer justifyBarry Warsaw2010-06-291-3/+1
| | | | | setting these based on RFC 5322, and modern MTAs no longer need us to do so for proper bounce processing.
* List-Post header should be retained in MIME digest messages (LP: 526143)Barry Warsaw2010-03-281-4/+6
|
* More refactoring:Barry Warsaw2010-02-251-1/+0
| | | | | | | * Rework the [webservice] section to be more useful to restish. * Get rid of the resource_type_link * Add and test some helpers. * Move the root resources to a different module.
* Refactoring the REST support by removing unnecessary stuff and moving otherBarry Warsaw2010-02-251-1/+1
| | | | | | stuff around. We no longer need APIValueError or IResolvePathNames. Also, refactor the creation of the REST server so that it could be used with other WSGI frameworks.
* Checkpointint support for different path layouts, allowing us to optionallyBarry Warsaw2010-01-271-2/+3
| | | | | | | | support the Filesystem Hierarchy Standard. This is done through the configuration file. * add a --paths/-p option to 'bin/mailman info' * PIDFILE -> PID_FILE
* Happy New Year.Barry Warsaw2010-01-0115-15/+15
|
* * Add REST interface for joining a mailing list.Barry Warsaw2009-12-281-0/+6
| | | | | | * add_member() now returns the newly created IMember. * Reorganized several exceptions and exposed them to the REST API. * Added NoSuchListError.
* IRegistrar is now a utility; it doesn't need to be adapted from an IDomain.Barry Warsaw2009-12-121-2/+1
| | | | | | | This is because registration confirmation messages must come from the mailing list that the subscription request came from. Remove IDomain.confirm_address() since this lives only on the IMailingList now.
* * Refactor the language manager off of the config object and into a utility.Barry Warsaw2009-12-102-9/+20
| | | | | | | | * Fix a few small typos in exception handlers. * Move the initialization of the Zope Component Architecture into the first initialization step. The only reason we couldn't do that previously was because the domain object referenced the config, causing a circularity problem. Refactor the Domain implementation to avoid that.
* Add a 'name' attribute to the Switchboards. This is so 'whichq' can be setBarry Warsaw2009-12-103-14/+19
| | | | for bin/mailman unshunt.
* * Make IDomainManager a utility, since the config object is global.Barry Warsaw2009-12-082-7/+10
| | | | | | | | * Give IMailingList a .domain attribute which looks up the IDomain for its .host_name. This cleans up a lot of code. * Add a test for the 'confirm' email command. * Suppress blank lines in email command responses. * Make the IDomainCollection a utility.
* Fix some test failures:Barry Warsaw2009-12-081-2/+2
| | | | | * when postfix is not installed * because the Subject values are instances, not strings
* Fix a test.Barry Warsaw2009-12-061-2/+11
|
* * Fix a test based on updated output.Barry Warsaw2009-12-064-36/+358
| | | | | | | | | * Add a stub for the -confirm email command * Add stubs for -leave and -unsubscribe * Remove the crufty (and broken) 'tojoin' 'toleave' and 'toconfirm' metadata keys for synchronizing between lmtp and the command runner. Replace this by putting the subaddress recognized by lmtp into the metadata and having the command runner look at the subaddress.
* Revert the last changes. Calculate size where needed.Barry Warsaw2009-12-051-1/+0
|
* The delivery agent requires that the message object have a .original_sizeBarry Warsaw2009-12-041-4/+8
| | | | attribute. Add this for UserNotifications.
* Fix a test and make Sphinx happy.Barry Warsaw2009-11-281-40/+40
|
* Refactor VERP calculation so that it's now in the outgoing queue instead ofBarry Warsaw2009-11-272-29/+363
| | | | | | the to-outgoing handler. Putting it in the latter means that more delivery vectors can take advantages of the same mechanisms. to-outgoing is now really simple (but is it simple enough?).
* Factor out most of the i18n subsystem and convert to using the flufl.i18nBarry Warsaw2009-11-154-10/+10
| | | | package.
* Cleanup.Barry Warsaw2009-11-044-4/+0
|
* smtp_direct.py is dead and gone.Barry Warsaw2009-11-031-8/+8
|
* Add decorations.Barry Warsaw2009-11-016-14/+14
| | | | s/recips/recipients/
* Convert bin/inject to bin/mailman injectBarry Warsaw2009-09-191-1/+0
| | | | | | Expose the database store on .store instead of ._store A few other cleanups.
* Remove the user_manager attribute from config.db and expose it as anBarry Warsaw2009-08-261-1/+5
| | | | IUserManager utility.
* Instead of using an adapter, use a utility to get the mailing list manager.Barry Warsaw2009-08-262-2/+4
|
* TweaksBarry Warsaw2009-08-241-3/+4
|
* Test updates and repair.xoBarry Warsaw2009-08-214-6/+10
|
* Expose 'transaction' as an alias for config.db in the doctest globs. ThisBarry Warsaw2009-07-213-3/+4
| | | | | | | | means commit() is now transaction.commit() in doctests. Add tests of extended domain creation, which exposes the need to unlock the database (via transaction.abort()) so that the foreground test process doesn't lock the background REST server process.
* Our first writable REST API! You can now create new domains through theBarry Warsaw2009-07-201-3/+4
| | | | | | | | | webserver. Fix the https/http urls. Add the adapter for HTTPCharsets so POSTs work properly. Nice little cargo cult from lazr.restful.
* De-u-literal-ify our doctests.Barry Warsaw2009-07-188-47/+55
|
generated by cgit v1.2.3-70-g09d2 (git 2.51.2) at 2025-11-09 16:23:34 +0000