summaryrefslogtreecommitdiff
path: root/Mailman (follow)
Commit message (Collapse)AuthorAgeFilesLines
* SUBDIRS: Add Queue subdirectory.bwarsaw2000-12-201-1/+1
|
* RedeliverMessage(): Removed.bwarsaw2000-12-201-11/+3
| | | | | | | | DeliverToUser(): Don't put DELIVERY_MODULE in the pipeline; that's now handled by the OutgoingRunner. LIST_PIPELINE: Put ToUsenet at the end of the pipeline, and add ToOutgoing to make sure the message gets put in qfiles/out.
* Rewritten to use the new multiple-queue model.bwarsaw2000-12-201-85/+32
| | | | | | | | process(): Makes a copy of the message, munges the copy's headers and requeues the copy to qfiles/news. This no longer does any forking, since handoff from Mailman->NNTPD is performed by the NewsRunner. do_child(): Get rid off.
* intermediatebwarsaw2000-12-201-0/+23
|
* process(): Get rid of the string module in favor of string methods.bwarsaw2000-12-203-12/+10
|
* intermediatebwarsaw2000-12-201-2/+2
|
* New module for handing off messages from Mailman (in the qfiles/newsbwarsaw2000-12-201-0/+52
| | | | directory) to the nntpd news server. Used only by NewsRunner.
* process(): Get rid of the string module in favor of string methods.bwarsaw2000-12-201-7/+7
| | | | | Also, use list comprehensions instead of map where appropriate. And, wrap translatable strings in _().
* process(): Get rid of the string module in favor of string methods.bwarsaw2000-12-201-3/+4
| | | | | Also, don't set the X-BeenThere: header by appending to msg.headers; use the __setitem__() interface instead.
* process(): List comprehensions are fun!bwarsaw2000-12-201-4/+2
|
* process(): Get rid of the string module, in favor of string methods.bwarsaw2000-12-201-7/+2
| | | | | filterfunc(): Get rid of this too, since we can use a list comprehension instead (which I think is just as clear).
* process(): Use augmented assignments.bwarsaw2000-12-201-1/+1
|
* process(): Use string method startswith().bwarsaw2000-12-201-3/+2
|
* Multiple queue runner classes, in a new package Mailman.Queue.bwarsaw2000-12-207-0/+513
| | | | | | | | | | | | | | | Runner is the base class providing the framework for other qrunner specializations. IncomingRunner runs through qfiles/in taking messages from the smtpd, sending them through the standard message handler pipeline and potentially writing the message to qfiles/out or qfiles/news. OutgoingRunner takes message from qfiles/out and hands them over to the MTA. NewsRunner takes message from qfiles/news and hands them over to the news server.
* _(): new function to fake out i18n for now.bwarsaw2000-12-201-0/+3
|
* qfiles directory is just a holder of all the sub-queue directories.bwarsaw2000-12-201-1/+6
| | | | | | | | | Currently there's qfiles/in -> the incoming queue (from post, mailcmd, mailowner, auto) qfiles/out -> the smtp outgoing queue (handoff mailman->mta) qfiles/bounce -> for incoming bounce processing qfiles/news -> the nntp outgoing queue (handoff mailman->nntpd)
* Enqueue(): Allow for queuing to alternative queue directories. Ifbwarsaw2000-12-201-8/+20
| | | | | | | | | | | | | | | | | newdata or kws has an item called "_whichq", this specifies the directory to queue to, defaulting to INQUEUE_DIR. The metadata "filebase" is always set to __filebase, not just when the metadata is created for the first time. When writing the .db file (i.e. containing the metadata), always write it after the .msg text file, and always write it to a temporary file, with an os.rename() shuffle. This should avoid race conditions when queuing a message to another queue because the queue runners only key off of files ending in ".db". Requeue(): New method which calls Enqueue() after zapping self.__filebase so it'll be recalculated.
* We can use Python 2.0's standard nntplib.py now.bwarsaw2000-12-201-535/+0
|
* First integration patch for Juan Carlos's translatable string changesbwarsaw2000-12-078-391/+410
|
* First integration patch of Juan Carlos's i18n string marking changes.bwarsaw2000-12-072-17/+17
|
* First round integration of Juan Carlos's translatable string markings.bwarsaw2000-12-0713-325/+343
|
* intermediatebwarsaw2000-12-071-0/+75
|
* Bumping version number to 2.1a1bwarsaw2000-12-071-4/+4
|
* write(): No longer necessary; Python 2.0's extended print syntax isbwarsaw2000-12-071-20/+0
| | | | better.
* Bumping to 2.0 final.bwarsaw2000-11-211-3/+3
|
* Bumping to rc3bwarsaw2000-11-161-2/+2
|
* ScriptURL(): Hopefully fixes the last place that not havingbwarsaw2000-11-161-1/+5
| | | | | | | DEFAULT_URL end in a slash causes problems. The default value for web_page_url argument is now None, and if it's found to be None in the body of the function, it's set to mm_cfg.DEFAULT_URL (and coerced to always end in a slash).
* process(): "Fix" the Reply-To: header munging algorithm. If mungingbwarsaw2000-11-161-5/+10
| | | | | | | is enabled, we now unconditionally munge based on the reply_goes_to_list config variable setting. If the Reply-To: header is munged, the original Reply-To: header (if any) is put on X-Reply-To:
* GetConfigInfo(): Update the details for both the reply_goes_to_listbwarsaw2000-11-161-17/+29
| | | | and reply_to_address configuration options.
* Better description of DEFAULT_REPLY_GOES_TO_LISTbwarsaw2000-11-161-1/+9
|
* CheckVersion(): Do not call versions.Update, and do not set thebwarsaw2000-11-141-3/+4
| | | | data_version attribute, unless the list is locked.
* __init__(): Fix for case where Content-Transfer-Encoding header isbwarsaw2000-11-131-2/+2
| | | | missing. Submitted by Erik Forsberg (original patch author).
* __init__(): The values of Content-Type and Content-Transfer-Encodingbwarsaw2000-11-101-3/+6
| | | | are case insensitive according to RFC 1521. Closes patch #102268.
* Bumping to rc2bwarsaw2000-11-101-2/+2
|
* Added a Useful Comment.bwarsaw2000-11-101-1/+2
|
* get_archives(): Removed the line which specifically eliminatesbwarsaw2000-11-101-1/+0
| | | | | | | | messages with a Subject: in ('subscribe', 'unsubscribe'). This kind of filtering happens at higher levels in Mailman; if such a message is in the .mbox file, it should be in the html archive. Closes bug #121811 for real now.
* _rx_quote: Change the regexp so that it only matches legitimatebwarsaw2000-11-091-1/+1
| | | | | (uppercase) hex digits. This fixes SF bug #117548 and replaces the suggested patch in patch #102097. (patch approved by Jeremy.)
* process_form(): Don't pass the `user' argument to PrintResults whenbwarsaw2000-11-091-2/+1
| | | | | | the user is successfully unsubscribed. In that case there /is/ no options for the user to continue to edit. Fixes SF bugs 122017 and 121591.
* DEFAULT_AUTOMATIC_BOUNCE_ACTION: John A. Martin noticed that thebwarsaw2000-11-091-3/+3
| | | | | comment doesn't jive with the web options. It also doesn't jive with the actual implementation, so make the comment conform to reality.
* PUBLIC_ARCHIVE_FILE_DIR, PRIVATE_ARCHIVE_FILE_DIR: slightly morebwarsaw2000-11-091-2/+2
| | | | portable definitions.
* __handlepost(): When reconstituting the message from the requestbwarsaw2000-11-081-0/+6
| | | | | | | | | database, zap the `filebase' attribute and let it be recalculated when the message is enqueued. This fixes duplication bugs when an approved message has delivery problems, and get re-enqueued continually with a different filebase. This should close SF bug #117015.
* connect(): Wrap the sock.connect() in a try/except so that if thebwarsaw2000-10-311-1/+5
| | | | | connection fails with a socket.error, the file descriptors are reclaimed.
* GetConfigInfo(): In the reply_goes_to_list details, add a link tobwarsaw2000-10-271-1/+3
| | | | Simon Hill's dissenting opinion on Reply-To munging.
* HyperArchive.html_TOC(): Dan Mick recognized an indentation bug whichbwarsaw2000-10-271-1/+1
| | | | broke the archiver.
* process(): Patch by Fil to force listinfo urls to be absolute.bwarsaw2000-10-271-1/+1
|
* Bump versions to 2.0 release candidate 1.bwarsaw2000-10-231-3/+3
|
* Comment out a couple of print statements which can cause annoying cronbwarsaw2000-10-201-2/+2
| | | | emails. These should really be log file output.
* SF Patch #101846 by Sean Reifschneider (modified by bwarsaw) to addbwarsaw2000-10-201-0/+5
| | | | MTA_ALIASES_STYLE for the newlist script's output.
* lock(): Added a comment that the __touch() right after the initialbwarsaw2000-10-121-0/+4
| | | | | | __write() can fail with an EPERM. It seems to be a very rare occurrence, only happens from cron, and (only?) on Solaris 2.6. I'm too nervous to change this for 2.0.
* __handlepost(): If the message has been approved, and is beingbwarsaw2000-10-101-2/+10
| | | | | | | | | forwarded, we need to craft two completely separate messages for each path through the system. They can't share msgdata state or message text or the first will simply disappear. Hacky way of doing this is to rewind the file pointer and re-instantiate a new message object. Closes SF bug #116311.
generated by cgit v1.2.3-70-g09d2 (git 2.52.0) at 2026-01-06 18:42:37 +0000