summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
...
* pcre: Make the start line regexp even more liberal. Seems like any ofbwarsaw2002-10-211-1/+1
| | | | spaces or tabs can appear at the beginning of the line.
* main(): If we get an exception (any exception) while unshunting abwarsaw2002-10-211-4/+10
| | | | message, log the error and continue. Suggested by Peer Heinlein.
* maybe_forward(): Improve the grammatically anti-good message sent withbwarsaw2002-10-211-3/+3
| | | | | a forwarded bounce message. Complained upon by Daniel Buchmann, improved upon by Greg Ward. Further mungulated by Barry.
* GetConfigInfo(): Quick patch for invalid html by Daniel Buchmann.bwarsaw2002-10-211-2/+3
|
* Norwegian catalog update. Not complete yet, just submitting my latest changes.avalon2002-10-202-13/+72
|
* updated to last mail<->news gateway changespioppo2002-10-202-44/+54
|
* More catalog updatesbwarsaw2002-10-2029-3805/+5114
|
* A bunch of updates.bwarsaw2002-10-201-0/+6
|
* Some help for running Mailman on MacOSX.bwarsaw2002-10-200-0/+0
|
* intermediatebwarsaw2002-10-201-0/+31
|
* Create(): include the full posting address in the error message forbwarsaw2002-10-201-2/+3
| | | | when the posting address is invalid.
* dolist(): Add a check for consistency between the bounce_info andbwarsaw2002-10-201-5/+28
| | | | | | | | | | | | | delivery_status, as necessary for cron/disabled. Some older betas (alphas?) could have a situation where an address is disabled BYBOUNCE but there was no bounce information available. The "fix" is to re-enable such addresses. There should be almost none that fall into this category, but I think it's fine to re-enable them and let the normal bounce processing re-disable them in the normal way. Also, some minor formatting change and whitespace normalization.
* __getstate__(), __setstate__(): Get rid of the lock object attributebwarsaw2002-10-201-2/+13
| | | | on both pickling and unpickling, for legacy archive pickles.
* prefix_subject(): The charsets coming from headerbits can be stringsbwarsaw2002-10-201-0/+2
| | | | in some cases; coerce them to Charset instances.
* checkdata(): This should probably check request.db.tmp too.bwarsaw2002-10-201-1/+2
|
* __getstate__(self): Never pickle the lock object! I believe thisbwarsaw2002-10-191-0/+6
| | | | fixes SF bug # 625516, but more testing needs to happen.
* Get rid of all the manipulations of the MailList object's lock file.bwarsaw2002-10-191-6/+2
| | | | HyperArch should have nothing to do with it.
* __writelog(): It turns out to be really helpful to print the stackbwarsaw2002-10-191-1/+4
| | | | trace when we do logging.
* prepare_message(): If the newsgroup is moderated, news_moderation is 1bwarsaw2002-10-191-1/+7
| | | | | | | or 2, and we remove any existing Approved header, replacing it for one of our own, pointing to the posting address. Also, a pychecker fix.
* ModeratedNewsgroup: New hold class.bwarsaw2002-10-191-5/+12
| | | | | | | | process(): If news_moderation == 2, then always hold the message for moderation, unless of course the message was pre-approved, via an Approved header. Also, whitespace normalization and some pychecker fixes.
* GetConfigInfo(): Added news_moderation variable, which is used tobwarsaw2002-10-191-12/+52
| | | | | select the moderation policy of the gated-to newsgroup. Also, re-arranged and cleaned up the interface by adding section headers.
* add_options_table_item(), get_item_gui_value(): Pass the `extra' bitbwarsaw2002-10-191-12/+15
| | | | | | | | of information in the gui item description tuple to get_item_gui_value(). This latter is usually ignored, but for radio buttons, can be a flag specifying whether horizontal or vertical buttons should be used. The default is 0 for backwards compatibility, signifying horizontal radio buttons.
* NewVars(): Pick up news_moderation.bwarsaw2002-10-191-0/+1
|
* Whitespace normalization.bwarsaw2002-10-191-3/+3
|
* DATA_FILE_VERSION: Bump to pick up news_moderation.bwarsaw2002-10-191-1/+1
|
* InitVars(): New config variable news_moderation, which is a 3-way flagbwarsaw2002-10-191-3/+6
| | | | | | | | | | | | | | | | | controlling how the mail->news gateway should handle moderation. This supplants MM2.0's SF patch #401270 in a much simpler way. Possible values are: - 0 means we're not gating to a moderated newsgroup - 1 means "open list, moderated group", essentially we just add an Approved header to the gated message, but do nothing else special - 2 means "moderated" which adds an additional hold criteria to the message. Once the message is approved, it gets an Approved header and is gated. Note: this was a top request by the FSF, and essential for python.org. I had a brainstorm about how to make this fairly simple to add so we're thawing the release for this one.
* Added Greg's distillation and variation on my riff about upgradingbwarsaw2002-10-191-6/+107
| | | | | | | | | lists one at a time. I did a little bit of reformatting, preferring one line Apache rewrite rules where possible. I hope I didn't break anything. ;) We'll give his cool ideas about sharing url space a try when we upgrade python.org to MM2.1b4.
* Whitespace normalization.bwarsaw2002-10-191-4/+4
|
* SendSubscribeAck(), SendUnsubscribeAck(), MailUserPassword(): Sincebwarsaw2002-10-181-3/+3
| | | | | these messages are destined for one recipient anyway, we'll turn on VERP if VERP_PERSONALIZED_DELIVERIES is enabled.
* Add a work around for the default small stack size on MacOSX.bwarsaw2002-10-181-0/+19
| | | | | Apparently pipermail tickles a well-known bug on OSX related to deeply recursive regular expressions.
* Another update from Gregbwarsaw2002-10-181-9/+2
|
* Whitespace normalizationbwarsaw2002-10-181-3/+3
|
* A bouncer that recognizes (some) MS Exchange bounces.bwarsaw2002-10-180-0/+0
|
* intermediatebwarsaw2002-10-181-0/+47
|
* DATA: Added tests for MS Exchange.bwarsaw2002-10-181-4/+7
| | | | Also, whitespace normalization.
* Whitespace normalization and pychecker fixes.bwarsaw2002-10-181-5/+7
|
* ScanMessages(): Added the Exchange (MS Exchange) recognizer.bwarsaw2002-10-181-3/+4
| | | | Also, whitespace normalization.
* Another MS Exchange sample bounce.bwarsaw2002-10-181-0/+119
|
* An MS Exchange sample bounce.bwarsaw2002-10-181-0/+108
|
* Updated Spanish catalogs from Juan Carlos Rey Anaya. Also committingbwarsaw2002-10-183-222/+63
| | | | the .mo version of the Russian catalog.
* send_digests(): Tokio Kikuchi patch of a bug reported by Chuq thatbwarsaw2002-10-181-5/+5
| | | | | caused a daily ratchet of the volume number. Also, one minor refactoring in the bump digest number conditional.
* One more patch from Greg.bwarsaw2002-10-171-2/+2
|
* FakeDoc.addError(): Make the API really equivalent tobwarsaw2002-10-171-3/+5
| | | | Document.addError()
* prefix_subject(): Do byte string -> unicode string conversionbwarsaw2002-10-171-3/+5
| | | | | correctly, i.e. through the input_codec of the charset, not using the charset directly.
* Nigel's suggested tweaks, c.o. Greg.bwarsaw2002-10-171-43/+30
|
* process(): Two important fixes. First, when adding Cc headers tobwarsaw2002-10-161-16/+26
| | | | | | | | | | personalized messages, be RFC 2822 compliant in that a maximum of one Cc header is allowed. Fortunately we can use the same algorithm as we use for munging Reply-To. Second, make sure the Cc header adding goes in the "if not fasttrack" section so stuff like password reminders and confirmation messages don't get a Cc to the list tacked on. :-o
* Greg Ward and I trade spell checks. :)bwarsaw2002-10-151-1/+1
|
* Greg Ward's rewrite, with spell checking and a few minor updates bybwarsaw2002-10-151-237/+317
| | | | Barry.
* open_newsgroup(): SF bug # 620033 indicates that an IOError can bebwarsaw2002-10-151-5/+5
| | | | | | raised in the NNTP() constructor. Just catch it, log it, and move on. Also, whitespace normalization.
* ForwardMessage(): Typo. OwnerNotification's don't take a langbwarsaw2002-10-151-1/+0
| | | | | argument (since they can calculate that from the mlist argument). Found by Mentor Cana.
generated by cgit v1.3.1 (git 2.54.0) at 2026-08-10 00:39:37 +0000