aboutsummaryrefslogtreecommitdiffstats
path: root/Mailman (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Introduce new attribute (collapse_alternatives) to allow HTML intkikuchi2005-07-136-7/+25
| | | | multipart/alternative message after content filtering.
* Backout last checkin. The fix was already coded in the following part. ^^;)tkikuchi2005-07-131-3/+0
|
* A part can be None type in Python2.4 if multipart/mixed without real content.tkikuchi2005-07-131-0/+3
| | | | Temporary work around to ignore such case.
* Avoid infinite loop of held message and its notifications.tkikuchi2005-07-041-5/+6
|
* Updated the link to the "Reply-To munging considered useful" essay.bwarsaw2005-07-011-4/+3
|
* Fix SF BugID 1179487:tkikuchi2005-06-261-3/+14
| | | | | | | | | Although the RFC2231 bug example in the tracker has been solved in mailman-2.1.6, there may be more cases where ToDigest.send_digests() can block regular delivery. I put the send_digests() calling part in try - except clause and leave a message in the error log if something happened in send_digests(). Daily call of cron/senddigests will notify more details to the site administrator.
* The usual post-release version bump.bwarsaw2005-05-311-3/+3
|
* Updates for 2.1.6 final.bwarsaw2005-05-301-3/+3
|
* We're a little bit more release candidate than that. :/bwarsaw2005-05-221-1/+1
|
* process(): One more bug fix, though this one should be safe. Under email 2.5,bwarsaw2005-05-221-1/+8
| | | | | | | | | | it is possible for a message to be "not is_multipart()" but to have a None payload. I know of at least one situation where this can happen: you have a Content-Type: multipart/* but there is nothing but two blank lines between the first boundary and the end boundary. Under email 3, you get a string payload but under email 2.5 you get None. Because there's nothing in such parts, they should be safe to ignore.
* Bump to 2.1.6rc4bwarsaw2005-05-141-2/+2
|
* I finally find out why re.escape() doesn't work properly. '%' should nottkikuchi2005-05-132-11/+12
| | | | | be escaped for it is the insert directive. It was my trial and error in the comments in previous version. Well, I didn't think of '++' ...
* prefix_subject(): Don't double escape the prefix_pattern.bwarsaw2005-05-131-3/+0
|
* Copyright years and whitespace normalization.bwarsaw2005-05-131-4/+3
|
* Copyright years.bwarsaw2005-05-131-1/+1
|
* strip_subject() in HyperArch.pybwarsaw2005-05-133-15/+16
| | | | | | | | | | prefix_subject() in CookHeaders.py When calling re.sub() to substitute the subject prefix, you have to escape the pattern, otherwise prefixes like [C++] cause "multiple repeat" exceptions in re. Also, whitespace normalization and copyright years updates.
* Fix UnicodeDecodeError and UnicodeEncodeError to UnicodeError.tkikuchi2005-05-082-2/+2
| | | | Both are introduced in Python2.3. UnicodeError catches both.
* Copyright years.bwarsaw2005-05-021-1/+1
|
* ChangeHTML(): Set umask to 0 while creating the langdir. Fixes SF bugbwarsaw2005-05-021-3/+7
| | | | #1190404
* Set DEFAULT_MAX_DAYS_TO_HOLD to 0 to disable it, since this is a new featurebwarsaw2005-05-021-4/+4
| | | | in a maintenance release.
* process(): For some reason, the part's payload can be None. Defend againstbwarsaw2005-05-011-6/+7
| | | | trying to call None.splitlines().
* prepare_message(): Fixed typo.bwarsaw2005-05-011-2/+2
|
* bumping to 2.1.6rc3bwarsaw2005-04-291-2/+2
|
* get_content_charset(): This is fixed in email-2.5.6 so remove it from herebwarsaw2005-04-291-19/+1
| | | | because it's better not to have redundant code.
* Add TypeError for possible error when checking charset availablity.tkikuchi2005-04-251-1/+1
|
* Bumping version 2.1.6rc2tkikuchi2005-04-221-2/+2
|
* Brushing up the override function.tkikuchi2005-04-221-4/+13
|
* get_content_charset() should return failobj on error.tkikuchi2005-04-211-0/+9
|
* RFC2231 charset is in email package.tkikuchi2005-04-211-4/+1
|
* Bump version to 2.1.6rc1tkikuchi2005-04-141-3/+3
|
* zh_CN new translation. moving to utf-8 also.tkikuchi2005-04-071-1/+1
|
* Typo in comment. Thanks Mr. Ikeda.tkikuchi2005-03-281-1/+1
|
* Username may contain non-ascii character for English-US.tkikuchi2005-03-281-3/+3
| | | | Fall back to email address.
* increment REL_SERIALtkikuchi2005-03-241-1/+1
|
* Bumping version number to 2.1.6b5tkikuchi2005-03-241-1/+1
|
* Change Big5 to UTF-8 for zh_TW.tkikuchi2005-03-231-1/+1
|
* Fix bug in patch #1032434: You should accumulate subpart headers.tkikuchi2005-03-211-6/+10
| | | | (should have went out an intermediate code) :-<
* Use of websafe() here is too severe. Only '<' is escaped.tkikuchi2005-03-101-1/+2
|
* The first header_filter_rules can be empty. Thanks Bryan Fullerton.tkikuchi2005-02-231-0/+3
|
* Bump to 2.1.6b4. Note that I rewrote the index.html page slightly, both tobwarsaw2005-02-171-2/+2
| | | | | indicate that 2.1.5 is still the current stable release and to point people at the security page.
* Anonymous list fix up: Add internal_name in the From: header.tkikuchi2005-02-171-1/+9
| | | | Log anonymize record in 'post' for admin's inspection convenience.
* Forget to add new attribute for new lists.tkikuchi2005-02-161-0/+1
|
* Request from Brad Knowles. We need subject prefix strippedtkikuchi2005-02-122-4/+18
| | | | from the posts from mailling list for gatewaying to nntp.
* adjust SERIALtkikuchi2005-02-111-1/+1
|
* bumping 2.1.6b3tkikuchi2005-02-112-2/+5
|
* Spelling and copyright years updates.bwarsaw2005-02-101-3/+3
|
* Checkin for initial workaround for directry traverse flaw in private.py.tkikuchi2005-02-101-3/+6
| | | | This is for the people who think 'CVS should be safe' and not final solution.
* sourceforge patch id=1107169, re-use member_moderation_notice ...tkikuchi2005-02-064-2/+13
| | | | | | | Instead of re-use, a new variable 'nonmember_rejection_notice' was introduced. For compatibility with my patch for subject numbering, 'data_version' was raised by 3. Sorry for the leap but I want MY user to get smooth update. (TK)
* Fix pipermail.py bug for not retainning original date in .txt archivetkikuchi2005-02-051-1/+4
| | | | | if the archive was regenerated by 'arch --wipe'. Thanks SHIGENO san for reporting bug and essential fix. Code was rewritten for simplicity.
* Remove misleading detail for respond_to_post_requests.tkikuchi2005-02-051-6/+2
|
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-27 13:40:41 +0000