From tkikuchi at is.kochi-u.ac.jp Tue Aug 2 14:54:13 2005 From: tkikuchi at is.kochi-u.ac.jp (Tokio Kikuchi) Date: Tue, 02 Aug 2005 21:54:13 +0900 Subject: [Mailman-Developers] [Mailman-Users] header_filter_rules possible bug In-Reply-To: <54F3B081-F921-490C-B20D-86F95EA5A4D3@unimelb.edu.au> References: <54F3B081-F921-490C-B20D-86F95EA5A4D3@unimelb.edu.au> Message-ID: <42EF6CF5.6000107@is.kochi-u.ac.jp> Hi, I think you are right! I have to make warning when checking in because this may cause incompatible behavior for already installed other patterns. :-( Iain Pople wrote: > Hi, > > I have noticed some unexpected behaviour with header_filter_rules. > > I wanted to reject messages with an empty subject line so i came up > with the following regex: > > ^subject:\s*$ > > However it wasn't working as desired. After delving around in the > SpamDetect.py code i noticed that the re.search call is not using > MULTILINE mode. So the headers are treated as one long string and ^ > would only match the beginning of the string. > > Is this the desired behaviour? To me it makes more sense that it > should use multiline mode. > > i.e.: > > --- SpamDetect.py.bak 2005-08-02 15:56:42.083270000 +1000 > +++ SpamDetect.py 2005-08-02 17:33:59.698226000 +1000 > @@ -127,7 +127,7 @@ > for pattern in patterns.splitlines(): > if pattern.startswith('#'): > continue > - if re.search(pattern, headers, re.IGNORECASE): > + if re.search(pattern, headers, re.IGNORECASE|re.MULTILINE): > if action == mm_cfg.DISCARD: > raise Errors.DiscardMessage > if action == mm_cfg.REJECT: > @@ -137,3 +137,4 @@ > hold_for_approval(mlist, msg, msgdata, > HeaderMatchHold) > if action == mm_cfg.ACCEPT: > return > > -- > Iain Pople > Systems Programmer > University of Melbourne > > ------------------------------------------------------ > Mailman-Users mailing list > Mailman-Users at python.org > http://mail.python.org/mailman/listinfo/mailman-users > Mailman FAQ: http://www.python.org/cgi-bin/faqw-mm.py > Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ > Unsubscribe: http://mail.python.org/mailman/options/mailman-users/tkikuchi%40is.kochi-u.ac.jp > > Security Policy: http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq01.027.htp > > > -- Tokio Kikuchi, tkikuchi@ is.kochi-u.ac.jp http://weather.is.kochi-u.ac.jp/ From shaikli at yahoo.com Thu Aug 4 08:04:06 2005 From: shaikli at yahoo.com (Nadim Shaikli) Date: Wed, 3 Aug 2005 23:04:06 -0700 (PDT) Subject: [Mailman-Developers] DomainKeys support Message-ID: <20050804060406.10571.qmail@web31809.mail.mud.yahoo.com> I've noticed of late that emails sent by yahoo users that get relayed by mailman end-up without the domainkey header entry [1] and thus in various yahoo users' bulk (ie. spam) folder. Is there anything that can be done to remedy this issue ? Can a site's mailman application add a locally qualified domainkey entry header (or keep the original entry as-is) ? If so, how and/or is there a place I can read more about this (didn't see any mention in the FAQ). [1] http://antispam.yahoo.com/domainkeys Regards, - Nadim ____________________________________________________ Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs From jarrell at vt.edu Thu Aug 4 17:18:47 2005 From: jarrell at vt.edu (Ron Jarrell) Date: Thu, 4 Aug 2005 11:18:47 -0400 (EDT) Subject: [Mailman-Developers] TypeError: cannot concatenate 'str' and 'NoneType' object Message-ID: <200508041518.j74FIlIp004661@babylon5.cc.vt.edu> Well, it came back... I was poking around because I was having cron problems, and some jobs weren't running.. There's nothing in error, but if you run disabled by hand, you get: /usr/local/bin/python -S /home/mailman/cron/disabled Traceback (most recent call last): File "/home/mailman/cron/disabled", line 220, in ? main() File "/home/mailman/cron/disabled", line 204, in main mlist.sendNextNotification(member) File "/home/mailman/Mailman/Bouncer.py", line 270, in sendNextNotification msg['Subject'] = 'confirm ' + info.cookie /usr/local/bin/python -S /home/mailman/cron/disabled Traceback (most recent call last): File "/home/mailman/cron/disabled", line 220, in ? main() File "/home/mailman/cron/disabled", line 204, in main mlist.sendNextNotification(member) File "/home/mailman/Mailman/Bouncer.py", line 270, in sendNextNotification msg['Subject'] = 'confirm ' + info.cookie TypeError: cannot concatenate 'str' and 'NoneType' objects It worked fine for a while. From stephen at xemacs.org Fri Aug 5 07:49:39 2005 From: stephen at xemacs.org (Stephen J. Turnbull) Date: Fri, 05 Aug 2005 14:49:39 +0900 Subject: [Mailman-Developers] DomainKeys support In-Reply-To: <20050804060406.10571.qmail@web31809.mail.mud.yahoo.com> (Nadim Shaikli's message of "Wed, 3 Aug 2005 23:04:06 -0700 (PDT)") References: <20050804060406.10571.qmail@web31809.mail.mud.yahoo.com> Message-ID: <878xzhdjbg.fsf@tleepslib.sk.tsukuba.ac.jp> >>>>> "Nadim" == Nadim Shaikli writes: Nadim> I've noticed of late that emails sent by yahoo users that Nadim> get relayed by mailman end-up without the domainkey header Nadim> entry and thus in various yahoo users' bulk (ie. spam) Nadim> folder. Is there anything that can be done to remedy this Nadim> issue ? Can a site's mailman application add a locally Nadim> qualified domainkey entry header (or keep the original Nadim> entry as-is) ? According to the DomainKeys FAQ: How does DomainKeys work with mailing lists? Mailing lists that do not change the content or re-arrange or append headers will be DomainKey compatible with no changes required. Mailing lists that change the message and headers should re-sign the message with their own private key and claim authorship of the message. Unfortunately, standard mailing lists will change/append certain headers, breaking the signature. Specifically, Mailman does change the Sender header, which means DomainKeys can't just pass through. You need to re-sign.[1] Based on that page, AFAICT it would be a bad idea for list management software like Mailman to support DomainKeys itself[1], except that it should optionally be configured to check for DomainKeys flags from the incoming MTA, and optionally submit the mail to the DomainKeys submission service port instead of SMTP for outgoing nail. Mailman supports both of those configurations already. Then you should get an MTA that supports DomainKeys (see the DomainKeys FAQ for a list), and you'll also have to fix up your DNS to publish the keys. Note that it's unclear whether implementing DomainKeys yourself will help very much, as it depends on whether the users care which domain has been authenticated, or if simply proving that you're not a spoof is enough. Probably most users will just look for unspoofed mail and let it through, and you'll be fine, but that depends on your user base. You may have to educate them to add your domain to the list they accept. Footnotes: [1] It looks to me like "claim authorship" is in error. As far as I can tell from the DomainKeys page, DomainKeys verifies the sending domain, not the author's domain, although the page refers to authors and From several times. [2] Mailman is just one user of the typical system, and is not the domain "owner". Since DomainKeys authenticates domains rather than users, it should be done by the domain's mail server, not by user mail agents. (You have to reconfigure the DNS even if the mailing list manager does the signing, so even with signing implemented in Mailman you would need very high administrative privilege to implement DomainKeys.) -- School of Systems and Information Engineering http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software. From lcarlson at d.umn.edu Sat Aug 6 00:23:46 2005 From: lcarlson at d.umn.edu (Laura Carlson) Date: Fri, 05 Aug 2005 15:23:46 -0700 Subject: [Mailman-Developers] Wishlist Request for Mailman Web Acessibility Message-ID: <2268962264.1123255426@dial34-63.d.umn.edu> We would like to request that Web Accessibility be added to the Mailman wishlist. The Mailman user community includes an increasing number of individuals. Some have disabilities and rely on adaptive technologies. Incorporating accessibility would help increase the availability and accessibility of your software to people of all needs. It would be great to have Mailman at a minimum meet World Wide Web Consortium (W3C) Web Content Accessibility Single A Guidelines[1] and adhere to Section 508[2] of the Americans with Disabilities Act. There are some simple things that can help. For instance when the end user views templates like the information page and the user specific options page they get form fields that are not accessible. Forms are a problem because the extra accessibility tags are simply not known to many developers. Forms aren't the easiest of things to use for people with disabilities. Navigating around a page with written content is one thing, hopping between form fields and inputting information is another. Because of this, it is a good idea to add a number of elements to the form. Most of culprit HTML in question is in the python code. One of the main things that needs to be done is to associate form labels explicitly with their controls. Markup for labels needs to have the