Mailman 3 August 2001 - Mailman-Developers

Re: [Mailman-Developers] MM2.1 DemoList
by Dan Mick 04 Aug '01

04 Aug '01

> At 12:50 AM 8/3/01 -0400, Barry A. Warsaw wrote: > > MY> The abbreviation for kilobyte is kB (1000 or 1024) or KB > > MY> (1024). Perhaps it should be spelled out to avoid confusion. > > Actually KB isn't an official standard for multiples of 1024. Computer > people are just supposed to know when a K means 1000, and when > it means 1024. That's why Nist invented Kebi. You mean "NIST". ;) > A Kebibyte, or KeB is > based on 1024 (As are Mebibytes, and Gebibytes.) Not, mind you, that > you'll ever catch me pronouncing them outloud. Not that you'll ever catch me using them, either. The only people who care are disk drive manufacturers, who lie like rugs on a continuing basis. Everyone else knows what a kilobyte or a megabyte really is, despite the otherwise-annoying imprecision, and has been using it for more than 30 years, so NIST can go pound sand, IMO.

1 0

Re: [Mailman-Developers] MM2.1 DemoList
by Dan Mick 03 Aug '01

03 Aug '01

> MC> The usual feedback I get from subscribers is the inability for > MC> them to find the way to unsubscribe. > I've reorganized things a bit so that it's clear the "Edit Options" > button can be used for unsubscribing as well (the button now reads > "Unsubscribe or edit options"). Also, if you leave the field blank, > you'll be shown the options login page with an extra field prompting > you for your email address. So the "simple unsubscribe directions" > can now be explained as: > > 1. Go to http://www.dom.ain/mailman/options/listname > 2. Enter your email address > 3. Click on "Unsubscribe" > 4. Check your email for confirmation instructions > > #4 may still be optional if I get to implement open unsubscribes. Strong vote here for moving the "options" further up the page. Those who are just looking can scroll past the "if you're a current member" stuff or not, I don't care, but I want my subscribers to see the "<listname> Subscribers" section immediately without scrolling. Also, yes, I think the title should be clear. In fact, I think the "Visit Subscriber List" should not be in this section. In fact, I'd go as far as, up near the top, adding a new major section (perhaps right after About <listname>): --- Unsubscribing To unsubscribe from <listname>, enter your subscription email address: <textbox> <Unsubscribe button> ---- and then I'd probably move "<listname> Subscribers" up there too. Service your current customers higher on the page; let people who want to subscribe be the ones who scroll...

2 1

cron/mailpasswds
by Michael Balzer 03 Aug '01

03 Aug '01

dear mm developers, may i suggest a change to cron/mailpasswds? instead of using the same from address for all passwd notifications, use a from address that reflects the domain (virtual host). reason: all replies to passwd reminders now go to the main mailman-owner. it should be possible to direct them to the mailman-owner of the domain instead. diff: 61d60 < mailman_owner = mm_cfg.MAILMAN_OWNER 62a62 > mailman_owner = 'mailman-owner@' + host with a bit more context: ... for host, users in hosts.items(): mailman_owner = 'mailman-owner@' + host subj = host + ' mailing list memberships reminder' ... this change should be accompanied by a config var for the 'mailman-owner' part + a documentation change, so users know they can add some mailman-owner(a)domain.com address for each virtual list domain. mfg, michael balzer ps: please reply to me directly, i'm not subscribed. -- 2BCom IT Solutions / Hasslinghauser Str. 156 / D-58285 Gevelsberg Fon +49 2332 7588-0 / Fax -11 / http://www.bbcomp.de

1 0

Demolist feedback
by Edmund Lau 03 Aug '01

03 Aug '01

Under Digest Options for the demolist, how about changing "Should Mailman start a new digest volume?" to "Should Mailman start a new digest volume right now?" I'm assuming that's what it means. -Ed

2 1

Relax, NEW! Electronic Pulse Massager
by Discount Web Shopping 03 Aug '01

03 Aug '01

1 0

Re: [Mailman-Developers] Clean output with bin/withlist??
by Dan Mick 02 Aug '01

02 Aug '01

> def hostname(m): > print m.host_name > > % bin/withlist -q -r hostname mylist > wooz.org Who needs -q? Everything but the output comes to stderr; the desired output comes to stdout. This has been true for some time now, no?

1 0

Looping digest - mailman bug?
by Kaja P. Christiansen 02 Aug '01

02 Aug '01

Hi, On Wed, 20 Jun 2001 19:12:00 +0200, I wrote: > Last Monday, one of our mailing lists went mad and sent multiple copies > of the same digest to the list's digest users. > > [...] > > The traceback in the errors log shows: > > Jun 19 00:01:01 2001 (18618) Traceback (innermost last): > File "/usr/local/mailman203/Mailman/Handlers/HandlerAPI.py", line 82, in do_pipeline > func(mlist, msg, msgdata) > File "/usr/local/mailman203/Mailman/Handlers/Sendmail.py", line 86, in process > fp.write(msgtext) > IOError: [Errno 32] Broken pipe > > We run Mailman 2.0.3 with Postfix. I addressed this problem earlier > (March 26th), but there was no reply. Several of our lists suffered from Mailman's mail bombing and we turned the digest option off in hope it'll help. It did, for a while, until the same happened with a non-digest message being send over and over again (once per minute, by qrunner). We were able to locate where and how it happens, and to 'reproduce' the error (in laboratory conditions :-) both under mailman 2.0.3 and 2.0.6. The mail looping occurs when there is a MIME message with a single . (dot) in a line; Mailman sends it to sendmail 'as is'. But since sendmail/postfix interprets a line with single dot as the end of the message, it sends everything before the dot and exits with 'Broken pipe'. Mailman, however, still has the message in it's queue and sends it all over again... When I tried sending non-MIME message with single-dot-line in it to a Mailman list, there was no looping, but the message body after the single dot was missing. Until there is something better, I suggest adding to Sendmail.py a patch which perhaps is not pretty (it adds a space before the infamous dot), but it works: --- ./Mailman/Handlers/Sendmail.py.orig Fri Jul 27 13:40:31 2001 +++ ./Mailman/Handlers/Sendmail.py Fri Jul 27 14:11:46 2001 @@ -31,6 +31,7 @@ import string import os +import re import HandlerAPI from Mailman import mm_cfg @@ -80,6 +81,8 @@ msgtext = str(msg) # cycle through all chunks failedrecips = [] + bar = re.compile('\n\.') + msgtext = re.sub(bar, '\n .', msgtext, 0) for chunk in recipchunks: # TBD: SECURITY ALERT. This invokes the shell! fp = os.popen(cmd + chunk, 'w') Kaja

4 9

Re: [Mailman-Users] modules relationship
by J C Lawrence 01 Aug '01

01 Aug '01

On Wed, 1 Aug 2001 15:51:20 +0100 Yanuar Nugroho <y.nugroho(a)student.umist.ac.uk> wrote: > If I would like to know how the modules are interrelated each > other, along with their functionalities in Mailman, what > documentation should I consult? The source? -- J C Lawrence )\._.,--....,'``. ---------(*) /, _.. \ _\ ;`._ ,. claw(a)kanga.nu `._.-(,_..'--(,_..'`-.;.' http://www.kanga.nu/~claw/ Oh Freddled Gruntbuggly

1 0

modules relationship
by Yanuar Nugroho 01 Aug '01

01 Aug '01

Dear Barry et al, If I would like to know how the modules are interrelated each other, along with their functionalities in Mailman, what documentation should I consult? Thank you, Yanuar Yanuar Nugroho http://www.yanuar-n.f2s.com/ http://yanuar-nugroho.mainpage.net/ ---------------------------------------- PG. Stud. MSc. Information Systems Engineering, Dept. of Computation, UMIST, Manchester - UK

2 1

new part for empty header (?)
by Georg Koch 01 Aug '01

01 Aug '01

Hi Barry, hi Developers, when I look into Mailman/Handlers/Decorate.py, it seems to me that it adds a new header part to a multipart/mixed even if the header is empty. Is it so? Best wishes Georg -- -- Georg Koch ( mailto:gorg@cochrane.de ) | Phone: +49 761 203 6710 German Cochrane Centre | Fax: +49 761 203 6712 Mail: Institute of Medical Biometry and Medical Informatics Stefan-Meier-Strasse 26, D-79104 Freiburg, Germany ...................................................................... Crossing the Quality Chasm: A New Health System for the 21st Century http://www.nap.edu/books/0309072808/html/

2 1