Mailman 3 February 2000 - Mailman-Developers

Re: [Mailman-Developers] editing email addresses
by Harald Meland Feb. 13, 2000

Feb. 13, 2000

[Ola Flygt] > Hello > > At Växjö University we are using Mailman for a number of lists. We > are now in the process of changing email addresses for a lot of users > since we have a change in the organization. Is there a way to change > all the users email addresses in the database instead of deleting and > adding them again? There's bin/clone_member, which will clone an address. Combined with bin/remove_members this would makes for "change_member_address" functionality. You'd probably want to use the output of bin/list_members run on the various list if you want to script all of this. -- Harald

1 0

Re: [Mailman-Developers] Juat an Idea - HTML
by Harald Meland Feb. 13, 2000

Feb. 13, 2000

[Ricardo Kustner] > On Sat, Feb 12, 2000 at 05:20:52PM -0800, J C Lawrence wrote: > > On Fri, 11 Feb 2000 10:29:54 +0000 > > David Sexton <david.sexton(a)sapphire.net> wrote: > > > Would it be feasable for MM to strip out HTML tags from a > > > message for the whole list (especially for when no HTML is in the > > > list's acceptable use policy)? > > There are external tools that can do this quite well. See freshmeat > > for a couple examples. You can then just have your MailMan lists > > run thru procmail which invokes the HTML stripper before handing off > > while this is a good idea/solution, I still think it would be very > useful to have this feature inside MM... you can't expect every admin > to start using procmail filters... I agree to the extent that list-specific configuration issues should rather be taken care of inside Mailman than rely on whatever tricks the list admins can harass their site admins into doing for them. However, I suspect that implementing this specifically for HTML-containing email would be a too specific solution to a more general problem -- it seems to me that list admins want to block incoming email on a whole lot of different criteria, and providing a separate configuration button for each and every of these specific criteria will bloat the admin pages a lot (they're actually too unwieldy as they are now, IMHO). So, I think we need to decide what classes of blocking criteria we want to support, and then find a way of implementing each class in a suitable fashion. In the particular case of HTML-containing mail, I'd love to see a general solution to transforming/blocking on/throwing away/whatever MIME parts of specific types. I'm not sure how the configuration interface to this functionality should be done, though -- the goal being to un-bloat the list admin pages while adding functionality... -- Harald

1 0

admindb.py patch
by Ricardo Kustner Feb. 13, 2000

Feb. 13, 2000

I know i've posted this before, but this time i've made a few little changes... below is a patch to Mailman/Cgi/admindb.py which 1) seperates the heldmsg header from the body 2) puts the header in a separate 3 row textarea 3) moves the "approve/reject/discard" line to the below the message, since that's much more convinient when approving many messages this way you can immediately start reading the message instead of having to scroll down the header lines first. ps: maybe the header/body seperate code should be moved to a function instead... Ricardo. *** /usr/local/cvs/mailman/Mailman/Cgi/admindb.py Mon Nov 15 23:29:46 1999 --- admindb.py Sun Feb 13 16:47:07 2000 *************** *** 195,201 **** form.AddItem(Center(Header(2, msg))) try: fp = open(os.path.join(mm_cfg.DATA_DIR, filename)) ! text = fp.read(mm_cfg.ADMINDB_PAGE_TEXT_LIMIT) fp.close() except IOError, (code, msg): if code == ENOENT: --- 195,211 ---- form.AddItem(Center(Header(2, msg))) try: fp = open(os.path.join(mm_cfg.DATA_DIR, filename)) ! msg_header = "" ! msg_body = "" ! for line in fp.readlines(mm_cfg.ADMINDB_PAGE_TEXT_LIMIT): ! if msg_body == "": ! if line != "\n": ! msg_header = msg_header + line ! else: ! msg_body = line ! else: ! msg_body = msg_body + line ! fp.close() except IOError, (code, msg): if code == ENOENT: *************** *** 214,223 **** t.AddRow([Bold('Reason:'), reason]) t.AddRow([Bold('Subject:'), subject]) t.AddRow([ - Bold('Action:'), - RadioButtonArray(id, ("Approve", "Reject", "Discard (eg, spam)")) - ]) - t.AddRow([ Bold('If you reject this post, explain (optional):'), TextArea('comment-%d' % id, rows=4, cols=60, text=("Please do *not* post administrative requests" --- 224,229 ---- *************** *** 229,236 **** ]) row, col = t.GetCurrentRowIndex(), t.GetCurrentCellIndex() t.AddCellInfo(row, col, colspan=3) t.AddRow([Bold('Message Excerpt:'), ! TextArea('fulltext-%d' % id, text, rows=10, cols=60)]) row, col = t.GetCurrentRowIndex(), t.GetCurrentCellIndex() t.AddCellInfo(row, col, colspan=3) form.AddItem(t) --- 235,247 ---- ]) row, col = t.GetCurrentRowIndex(), t.GetCurrentCellIndex() t.AddCellInfo(row, col, colspan=3) + t.AddRow([Bold('Message Header:'), TextArea('header-%d' % id, msg_header, rows=3, cols=60)]) t.AddRow([Bold('Message Excerpt:'), ! TextArea('fulltext-%d' % id, msg_body, rows=10, cols=60)]) ! t.AddRow([ ! Bold('Action:'), ! RadioButtonArray(id, ("Approve", "Reject", "Discard (eg, spam)")) ! ]) row, col = t.GetCurrentRowIndex(), t.GetCurrentCellIndex() t.AddCellInfo(row, col, colspan=3) form.AddItem(t) --

1 0

heldmsg
by Ricardo Kustner Feb. 13, 2000

Feb. 13, 2000

just a quick thought... currently, the 1.2 CVS version puts held messages for all lists into ~mailman/data ... wouldn't it be a bit cleaner to put these messages in some place like ~mailman/lists/listname/heldmsg/ instead? Ricardo. --

1 0

Stripping HTML
by Bill Bumgarner Feb. 13, 2000

Feb. 13, 2000

Ricardo, While you are at it, add the ability to strip out other formatting-- like the non-HTML <param>font</param stuff you find from certain clients [it isn't HTML, is it]? This is not quite as straightforward as one might expect. In particular, you need to be able to actually rewrite attachments in the MIME body of the message-- something that the "out of the box" Mailman cannot do. However, I had to do exactly that-- including reencoding the rewritten message back into a MIME compliant beast-- to make Mailman rip apart messages with attachments, file the attachments to a WebDAV server, and rewrite the attachment such that it is text/plain with a body that has an URL pointing to the filed attachment. It would be very useful to the project that started all of this if Mailman could "clean" the messages of formatting information. If you send me a function/class/code snippet that does the stripping in (as someone else said) "just the right way", I would be happy to integrate it as a configurable option in my Rewrite module... and, of course, make the whole ball of wax available to the community at large (including helping Barry integrate all of this stuff into the core distribution). thanks! b.bum

4 3

Re: [Mailman-Developers] Juat an Idea - HTML
by J C Lawrence Feb. 13, 2000

Feb. 13, 2000

On Fri, 11 Feb 2000 10:29:54 +0000 David Sexton <david.sexton(a)sapphire.net> wrote: > Would it be feasable for MM to strip out HTML tags from a > message for the whole list (especially for when no HTML is in the > list's acceptable use policy)? There are external tools that can do this quite well. See freshmeat for a couple examples. You can then just have your MailMan lists run thru procmail which invokes the HTML stripper before handing off to mailMan. -- J C Lawrence Home: claw(a)kanga.nu ----------(*) Other: coder(a)kanga.nu --=| A man is as sane as he is dangerous to his environment |=--

2 1

Bug#57596: mailman: Here we go again... post: nntplib.error_perm : 500 "POST" not implemented; try "help". (fwd)
by Gergely Madarasz Feb. 13, 2000

Feb. 13, 2000

Another day, another bugreport :/ Lazarus, can you tell me if this is after you applied the last patch suggested by Thomas, for gate_news ? Package: mailman Version: 1.1-2 Severity: important Feb 09 03:47:55 2000 post: Traceback (innermost last): post: File "/var/lib/mailman/scripts/post", line 73, in ? post: mlist.Post(msg, approved=fromusenet) post: File "/usr/lib/mailman/Mailman/MailList.py", line 1331, in Post post: self.SendMailToNewsGroup(msg) post: File "/usr/lib/mailman/Mailman/GatewayManager.py", line 204, in SendMailToNewsGroup post: con.post(msg) post: File "/usr/lib/python1.5/nntplib.py", line 417, in post post: resp = self.shortcmd('POST') post: File "/usr/lib/python1.5/nntplib.py", line 158, in shortcmd post: return self.getresp() post: File "/usr/lib/python1.5/nntplib.py", line 134, in getresp post: raise error_perm, resp You get the idea. Gatewaying seems completely broken.

3 2

nightly_gzip - more than a little broken...
by Nigel Metheringham Feb. 13, 2000

Feb. 13, 2000

I wondered why the nightly_gzip program was not working correctly on some of my lists - one (a high traffic list) has weekly based archiving, the other (an announce list) has quarterly based archiving. A very quick look at the code shows:- for f in allfiles: try: time.strptime(f, '%Y-%B.txt') except ValueError: continue that strptime seems to be ensuring that the archive mbox files have a year and month in their names before a .txt - and this only applies to monthly archives in practice. Would there be any problems with just changing this to be a regexp match for '\.txt$' instead? If you really want to check its an archive then look for a directory x with a file x.txt . Nigel. -- [ - Opinions expressed are personal and may not be shared by VData - ] [ Nigel Metheringham Nigel.Metheringham(a)VData.co.uk ] [ Phone: +44 1423 850000 Fax +44 1423 858866 ]

2 1

cron/gate_news watermark == 0
by Jim Tittsler Feb. 13, 2000

Feb. 13, 2000

A watermark value of 0 is used for two different purposes in cron/gate_news: 1. the group has never been gatewayed, so "catch up" to the highest article number 2. the highest article number seen for the group is 0 (which can be a valid "high article number" before the first posting in a newly created newsgroup) The result of this is the first article posted to the newly created newsgroup fails to get gatewayed to the mailing list because the watermark for that group is (still) 0 which is used as a flag to "catch up" so all that happens is the watermark gets set to 1. I believe I have fixed that in my copy by using the value None for the first time a group is gatewayed, allowing proper processing when the watermark is truly article number 0. Jim

1 0

Mailman limitations
by Edward Elhauge Feb. 12, 2000

Feb. 12, 2000

Hi developers, I transfered the ten mailing list I run from Smartlist to Mailman about two months ago and don't regret it. But I can also see some improvements needed in Mailman to get it generally accepted. I've been on the list a couple of weeks and posted a request on the mailman-users list about these limitations and haven't seen any solutions forthcoming so I presume what I want is not available in Mailman right now. I'm a fairly senior Perl programmer and would love to learn Python so now might be the time for me to find out if there is something the Mailman team needs in terms of Python programming. What I find great about the system is the authentication of the address required to get on the list and the monthly reminder for users on how to get off. Also the Web user config interface and admin interface along with the Web archive. What I don't like is that I can't run the lists on autopilot mode. My lists are fairly small, between 20 and 100 mail address in each. A lot of the subscribers on one list are from other countries and post in French or Spanish. A lot of political activists post to these lists about their own interests without staying on topic for the lists. I have the lists set up so that you have to subscribe to post; that deters some of the huge political mailers. The authentication step, I think, stops some people who use their addresses to post but never read their mail. I have the too many addresses filter in, although I bumped the limit to about 15. I also set the implicit address filter. Both of these last two steps promote the interactive list atmosphere that I am trying to promote. The bad thing is that I then have at least 10 messages to approve throughout the day. It is a real hassle. Most of these people seem to ignore my rejection notices; perhaps some of them don't understand the generic message sent back that talks about 'not posting administrative requests' etc, since it isn't to the point of why their post was rejected, and it may not be in a language they understand. Editing each message would take up even more of my time. Some of them may never read the mail received at their address. I feel that a lot of sites that have dropped Mailman, after giving it a try, might have done so because they can't run their lists on autopilot. I think was is needed is to keep these filters, but to have the option to have an automatic rejection message specific to the rejection reason and to the list. As a further refinement the message might be specifically in the language that the poster used. Furthermore there should be a database of rejects by address, so that persistent rejects can be put on a reject spam list at the the MTA level. Let me know if these capabilities are already in Mailman or if there is already ongoing development along these lines. If not, I would be interested in working on enhancements along these lines. -- Edward Elhauge <ee(a)uncanny.net> | "War is like love; Uncanny Inc., San Francisco | it always finds a way." | -- Bertold Brecht

2 1