Gnus development mailing list
 help / color / mirror / Atom feed
From: "Edward J. Sabol" <sabol@alderaan.gsfc.nasa.gov>
Subject: Re: Posting styles
Date: Thu, 30 Jan 1997 16:47:44 -0500	[thread overview]
Message-ID: <199701302147.QAA01798@alderaan.gsfc.nasa.gov> (raw)
In-Reply-To: <whu3nzxe8w.fsf_-_@tyr.metis.no> (message from Steinar Bang on 30 Jan 1997 18:30:07 +0100)

Excerpts from mail: (30-Jan-97) Posting styles by Steinar Bang
Steinar> Could someone summarize the posting styles of GNUS?

Gnus-posting-styles was originally implemented in September Gnus 0.10 and
later removed right before the Gnus 5.2 release (I assume due to some
incompatibility with message-mode which was added late in the September Gnus
development period) under the cloak of darkness and without even a ChangeLog
entry. See the Ding nnml archive circa September 28th, 1995 for the whole
story of the origin of gnus-posting-styles.

The gnus-posting-styles variable was an alist of `(MATCH . STYLE-LIST)'
pairs where MATCH could be either a regexp or a function/form to be eval'ed
and STYLE-LIST was an alist of `(ATTRIBUTE . VALUE)' pairs. Valid attributes
were supposed to be one of `signature', `from', `organization', and `xface'
or a string signifying an abtitrary message header. VALUE could also be a
function or form to be eval'ed. For example:

 (setq gnus-post-style
       '((".*" . ((signature . nil)
                  (xface . "~/.xface")
                  (organization . "Rebel Alliance")
		  ("X-Home-Page" . (getenv "WWW_HOME"))
                  ("X-Arbitrary-Header" . "blah, blah, blah")
                  ))
         ((posting-from-work-p) . ((signature . "~/.work-signature")
                                   (from . "user@bar.com")))
         ((not (usenet-post-p)) . ((signature . "~/.mail-signature")))
         ("^rec.humor" . ((signature . (my-funny-signature-randomizer))
                          (from . "user@foo.com")))
         ("^alt.quotations" . ((signature . (my-quote-randomizer))
                               (from . "user@foo.com")))
         ("^comp" . ((signature . "My computer is better than yours!")
	             (organization . "Whiz-Bang Computer, Inc.")))
         (".*emacs.*" . ((signature . "Give me Emacs or give me death!")))
         ("^nnml: . ((signature . "~/.mail-signature")))
         ))

Gnus iterated through each element of the alist in order and used all the
styles that match. Thus, the order of items in the list was significant. In
the above example, the signature "Give me Emacs or give me death!" would be
used in both comp.editors.emacs.* and alt.religion.emacs, but not
comp.unix.programmer.

I think this has a *lot* of intrisic freedom that using group parameters
doesn't give you. See the above example where in two cases the MATCH portion
of the alist are `(posting-from-work-p)' and `(not (usenet-post-p)'. Group
parameters couldn't do that.

Excerpts from mail: (30-Jan-97) Re: GNUS... by Per Persson
Per> My guess is that it should be a part of the Group Parameters thingiebob
Per> now, no? This seems like the most sensible thing to me.

The problem with using group parameters for posting styles comes when you
have well over 200 hundred groups that you are subscribed to. Setting a
specific style for say, all the comp.* groups gets pretty tedious. Topics can
help with this, but then you're forcing everyone to use topics.

For the best of both worlds, maybe if similar group parameters exist they
could override the variable settings?

My suggestion is to keep it as a variable as defined above and possibly
rename it message-posting-styles. I can see people wanting to use it or
something like it outside of Gnus whereever there is message-mode. The
definition of this message-posting-styles variable would probably have to be
tweaked in some manner to make it non-Gnus-specific. For example, instead of

 (setq gnus-posting-style
       '(("^rec.humor" . ((signature . (my-funny-signature-randomizer))
                          (from . "user@foo.com")))))

there would be

 (setq message-posting-style
       '((gnus-group-name-matches "^rec.humor") .
           ((signature . (my-funny-signature-randomizer))
            (from . "user@foo.com")))))

This way it could be generic.

Later,
Ed


  parent reply	other threads:[~1997-01-30 21:47 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <iv4gkuup.fsf@swip.net>
     [not found] ` <m2zpxs7yp6.fsf@proletcult.slip.ifi.uio.no>
1997-01-30 11:19   ` GNUS Per Persson
1997-01-30 14:36     ` GNUS Lars Magne Ingebrigtsen
1997-01-30 17:30       ` Posting styles Steinar Bang
1997-01-30 21:47       ` Edward J. Sabol [this message]
1997-01-31 22:08         ` Lars Magne Ingebrigtsen
1997-02-03 19:33           ` Edward J. Sabol
1997-02-04  0:49             ` Lars Magne Ingebrigtsen
1997-02-01 18:40       ` anonymous
     [not found]       ` <199701302147.QA <19970201184031.24574.qmail@sunsite.auc.dk>
1997-02-03  4:39         ` Lars Magne Ingebrigtsen
1999-03-14 18:48 Harry Putnam
1999-03-14 20:04 ` Jason R Mastaler
1999-09-24 21:24 posting styles Greg Stark
1999-09-24 22:33 ` Jack Vinson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=199701302147.QAA01798@alderaan.gsfc.nasa.gov \
    --to=sabol@alderaan.gsfc.nasa.gov \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).