p0.77 misplaces X-Sent:

p0.77 misplaces X-Sent:

[Karl Kleinpaste]

p0.77 is consistently misplacing the X-Sent: header, being immediately
below the first line of the message body.  (gnus-treat-date-lapsed is t.)

Example *Article* buffer displayed:

From: Lars Magne Ingebrigtsen <lmi@gnus.org>
Subject: Re: gnus hangs xemacs when news server is kaput
Newsgroups: gnu.emacs.gnus
Date: 20 Feb 1999 23:27:12 +0100
Organization: Dept. of Informatics, University of Oslo, Norway
Mail-Copies-To: never

Les Schaffer <godzilla@netmeg.net> writes:
X-Sent: 1 hour, 40 minutes, 29 seconds ago

> any ideas how i could at least try to debug the situation. seems odd
> if no one else has this problem, maybe its something stupid i did when
> i compiled xemacs.

I actually had a similar hang today -- the network was down, and
XEmacs hung while trying to do DNS lookups.  When the network came up,
the `C-g's I had typed were processed...

-- 
(domestic pets only, the antidote for overdose, milk.)
  lmi@gnus.org * Lars Magne Ingebrigtsen

Re: p0.77 misplaces X-Sent:

[Karl Kleinpaste]

/mutter/  It just figures that I realize the actual scenario, *just*
after sending the bug report.

The flaw is that X-Sent is placed at the bottom of the first block of
text, that is, above the first empty line.  Looks like p0.77 probably
narrows to message body when inserting X-Sent before "\n\n".

Re: p0.77 misplaces X-Sent:

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> /mutter/  It just figures that I realize the actual scenario, *just*
> after sending the bug report.
> 
> The flaw is that X-Sent is placed at the bottom of the first block of
> text, that is, above the first empty line.  Looks like p0.77 probably
> narrows to message body when inserting X-Sent before "\n\n".

I'm unable to reproduce this -- the function calls
`(article-narrow-to-head)', which should work OK, I think.  But I did
do rather a lot of fixes to the display functions to make it possible
for them to work recursively on narrowed regions, and there may very
well be some interference from other functions.  That is, could you
try switching off (or on :-) some of the other `gnus-treat-*'
functions and see what happens?

Oh, and what is your `gnus-treat-date-lapsed'?  It should probably be
`head'. 

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen

Re: p0.77 misplaces X-Sent:

[Kai Haberzettl]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Karl Kleinpaste <karl@justresearch.com> writes:
> >
> > The flaw is that X-Sent is placed at the bottom of the first block of
> > text, that is, above the first empty line.  
>
> well be some interference from other functions.  That is, could you
> try switching off (or on :-) some of the other `gnus-treat-*'
> functions and see what happens?

I've had the same problem as Karl, and the X-Sent goes away when I set 
(gnus-treat-date-local) to nil.

Kai

Re: p0.77 misplaces X-Sent:

[Karl Kleinpaste]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:
> Oh, and what is your `gnus-treat-date-lapsed'?  It should probably be
> `head'. 

That was the problem.  I had it simply `on'.  Customizing it to use
`head' makes X-Sent appear properly.

Perhaps a comment in the doc string is in order.  Or alternatively a
means to limit the available choices for gnus-treat-* to things that
make sense for the intended context of each variable.

Re: p0.77 misplaces X-Sent:

[Lars Magne Ingebrigtsen]

Karl Kleinpaste <karl@justresearch.com> writes:

> Perhaps a comment in the doc string is in order.  Or alternatively a
> means to limit the available choices for gnus-treat-* to things that
> make sense for the intended context of each variable.

Perhaps the gnus-treat-* variables that control functions that are
primarily meant for the headers should have a customize spec that only 
allows nil or `head'?  Yes.  Fix in Pterodactyl Gnus v0.78.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen