Mail body escape sequence fouls display

Mail body escape sequence fouls display

[Karl Kleinpaste]

After a very long hiatus, yesterday I upgraded to XEmacs 21.5.20 and
No Gnus v0.03.

When I get mail which contains literal escape characters in the text,
it can make all text following that point, and all further mail in the
same group, be grey or reverse video.

Specific circumstance is that I have environment variables TERM_SO and
TERM_SE, which contain the libcurses "standout" and "standout end"
(rev video) sequences for the terminal.  An automatic self-mail-
sending function in my shell emails me dumps of my shell state in
terminals, including env.variables.  When these mails are subsequently
seen in Gnus, the literal escape sequence of TERM_SO causes all further
text downward in that email to be rev video.

Weirder still, when reading the _next_ message in the same group,
*all* text is rev video, *except* for the line following the TERM_SE
sequence, which of course turns off rev video in a terminal, but then
(because of the order of the dump) the following TERM_SO turns rev
video back on again.

But the _header_ is never rev video, even in 2nd-thru-Nth items.

Since when does Gnus, or XEmacs in general, react to embedded escape
sequences in text being displayed?  (As opposed to using escape
sequences heavily for terminal control, of course.)

Screenshots...
First message in group, text goes rev video at the TERM_SO:
http://www.charcoal.com/~karl/gnus/screenshots/summary-personal.recall-1.png
Second message, text is rev video from beginning;
http://www.charcoal.com/~karl/gnus/screenshots/summary-personal.recall-2.png

--karl

Re: Mail body escape sequence fouls display

[Reiner Steib]

On Wed, Apr 06 2005, Karl Kleinpaste wrote:

> Since when does Gnus, or XEmacs in general, react to embedded escape
> sequences in text being displayed?

,----[ (info "(gnus)No Gnus") / GNUS-NEWS ]
|    * ANSI SGR control sequences can be transformed using `W A'.
| 
|      ANSI sequences are used in some Chinese hierarchies for
|      highlighting articles (`gnus-article-treat-ansi-sequences').
`----

Does it help to set `gnus-treat-ansi-sequences' to nil?

Bye, Reiner.
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/

Re: Mail body escape sequence fouls display

[Karl Kleinpaste]

Reiner Steib <reinersteib+gmane@imap.cc> writes:
> Does it help to set `gnus-treat-ansi-sequences' to nil?

Much better, thank you.  It would never have occurred to me that this
was deliberate.  I will look through the news file more.

I suspect it's (still) a bug for this to affect subsequent email viewing.

Re: Mail body escape sequence fouls display

[Reiner Steib]

On Wed, Apr 06 2005, Karl Kleinpaste wrote:

> Reiner Steib <reinersteib+gmane@imap.cc> writes:
>> Does it help to set `gnus-treat-ansi-sequences' to nil?
>
> Much better, thank you.  It would never have occurred to me that this
> was deliberate.

It's for ANSI escape sequences used in Chinese hierarchies, see [1].

> I suspect it's (still) a bug for this to affect subsequent email viewing.

Sure, but the question is whether the bug is in Gnus or XEmacs
21.5-b20.  Could you please post such an article in gmane.test
(news.gmane.org) or gnus.test?

Bye, Reiner.

[1] <URL:http://thread.gmane.org/v9brpg29nm.fsf%40marauder.physik.uni-ulm.de>
    <news:87smm53pew.fsf@jidanni.org> <news:iluk7700y82.fsf@extundo.com>
-- 
       ,,,
      (o o)
---ooO-(_)-Ooo---  |  PGP key available  |  http://rsteib.home.pages.de/