Subj : Talisman: bbs_display_sixel
To   : Digital Man
From : bugz
Date : Mon Jul 10 2023 08:54 pm

-=> Digital Man wrote to bugz <=-

 >  DM> SyncTERM in non-graphics modes (e.g. curses or ANSI) doesn't do
 >  DM> graphics (e.g. RIP, Sixel). That's expected. --
 >  DM>                                             digital man (rob)
 >
 > Should SyncTERM send the cterm response that it supports Sixel when
 > in curses mode?

 DM> I don't think so.

Here's what I'm seeing with Wireshark

From SyncTERM curses:  (syncterm -iC)

SEND >> 1b 5b 3c 30 63
RECV << 1b 5b 3c 30 3b 32 3b 33 3b 37 63
\x1b[<0;2;3;7c

The CTerm Device Attributes, 3 - Pixel operations supported
(sixel and PPM graphics).  I didn't expect it here.

SyncTERM X11: (syncterm -iX)

SEND >> 1b 5b 3c 30 63
RECV << 1b 5b 3c 30 3b 31 3b 32 3b 33 3b 34 3b 35 3b 36 3b 37 63
\x1b[<0;1;2;3;4;5;6;7c

I definitely would expect it here.

 > I didn't expect it to say it supported it, when
 > it didn't (or more specifically couldn't).

 DM> I agree.

Did I find a bug?

 > Is this a detection issue?  Is there a better way to detect Sixel support
 > in SyncTERM?  (Or is the answer -- just don't use curses mode?!)

 DM> Better than what?
 DM> I use CTDA query and I think that works correctly:
 DM> https://gitlab.synchro.net/main/sbbs/-/blob/master/exec/load/cterm_lib.j
 DM> s

Ok, that's what I've seen used.  I wasn't sure if there was a better way
to detect sixel support then the CTDA query.

Take care,

bugz (Steve)

--- Talisman v0.47-dev (Linux/x86_64)
 * Origin: Lunar Mod (21:1/236)

.