Hello,

I have been struggling for a while with all sorts of termcap issue when
trying to use solarized theme with rxvt-unicode (9.12, latest from
ports) under FreeBSD (8.2-RELEASE), which means no terminfo and only
termcap (in case it's relevant).

Using 256 color support, I ended up with something almost working,
except a little glitch with tmux, that is sumuarized in the following
screenshot:
http://instinctive.eu/screenshots/20111026-122711-tmux-sgr0-bug.png

The upper window is my rxvt-unicode (with 256 color support compiled in,
though I would have loved to have something working in default 88 color
mode, that's already plenty for a 16-color palette).

The lower window is exactly the same but with tmux running in the
rxvt-unicode, and a different TERM value.

It seems that "something" makes tmux display characters in a brighter
color, and that something is triggered by my fancy prompt, which
accounts for the differences between bare rxvt-unicode and the first
"printf". The second one has the prompt slightly modified (reporting an
error, that's the bold red "1" there) which does not trigger it.

So changing the color (using \033[36m in my example, though it is
trigger by all 30-37 and 90-97 codes) somehow trigger the switch from
the correct rendering to the too-bright rendering, as can be seen at the
end of the printf'ed line.

Messing with display modes seems to somehow revert it back to the
correct rendering: in my example any text between \033[7m and \033[27m
puts the rendering back into "correct" mode when leaving it. I see the
same effect with any text between \033[3m and \033[23m and with any text
between \016 and \017. However there has to be some text, even only
spaces, as having the starting control sequence and immediately
afterwards the closing control sequence does not put the display back to
correct state.

To my surprise, sgr0 sequence \033[m does not help: if display is
correct, it stays so, and if it's too bright, it also stays so.

I tried to investigate what is going on in tmux 1.5 source code, and I
reached input_csi_dispatch_sgr() in input.c. Considering the issue
resists to sgr0, I guess it has something to do with grid_cell.attr
however color change does not actually affects it. So I'm stuck at that
point.

Would anyone have insights on that part of the code to help me diagnose
further the issue?
I would gladly provide any additional system-specific configuration
(like termcap entries, X resources for rxvt-unicode, etc) if needed, but
there is nothing fancy outside of FreeBSD defaults and straightforward
solarized stuff.


Thanks in advance for your help,
Natacha Porté

------------------------------------------------------------------------------
The demand for IT networking professionals continues to grow, and the
demand for specialized networking skills is growing even more rapidly.
Take a complimentary Learning@Cisco Self-Assessment and learn 
about Cisco certifications, training, and career opportunities. 
http://p.sf.net/sfu/cisco-dev2dev
_______________________________________________
tmux-users mailing list
tmux-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/tmux-users

Reply via email to