Fixed [v36.50.68]: Prompt problem in TCMD with ConPTY

[Alpengreis]

In v36.50.68 I have a prompt problem in TCMD tab with ConPTY:

First after loading TCMD it looks good ...

But then after I entered a command it looks like this ...

This is the problematic prompt:

$e[1;34m%USERNAME%$e[1;37m@$e[1;34m%USERDOMAIN%$s$e[1;33m$g$e[1;36m[$e[1;33m"$e[1;37m$p$e[1;33m"$e[1;36m]$_$e[1;37mADM:$s$e[0;37mN$s$e[1;36m$b$s$e[1;37mZeit:$s$e[0;37m$t$h$h$h$s$e[1;36m$b$s$e[0;37m$e[1;35m$$$s$e[0;37m

With prompt:

$p$g

I have no problem.

EDIT: It's because the prompt is over TWO LINES ...

 

[vefatica]

My prompts always end with $e[0m. That says default foreground on default background. I believe defaults are set here.

In TCC I say to use the defaults.

 

[Alpengreis]

Ty! Unfortunately the $e[0m doesn't help in that case ...

PS: Also, my prompt was ok before v36.50.68 ...

PPS: Scheme has no influence, colors are standard else ... and last but not least happens in ConTPY tab only (not in Conhost).

 

[vefatica]

What you see happens here also. The unwanted color is 80/80/80; I'd guess ^e[1;40 (bri bla background?). It looks like a clear_to_EOL that's using a wrong color.

After restarting TCMD it doesn't happen.

Note that I am setting a theme by brute force in TCStart.btm. I think it's this one.

^e]4;0;rgb:0c/0c/0c^e\^e]4;1;rgb:be/2c/21^e\^e]4;2;rgb:3f/ae/3a^e\^e]4;3;rgb:be/9a/4a^e\^e]4;4;rgb:20/4d/be^e\^e]4;5;rgb:bb/54/be^e\^e]4;6;rgb:00/a7/b2^e\^e]4;7;rgb:be/be/be^e\^e]4;8;rgb:80/80/80^e\^e]4;9;rgb:ff/3e/30^e\^e]4;10;rgb:58/ea/51^e\^e]4;11;rgb:ff/d1/4e^e\^e]4;12;rgb:2f/6a/ff^e\^e]4;13;rgb:fc/74/ff^e\^e]4;14;rgb:00/e1/f0^e\^e]4;15;rgb:ff/ff/ff^e\^e]10;rgb:be/be/be^e\^e]11;rgb:0c/0c/0c^e\^e]12;rgb:ff/ff/ff^e\^e]17;rgb:92/a4/fd^e\^e[2J^e[3J^e[H

 

[Alpengreis]

Yes, after restarting it's okay. BUT after I entered a command the problem appears again.

I also had set a theme by "brute force via tcstart - same problem.

 

[vefatica]

Any command? I can't get the problem again (at all)>

 

[vefatica]

Mildly interesting. This leads to corruption.

set prompt=`$e[1;33m$g$e[1;36m[$e[1;33m"$e[1;37m$p$e[1;33m"$e[1;36m]$_$e[1;37mADM:$s$e[0;37mN$s$e[1;36m$s$e[0m`

This doesn't:

set prompt=`$e[1;33m$g$e[1;36m[$e[1;33m"$e[1;37m$p$e[1;33m"$e[1;36m]$_$e[1;37mADM:$s$e[0;37mN$s$e[1;36m$e[0m`

The difference is $s (or not) before the final $e[0m

 

[Alpengreis]

Ahhh, thank you very much for investigating! I appreciate that!

Will test that later ...

 

[Alpengreis]

Yes, same result here too. First corrupted, second not.

Unfortunately, a remove of $s in MY prompt above don't works ...

 

[Alpengreis]

I have it!

IF I place a NON COLOR char at the end it works, regardless with or without $s before or not!

Means the following will work:

prompt $e[1;34m%USERNAME%$e[1;37m@$e[1;34m%USERDOMAIN%$s$e[1;33m$g$e[1;36m[$e[1;33m"$e[1;37m$p$e[1;33m"$e[1;36m]$_$e[1;37mADM:$s$e[0;37mN$s$e[1;36m$b$s$e[1;37mZeit:$s$e[0;37m$t$h$h$h$s$e[1;36m$b$s$e[0;37m$e[1;35m$$$s$e[0;37m$g$s[CODE\]

 

[vefatica]

I wonder if that's why I had no problem with my prompt. I have $e[?25h at the end (cursor on).

 

[rconn]

• The PROMPT ends with a foreground color SGR while bold (1) is still active. When TCC then emits a 4x background code for the black input background, the bold flag pushed the background to gray 0x808080.
• Windows Terminal ignores bold for the background → renders black.
• Non-ConPTY tabs use real console cell attributes
• $p$g never turns on bold, and ending the prompt with a non-color sequence changes the trailing bold state — so neither reproduces the bold+background combination.

I have made a change for build 69 to match Windows Terminal's behavior and ignore bold for backgrounds.