Re: Ansii auto detection

From: Jeremy (
Date: 04/27/02

On Sat, 27 Apr 2002, Templar Viper wrote:

> Hello,
> I have downloaded the auto ansii detection from the developer
> site, and took at look at it. It sends an ascii code to the client,
> and if it returns the correct response code, the patch assumes
>  the client is ascii (and color) capable. However, that the client
> suports ascii does not always mean that it supports color too.
> The windows telnet client for instance, is detected as ascii
> supportive. Although it recognises the ascii character set, it
> still doesn't have color.

Windows telnet, while it doesn't support colour, doesn't barf on it either
- it just loses the colour information and displays everything as black on
white.  (I think newer windows (2k, XP, etc) have a colour-enabled telnet,
but I could be wrong.)

This may or may not be a problem - the biggest reason to check if a
terminal is ANSI-capable is to avoid control characters being printed in
the mud's output.  Unless you're actually conveying information in the
colour that can't be gotten any other way, sending the colour codes to an
ANSI-capable monochrome terminal isn't really a problem.

> So, my question is: Is there a way to check if the client is
> _really_ capable of displaying colors automaticly? (So not
> by querying "do you want color? Y/n")

Not that I know of (or can think of) except possibly querying what
terminal is being used on the other side and looking up somewhere (termcap
or whatever system your OS uses) whether the terminal can do colour.  A
lot of terminals and mud clients either lie about this or don't understand
the question.

If the terminal lies about whether it can do ANSI, I'd put that as a
problem on the players side for using broken software.  I draw the line at
supporting severely broken clients (unless they're provided standard with
an operating system, like windows telnet).

Just my 2 coppers.

-Jeremy (no, not that one)

   | FAQ: |
   | Archives: |
   | Newbie List:   |

This archive was generated by hypermail 2b30 : 06/25/03 PDT