X-Spam-Check-By: sourceware.org
Message-ID: <441F673C.8000206@fastmail.fm>
Date: Mon, 20 Mar 2006 21:38:52 -0500
From: Charles Wilson <cwilson AT fastmail DOT fm>
User-Agent: Thunderbird 1.5 (Windows/20051201)
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
Subject: Re: rxvt and line-drawing characters
References: <du56hr$aip$1 AT sea DOT gmane DOT org> <4406222B DOT F2EB92DA AT dessent DOT net> <du5pao$tnr$1 AT sea DOT gmane DOT org> <440669A1 DOT 65F0752C AT dessent DOT net> <4406767C DOT 5030109 AT cwilson DOT fastmail DOT fm>
In-Reply-To: <4406767C.5030109@cwilson.fastmail.fm>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com

Charles Wilson wrote:
> Strangely, if I set TERM=rxvt-cygwin-native , use luconP as my font , it 
> does NOT matter whether my CYGWIN variable has codepage:oem, 
> codepage:ansi, or neither.  The following behaviors are the same:
> 
> ascii.exe prints the line draw characters
> pstree -G prints garbage
> 
> It *could be* due to the fact that pstree uses ncurses, and ascii does 
> not.   Checking...
> 
> Hmmm.  /usr/lib/ncurses/test/ncurses.exe (part of the ncurses-demo 
> package) turns the screen black-on-black when I try to check the 
> line-draw characters.  So I can't even tell if ACS chars with ncurses is 
> broken -- first I gotta figure out why the colors are getting scrogged 
> by the test program!  Somebody, like the ncurses maintainer, should fix 
> that.
> 
> Oh.  That'd be me, then.  Ooops.  I'll try to look into it soon, but PTC.

Okay, so I've looked into this.

(1) the ncurses "problem" is not a problem.  The ncurses.exe test 
program leaves the decision as to whether to use colorfgbg() defaults up 
to the user (-a fg,bg to explicitly specify defaults, -d to inherit 
defaults from rxvt, or '' to use the DEFAULT defaults.  Which is, 
apparently, black on black -- at least when run in an rxvt shell.)

So, running ncurses.exe with -d, (or, for that matter, the hideous -a 
1,4 == red-on-blue, unless you've redefined rxvt's colors) works fine. 
With

   rxvt -fn 'Lucida ConsoleP-16' -tn rxvt-cygwin-native -e /bin/bash

I get nice line draw characters from the ncurses.exe test program.  With

   rxvt -fn 'Lucida Console-16' -tn rxvt-cygwin-native -e /bin/bash

I get "european" characters where the line drawing ones used to be. Just 
as expected.  Also, codepage:oem|ansi has no effect -- it only applies 
when using the windows console, not when using rxvt.

(2) So why does pstree misbehave? Because it's hardcoded to use VT-100 
control characters for line drawing.  But that's EXACTLY where rxvt 
differs from VT-100 (and where rxvt-cygwin differs from rxvt, and where 
rxvt-cygwin-native differs from them all).

pstree SHOULD be using tgetent() calls to dynamically obtain the 
appropriate line-drawing character codes and escape sequences from 
ncurses/terminfo.  But it doesn't.

(There's also a similar bug in that it doesn't always calculate 
accurately the width of strings in which these line-draw chars are 
embedded...but that's a whole 'nother thing.)

Basically, pstree -G should be rewritten; it's just not going to work on 
cygwin's rxvt (native or X) very well.

"It's surely evil code"
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=265552


Basically, you're stuck with 'pstree -A'.

--
Chuck


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/