Re: [screen-devel] terminfo smacs feature, and screen "charset" command

[Alain Bench]

Hello Micah, and welcome on board,

 On Thursday, June 19, 2008 at 16:58:39 -0700, Micah Cowan wrote:

> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=144178
>> screen only understands \E(0 and \E(B. It also emits these sequences
>> to instruct the terminal its running in to move in/out of the ACS.

    Not exactly:

 - Screen understands ^N/^O once it has been correctly initialised by
enacs=\E(B\E)0. Of course box drawing apps are expected to send enacs
once, at the beginning, before any smacs/rmacs.

 - Screen then translates ^N/^O to the underlying terminal's
smacs/rmacs, whatever they are, and sends this translation.


>> with TERM=xterm, as infocmp shows that the correct sequences for this
>> terminal type are ^N/^O.

    Bizare. In the latest terminfo from 2008-04-29:

| $ infocmp -1 xterm | grep acs=
|       rmacs=\E(B,
|       smacs=\E(0,


> [Micah] My response:
>> the fault is in the terminfo description for screen.

    Let's see:

| $ infocmp -1 screen | grep acs=
|       enacs=\E(B\E)0,
|       rmacs=^O,
|       smacs=^N,

    This looks perfectly right to me, and should work. I don't
understand the original problem, or rather the reported facts don't seem
to be a problem.


>> in UTF-8 mode, screen apparently uses the Unicode box characters,
>> rather than the parent terminal's smacs feature.

    Some UTF-8 terminals don't accept mode changes, on the principle
that UTF-8 is modeless. That's why all applications drawing boxes have
to special case UTF-8... Screen does the job for the few poorly written
apps not doing it themselves.


Bye!    Alain.
-- 
Unicode Integrism?
See the archives for more discussion on why this should,
like hydrogen for dirigibles, be relegated to the past.
        PCC DTG on MU. © August 2004.