This is the mail archive of the cygwin@cygwin.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: remapping Cygwin 'bash' readline functions to PC keys


re: using 'rxvt' to be able to discriminate
    between <C-left-arrow> ^ <left-arrow>, etc. in
    Cygwin 'bash' 'readline'

John Mapoles, et al.,

Are you saying that I should try using use 'rxvt'
in a console window as the interface to *Cygwin*?

Remember, I'm not interfacing to another host when
I run 'bash', in this instance. 'bash' is *not*
remote.  It's *not* running on *u*x. It's running
on the local W2K or WXP machine running Cygwin.

(BTW, I'm sorry if you knew this, but I've gotten
such strange responses from some folks to this
line of queries that I can't help but wonder how
many really understand what I'm asking.)

Can I only do that if I run the 'rxvt' server
(daemon) too?

And, if so, would I still be able to launch
windowed (GUI) Windows applications from a 'bash'
command prompt?

If this would actually work, what percentage of my
MPU, am I throwing down a rat hole? ;-)

If this were to work, wouldn't that mean that
'rxvt' server or client was looking at the scan
codes?  So 'rxvt' server operates at a lower
level than the console handler?

Or, is all of this a problem with the
implementation of the terminal (console) handler
(or, terminfo translator) in Cygwin?

Or, am I more confused than I thought I was?

BTW, running 'ssh' in a Cygwin window to a Linux
machine, does not fix the control/naked arrow key
discrimination problem and resetting the 'TERM'
environment variable on Cygwin to anything other
than 'Cygwin' generates errors in applications
like 'man' that use screen character cell control
and doesn't fix the problem for my terminal
window for the remote machine (I think).

See below.

Thanks,

Lee

At 2003-02-20 02:03 PM -0800, john mapoles wrote:
>Lee,

>This is very similar to a problem that I've had in
>Vim.  I build a library of maps from the Vim actions
>to simpler actions, i.e. I like to map F4 to delete
>line ( F4 is usually ^[[[D on a windows box ).  On
>most UNIX platforms things like right-arrow and
>control right-error are different.  As you pointed out
>this is not so on a windows box.  I've just begun to
>play with rxvt.  This has the feel of a UNIX xterm but
>is also similar to a command window.  But, rxvt uses a
>more robust map.  On my machine I get:
>    -->    ^[[C
>   ^-->    ^[Oc
>rxvt can probably re-map any key by the correct choice
>of termcaps (something I do not know how to do).  Try
>rxvt.

BTW, my guess is the problem is not the map. but
whether some level of console handler looks at
the character output from the keyboard rather than
the scan codes, and whether the Cygwin DLL writers
want to go down to the right level?

At any rate what is the local 'terminfo' database
on Cygwin for? ssh in? rxvt in? Incoming 'telnet'
traffic?

>jm  



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]