tlf-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Tlf-devel] Improved keyboard handling


From: Nate Bargmann
Subject: Re: [Tlf-devel] Improved keyboard handling
Date: Mon, 4 Jan 2016 13:56:59 -0600
User-agent: Mutt/1.5.23 (2014-03-12)

* On 2016 04 Jan 12:58 -0600, Thomas Beierlein wrote:
> Hi Nate, hi Ervin,

Happy New Year, Tom!

> having some common key handling is a great goal - but not so even to
> reach.

I'd be thrilled will 95% coverage of Tlf's keys across most all
terminals.  I understand this is likely an effort that will never be
entirely perfect and I expect the function to grow a bit more to cover a
few more corner cases.  since the tests are fairly well defined I don't
expect speed issues for the most part.

> Thanks for the work Nate. I did some similar stuff some time ago, but
> got distracted by some more urgent problems in tlf. I will try to find
> the old branch and push it to github for reference. 
> 
> I had a similar handling for the ESCDELAY and an activated keypad. Main
> difference - I did only handle the alt-keys by adding the +128.

In some cases a different value needs to be used (see the Xterm section)
to assign the value into the range known by keyname() and define_key()
as listed in the names.txt file.

> If I remember correctly by doing so I was able to map nearly all used
> key combinations in tlf to the new ncurses keys. The only ones which
> did not work where the Ctrl-PgUp and Ctrl-PgDn keys which are not
> handled by ncurses for the xfce terminal.

Those are going to be a problem in some cases.  The Gnome Terminal I
installed had a handy configuration selection to disable all shortcuts
which enabled nearly all the keys.  Alt-Space was still caught by Xfce
for the window menu.

I also found that Xfce Terminal gives the same key code for Alt-A as
with Alt-Shift-A while Gnome Terminal and Xterm gave separate codes for
each.

> I had hoped that with ncurses doing the key decoding that we could drop
> the different handling for different terminal emulations. But that
> seems to be not so.

The good thing is that we can limit the custom coding to just a few
corner cases which is a good thing.  I'm surprised that the current
batch of corner cases is already this large, but results from Ervin and
Pierre have been encouraging.

> Let me add two more terminal settings which we should check - that is
> plain linux console and screen and/or tmux.

I did a quick check on plain Linux Console on my T410 laptop but need to
do so with a 101 key keyboard.  I do use tmux (right now to access Mutt
remotely) so it is also an easy test.  In fact I did some tests this
morning from a computer running Windows 7 and Putty and found that
setting the keyboard entry to use Xterm gave me the F1-F4 keys.  I'd
been scratching my head over that for a long time which was a headache
for using Midnight Commander remotely.  Nice to know this exercise
helped me fix that.

> Maybe it is time to sort it out and get a solid solution. Nate it would
> be nice if you could lead that work. Ervin and me we are willing to test
> and give hints and support. would that be ok for you?

Absolutely.  I'll probably get a bit more done over the next couple of
weeks and get a branch pushed out for wider testing.

73, Nate

-- 

"The optimist proclaims that we live in the best of all
possible worlds.  The pessimist fears this is true."

Ham radio, Linux, bikes, and more: http://www.n0nb.us



reply via email to

[Prev in Thread] Current Thread [Next in Thread]