Re: Two little features (and a bounty?)

[Micah Cowan]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dan Mahoney, System Admin wrote:
> So from Micah's findings, it would almost appear that based on the
> application, there's a number of things that one needs to do to make
> this work.  (xterm value, fake display, manually send the escape
> sequence, etc).

No; to my knowledge, the only program I've ever seen require DISPLAY to
be set is alpine, which is kind of retarded.

Manually sending the escape sequence will _not_ work, as I mentioned
previously. It will get screen to send the sequences, to be sure, but it
will not put the application into an accepting state (nano -m would not
accept them properly, despite being in "mouse enabled" mode, unless the
term was "xterm").

> This is definitely something that fits well into some kind of
> documentation grid as to what-setup-requires-what...

Probably should go on the wiki. http://aperiodic.net/screen/

> I'm not sure if there's anything in screen that can fix it (although I
> think the window-title hack is a neat answer).

Won't work: that'll be equivalent to what I did with "sending the escape
sequence first"; nano still doesn't interpret them.

Even if it did work, it'd be a bad idea, except for your own personal
use. Even if you never suspend nano to work in the shell, other people
certainly may.

> I am also not sure if
> this should be brought to the nano/alpine/whatnot developers -- there
> are at LEAST a couple situations where you want a mouse without a
> display, i.e. console-access, where screen is ideal (but then, you have
> multiple virtual consoles).

AFAICT, nano doesn't explicitly look for xterm; it may well be an
ncurses issue, rather than nano.

> Let me think on this a bit more -- there's GOT to be a clean way to
> handle this -- Micah, do you think it's relevant to open bugs with the
> appropriate projects?

It can't hurt; I had some success with that in Vim. But adding "screen"
to the list of approved terminals is not a good general solution; it
really needs to be addressed in ncurses/terminfo.

> Or better still, can we come up with a way to define a "test case" for
> mouse under screenable events?
> 
> (I had a coworker tell me I should run gpm in my ssh session, to fix
> this...I wonder how right he might have been...other than that I'm under
> freeBSD).

Does gpm work that way? I wouldn't have thought so, but I don't really
know that much about gpm.

I thought it's intended to work on the actual console terms, though,
with the local mouse.

- --
Micah J. Cowan
Programmer, musician, typesetting enthusiast, gamer,
and GNU Wget Project Maintainer.
http://micah.cowan.name/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIihR07M8hyUobTrERApJTAJ4/QI0rfcHTRvVv1Hl/Qo/eI5bo5gCeNCt4
+ji1+8OfmFUov9sJhJoARLI=
=Hy2p
-----END PGP SIGNATURE-----