Several small issues

[Martin Pohlack]

Hi Marc,

first of all, thank you for your answers.

[...]

> Seems like a bug in midnight commander: midnight commander enables
> application keypad mode but doesn't cope with generated sequence. Worse,
> actually, it eats part of them and outputs others.
> 
> For example, numpad-* creates ESC O j, but mc eats everything but j.
> 
> 
>>How could I fix this?  (Btw., it works in xterm and konsole, so my mc
>>seems to be fine.)
> 
> 
> xterm and konsole don't emulate the vt100 app keypad mode and instead
> generate * etc. all the time.
> 
> You can break vt100 emulation if you want and remap the keys to * etc. if
> you like, have a look at the keysym resource.

This sounds reasonable to me, I will do this (in hope of not breaking
other apps this way ...).

>>2) I use minicom to talk to other machines over serial line.  In konsole
>>and xterm, minicom's history is scrolled out into the konsole's history
>>(the windows scrollbar size shrinks).  The terminals secondary screen
>>seems not to be used.
> 
> 
> If the secondary screenm isn't used then an incorrect terminfo is being
> used.

Gentoo seems to use the upstream terminfo entry.

> I can verify that the minicom in debian correctly switches to the
> secondary screen as the terminfo tells it so.

I'm not exactly sure how to verify that it does not use the secondary
screen, but the terminal's background is destroyed by starting --
stopping minicom.  This does not happen e.g. with less.

I talked to the minicom maintainer yesterday about this and he said that
minicom does _not_ utilize the secondary screen.

Maybe you have an enhanced (read patched) debian version?

> You again should have a look at the manpage (just search for scroll). In any
> case, read the manpage! Really, do it!

Rest assured that I did this (for quite some time) before posting here
(my posted config might indicate this too).

> You can switch the secondary screen on and off and scorlling, too, to your
> liking.

Yes, I tried this of course but it did not make a difference (which led
to my initial conclusion about minicom not using the sec. screen).  I
tried every combination of [+-]ssr and [+-]ssc.

I would like to suggest a slight improvement to the manpage here (and
other similar parts):

...
      -ssc|+ssc
           Turn on/off secondary screen (default enabled); resource
           secondaryScreen.
...

The order "-ssc|+ssc" and "on/off" is either wrong or the option naming
is misleading.  '-' should be associated with turning something 'off'
and '+' with turning something 'on', both should be listed in the same
order to reduce confusion -> off/on

> As both default to "enabled", I guess you run into the problem of minicom
> displaying the status bar. 7.0 will only scroll lines into scrollback if
> its a full-screen scroll (i.e. bottom row is included). 7.1 will very
> likely restore the old behaviour, though.

I suspected something like this after reading this:

http://lists.schmorp.de/pipermail/rxvt-unicode/2006q1/000177.html

>>Fiddling around with the TERM environment variable (xterm, vt100) did
>>not seem to make a difference.
> 
> Fiddling around with TERM is generally a very bad idea. It will destroy
> your keymap and result in at least suboptimal behaviour.

Yes, of course.  But for testing whether the usage of another terminfo
entry would make a difference it seemed like a good idea.

>>How could I fix this?
> 
> 
> First of all, _read the manpage_ :)

:-) Thanks again, but I don't see how spending even more time in the
manpage would have helped me here.

Cheers,
Martin