[illumos-Developer] webrev for 294, 1149

[Yuri Pankov]

On Tue, Jul 19, 2011 at 10:04:50AM +0200, Joerg Schilling wrote:
> Yuri Pankov <yuri.pankov at gmail.com> wrote:
> 
> > On Sat, Jul 16, 2011 at 10:13:33AM -0400, Gordon Ross wrote:
> > > The new file termcap.src says it's derived from the "master" terminfo.src
> > > so would it make sense to put only the master under SCM and generate
> > > the (derived) termcap one at build time?
> >
> > This would require the `tic' from ncurses able to translate terminfo ->
> > termcap (current one in cmd/tic can't do it).
> 
> As I explained last evening, the tic from ncurses also can't do it......
> 
> it just pretends to do so but outputs code that cannot be parsed by a termcap
> file parser.

Thanks for the feedback, Joerg.

Quoting the 20110723 ncurses patch changelog:

[...]
+ improve generation of termcap using tic/infocmp -C option, e.g.,
to correspond with 4.2BSD (prompted by discussion with Yuri Pankov
regarding Schilling's test program):
	+ translate %02 and %03 to %2 and %3 respectively.
	+ suppress string capabilities which use %s, not supported by tgoto
	+ use \040 rather than \s
	+ expand null characters as \200 rather than \0
[...]

This reduces the number of errors reported by `termcap' to 118:
$ schily-2011-06-22/termcap/OBJ/i386-sunos5-cc/termcap \
if=termcap-20110723.src 2>&1 >/dev/null | wc -l
     118

Could you please take a look at the new termcap.src
(ftp://invisible-island.net/ncurses/current/termcap-20110723.src.gz) and
comment on the issues reported by `termcap' against it?

BTW, `termcap' is segfaulting when TERMCAP isn't set..


TIA,
Yuri