In article <BA8D75F2.2FB1D%monroe@criticalpath.com> you write:
>What's the best way to deal with this?  I've currently got a ~/.termcap file
>containing selected entries created from the ncurses terminfo.src using
>infotocap.  This mostly works, but seems pretty cumbersome.  It also doesn't
>work for some of the variants, including "nsterm-c-s-acs",the one I'd prefer
>to use.  
>
>(By "doesn't work", I mean that programs which use termcap give either
>"tgetent failure: No such file or directory", fail to initialize the
>terminal (possibly the same failure with a different error message), or dump
>core.)

Dumping core just because you've got a termcap file they don't like is
wrong, and it might be an idea to file a PR against those programs that do
that.

>What's the netbsd policy on termcap updates and/or terminfo support?  Should
>the netbsd termcap file be brought up to date?

In theory, our termcap is derived from ESR's one, and
<http://www.netbsd.org/Documentation/software/3rdparty/> doesn't indicate
that there should be any problem with importing new ones.  No import has
actually happened since 1997, though, so it's likely to be a rather painful
process.

>Do the historical length
>limits on termcap entries apply to the netbsd implementation?

I believe there's a somewhat evil hack to work around this limit.  From
src/doc/HACKS:

        In order to extend the termcap string over 1023 bytes, a ZZ entry was
        introduced to point to a memory location containing the full entry.

-- 
Ben Harris                                                   <bjh21@netbsd.org>
Portmaster, NetBSD/acorn26           <URL:http://www.netbsd.org/Ports/acorn26/>