Re: lynx-dev curses on/off and HTTelnet (was: DOS patch for dev.8)

[Klaus Weide]

On Mon, 30 Aug 1999, Doug Kaufman wrote:

> I checked a trace file while telnet'ing. stop_curses() is called
> from getfile(). Within HTTelnet.c, in function remote_session(),
> LYCursesON is FALSE. It isn't clear to me why _eth_release is needed
> to allow MSKERMIT to function properly when called from within lynx,
> since stop_curses() calls _eth_release. This may be something within
> WATT-32, since it wasn't necessary with 2.0 dev.rel.4, but was with
> 2.0 dev.rel.5. I'll try to look, but perhaps Gisle knows what is
> happening. In any case, stop_curses() seems to be called properly from
> getfile().

OK, I mostly just wanted to point out that there was something strange
going on; I was also a bit concerned whether any of my changes had somehow
caused that the stop_curses in getfile() is skipped.  Apparently that's
not the case.

Could the trace log writing itself (having a file open) cause any problem
under DOS?  That's just a wild guess.

  Klaus