jdb@mordor.UUCP (John Bruner) (08/16/85)
My timing on this is pretty poor. I've been very busy with other
things recently and I'm about to leave for a conference. However,
it is apparent that a large number of people did not get "uw"
when I originally posted it (or did not receive it intact).
Several people pointed out problems with my original sources to
the 4.2BSD programs "uw" and "uwtool". In particular, the code
which found and opened a pseudo-terminal was very dumb. I am
grateful to Guido van Rossum (mcvax!guido) for the changes he
posted to "net.sources.mac".
I have been working on a rewrite of the pty-search code so that
(1) it is modular and shared by both "uw" and "uwtool" and
(2) it is smarter (i.e. it checks to ensure that both the master
and slave sides of the pty can be opened). I didn't want to repost
my original, deficient version, yet I really haven't tested this
new version very much yet.
My solution to this quandary is to post both my current sources
("uw.c", "uwtool.c", "uw.h", "openpty.c", "openpty.h") and my
original sources ("uw.old.c", "uwtool.old.c") along with Guido's
fixes to the original "uw.c". This way, all bets are covered --
if my new code isn't quite right yet then the old versions will
be available. (The file "uw.h" is the same for both versions.)
There are no changes to the Macintosh portion of the program. I've
been taking a break from it. I plan to start looking at it when
I'm back in town in another week. I have a rather lengthy list
of suggested features to think about.
If part of this posting still doesn't reach you, please send me
mail.
--
John Bruner (S-1 Project, Lawrence Livermore National Laboratory)
MILNET: jdb@mordor [jdb@s1-c.ARPA] (415) 422-0758
UUCP: ...!ucbvax!dual!mordor!jdb ...!seismo!mordor!jdb