PATCH: some more job-control-tests

PATCH: some more job-control-tests

[Sven Wischnowsky]

These are collected from some of the mails.

Peter: the file contains the sed-loop from 7056 which didn't work for
you -- could you reproduce the failure?

Bye
 Sven

--- Misc/job-control-tests.old	Tue Jul 13 12:35:30 1999
+++ Misc/job-control-tests	Tue Jul 13 12:00:13 1999
@@ -29,3 +29,48 @@
 # ignoring the error messages from sed.
 # ^Z is more of a problem since you have to catch the sed.
 while true; do sed -e 's/foo/bar/' non-existent-file >/dev/null; done
+
+# Try
+#   ^Z
+#   fg
+#   ^Z
+#   fg
+fn() {
+  local a
+  while read a; do :; done
+  less "$@"
+}
+cat foo | fn bar
+
+# Try
+#   ^Z
+#   fg
+fn() {
+  cat builtin.c
+}
+fn | while read a; do :; done
+
+# Try
+#   ^Z
+#   fg
+#   q
+#   ^Z
+#   fg
+#   q
+fn() {
+  less builtin.c
+  echo done
+}
+x=2; while (( x-- )); do f; done
+
+# Try
+#   ^C
+# This won't work because zcat doesn't tell us that it received a signal.
+# But
+#   ^Z
+#   fg
+#   ^C (probably a second ^C is needed: if the continued zcat is still running)
+# works.
+# (See also the file Etc/BUGS)
+while true; do zcat foo.gz > /dev/null; done
+

--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de

Re: PATCH: some more job-control-tests

[Peter Stephenson]

Sven Wischnowsky wrote:
> Peter: the file contains the sed-loop from 7056 which didn't work for
> you -- could you reproduce the failure?

It's still happening under AIX 3 here --- it takes a few ^C's to get it to
stop.  It doesn't seem to be happening with the same version under Irix 6,
SunOS5 or HPUX 10 --- although the response is a good deal slower, so I
can't be sure it wasn't happening at all.  But the whole thing is weird: if
I redirect the error messages from sed, it stops much more frequently,
though not every time, sometimes with status 1 and sometimes with status
130, which presumably means sometimes it is, after all, interrupting the
shell itself.  Setting a trap shows that it really is behaving as if it
hasn't got the signal.  As before, if it doesn't get it on the first few
tries it seems to hang up completely.  Maybe I can look and see what it's
waiting for, but I have to get it to happen first... oh, great, gdb's hung
up too.  Anyway, it looks like some AIX strangeness, maybe to do with I/O,
so probably we shouldn't worry too much.  Aha! bash has the same problem
here... and not elsewhere.  So it's almost certainly the system.

-- 
Peter Stephenson <pws@ibmth.df.unipi.it>       Tel: +39 050 844536
WWW:  http://www.ifh.de/~pws/
Dipartimento di Fisica, Via Buonarroti 2, 56127 Pisa, Italy