RE: workaround for rsh problem

RE: workaround for rsh problem

[Heading, Anthony]

> The following workaround for the rsh problem was in the
> zsh-2.5.03 sources.

That fixes it.  Splendid. Thanks.

Anthony

Re: workaround for rsh problem

[P.Stephenson]

coleman@math.gatech.edu wrote:
> The following workaround for the rsh problem was in the
> zsh-2.5.03 sources.

This is is of course obvious, but you can do it by hand:  I have some
code inside an rsh that does

exec <&- >$logfile 2>&1 3<&- 4<&- 5<&- 6<&- 7<&- 8<&- 9<&-

and it seems to work with all shells with proper syntax.

-- 
Peter Stephenson <pws@ifh.de>       Tel: +49 33762 77366
WWW:  http://www.ifh.de/~pws/       Fax: +49 33762 77330
Deutches Electronen-Synchrotron --- Institut fuer Hochenergiephysik Zeuthen
DESY-IfH, 15735 Zeuthen, Germany.

workaround for rsh problem

[Richard Coleman]

The following workaround for the rsh problem was in the
zsh-2.5.03 sources.  It was removed in the transition from
buildzsh to autoconf.  Currently this workaround is not defined
by default in the zsh baseline, so you will need to add the
RSH_BUG_WORKAROUND to the defines in the Makefile.  The optimal
thing would be to check for this problem in configure, but I don't
know how that could be done.  If we can determine which architectures
for which this is necessary, I'll put a note about this in the
MACHINES file.

rc


*** init.c	1995/10/06 04:08:12	1.55
--- init.c	1995/10/18 03:46:30
***************
*** 318,321 ****
  {
-     long ttpgrp;
      static char *outbuf, *errbuf;
  
--- 318,325 ----
  {
      static char *outbuf, *errbuf;
+     long ttpgrp;
+ 
+ #ifdef RSH_BUG_WORKAROUND
+     int i;
+ #endif
  
***************
*** 333,334 ****
--- 337,347 ----
      setbuffer(stderr, errbuf, BUFSIZ);
+ #endif
+ 
+     /* This works around a bug in some versions of in.rshd *
+      * Currently this is not defined by default.           */
+ #ifdef RSH_BUG_WORKAROUND
+     if (isset('c')) {
+ 	for (i = 3; i < 10; i++)
+ 	    close(i);
+     }
  #endif