PATCH Re: piping question

[Bart Schaefer <schaefer@brasslantern.com>]

On Oct 4,  2:01pm, Bart Schaefer wrote:
}
} For who may be interested, the parent zsh is holding open the file
} descriptor that serves as the read end of the pipe.  This happens only
} if you backgrou[n]d the entire pipeline, e.g.
} 
}    cat /dev/zero | xmost &
} 
} So I think we still have a few bugs along the lines of workers/32171 and
} 32176 and perhaps 31919.

This seems to take care of it.  PWS, should I push before 5.0.7?

diff --git a/Src/jobs.c b/Src/jobs.c
index 83a4d96..bd95afb 100644
--- a/Src/jobs.c
+++ b/Src/jobs.c
@@ -1627,8 +1627,10 @@ spawnjob(void)
     }
     if (!hasprocs(thisjob))
 	deletejob(jobtab + thisjob, 0);
-    else
+    else {
 	jobtab[thisjob].stat |= STAT_LOCKED;
+	pipecleanfilelist(jobtab[thisjob].filelist);
+    }
     thisjob = -1;
 }
 
diff --git a/Test/A05execution.ztst b/Test/A05execution.ztst
index 8d256ff..ca97f4f 100644
--- a/Test/A05execution.ztst
+++ b/Test/A05execution.ztst
@@ -217,6 +217,21 @@ F:This similar test was triggering a reproducible failure with pipestatus.
 F:This test checks for a file descriptor leak that could cause the left
 F:side of a pipe to block on write after the right side has exited
 
+  { setopt MONITOR } 2>/dev/null
+  if [[ -o MONITOR ]]
+  then
+   ( while :; do print "This is a line"; done ) | () : &
+   sleep 1
+   jobs -l
+  else
+   print -u $ZTST_fd "Skipping pipe leak test, requires MONITOR option"
+   print "[0] 0 0"
+  fi
+0:Bug regression: piping to anonymous function; piping to backround function
+*>\[<->\] <-> <->
+F:This test checks for two different bugs, a parser segfault piping to an
+F:anonymous function, and a descriptor leak when backgrounding a pipeline
+
   print "autoload_redir() { print Autoloaded ksh style; } >autoload.log" >autoload_redir
   autoload -Uk autoload_redir
   autoload_redir