Re: warning about closing an already closed file descriptor

[Peter Stephenson <p.stephenson@samsung.com>]

On Tue, 20 Jan 2015 12:02:03 +0100
Roman Neuhauser <neuhauser@sigpipe.cz> wrote:
> > -  ({ exec 3<&- } 2>/dev/null
> > -  exec 3<&-
> > -  read foo <&-)
> > +  (exec {varid}<&0
> > +  exec {varid}<&-
> > +  print About to close a second time >&2
> > +  read {varid}<&-)
> 
> sorry if i'm talking out of my ass, but does this mean that
> the non-{varid} syntax is now without a test?

For closing fd's, yes.  There's no output to test --- though there is a
status, so we could still test that.  And I suppose the lack of an error
message is itself a test...

pws

diff --git a/Test/A04redirect.ztst b/Test/A04redirect.ztst
index cb67788..13f1f7c 100644
--- a/Test/A04redirect.ztst
+++ b/Test/A04redirect.ztst
@@ -152,11 +152,16 @@
 >hello
 >goodbye
 
+  ({exec 3<&- } 2>/dev/null
+   exec 3<&-
+   read foo <&-)
+1:'<&-' redirection with numeric fd (no error message on failure)
+
   (exec {varid}<&0
   exec {varid}<&-
   print About to close a second time >&2
   read {varid}<&-)
-1:'<&-' redirection
+1:'<&-' redirection with fd in variable (error message on failure)
 ?About to close a second time
 *?\(eval\):*: failed to close file descriptor *