ed 0.5 bug

[John Gatewood Ham]

The Korn Shell INIT package has a shell script mamake that has
this questionable sequence:

if      ( ed ) < /dev/null > /dev/null 2>&1
then    STDED=ed
else    STDED=ex
fi

This causes ed to lock up the box at 99% cpu forever (or until
you kill the process).  This prevents building Korn Shell.

Yes, I know this is a dumb script, but it worked with ed 0.2
and repeated attemtps to get the Korn Shell people to change
this have failed.  Now that I know the problem (it took a while
for me to find this) I can hard-code around it with some patches
on this awful mamake tool.

But which is the correct behavior, the ed 0.2 behavior (exits
immediately) or the 0.[345] behavior that locks up the cup at 99%?

I'm using bash 3.2.15 if that matters.

I am not subscribed to the mailing list; please cc: me.

Thank you,

JGH