robotparser behavior on 403 (Forbidden) robot.txt files

[John Nagle]

   I just discovered that the "robotparser" module interprets
a 403 ("Forbidden") status on a "robots.txt" file as meaning
"all access disallowed". That's unexpected behavior.

   A major site ("http://www.aplus.net/robot.txt") has their
"robots.txt" file set up that way.

   There's no real "robots.txt" standard, unfortunately.
So it's not definitively a bug.

				John Nagle
				SiteTruth