cmd module documentation lying?

[pwang99c]

i'm using the cmd.Cmd interactive interpreter loop for the first time, using documentation out of the latest _Python Essential Reference_.  (i looked on the web page http://www.python.org/doc/current/lib/Cmd-objects.html but there wasn't as much detail.)

there seems to be a disparity between documentation and actual code behavior with regard to the return values of command functions (do_*).  according to the book, only negative return values are construed as errors and cause the cmdloop() function to return; however, in practice, any non-zero value seems to do this.

--------------------------------------------
import cmd, string

class myInterpreter(cmd.Cmd):
	def do_blah0(self, argline):
		return 0
	def do_blah1(self, argline):
		return 1
	def do_blahNeg(self, argline):
		return -1
	def do_blahArb(self, argline):
		return string.atoi(argline)

x = myInterpreter()
x.cmdloop()
--------------------------------------------

when the above code is run, one can type "blah0" all day long and the interpreter continues running; if one tries running any of the other functions (or blahArg with a non-zero value), the interpreter loop exits.

so... is this a known problem, or am i misunderstanding something?  should i be emailing David Beazley about his book?  or is there some place for more up-to-date documentation on the cmd module? (besides the source itself. :-)

TIA,
peter