[Python-Dev] test_minidom non-failure failure? (take 2)

[Neil Schemenauer]

On Fri, Oct 13, 2000 at 01:33:57AM -0400, Barry A. Warsaw wrote:
> I think (but don't really remember) that LEAK_DEBUG had
> different semantics than it does now.

Nope.  LEAK_DEBUG prints to stderr information about all garbage
the GC finds, even stuff the GC can free.  The thinking was that
GC would be an option and people would want to find applications
that leak when only using reference counting.

> No matter.  This was useful when gc was spankin' new and only
> semi-tested.  It's probably fine to turn off -l in "make test" by
> default for the 2.0 release.

I've left -l enabled for now.  Detecting tests that create
uncollectable garbage is probably a good thing.  It doesn't cost
much and the message should be fairly clear now.

  Neil