-;;; FIXME: Very likely we should be scavenging NLX-based dependencies
-;;; here too. OTOH, there's a lot of global weirdness in NLX handling,
-;;; so it might be taken care of some other way that I haven't figured
-;;; out yet. Perhaps the best way to address this would be to try to
-;;; construct a NLX-based test case which fails in the same way as the
-;;; closure-based test case on cmucl-imp 2001-11-29.)
-;;;