[cifs-protocol] RE: CAR - problem with MS-ADTS docs on possibleInferiors

tridge at samba.org tridge at samba.org
Thu Apr 9 05:01:50 GMT 2009


Hi Hongwei,

 >   We finished changing definitions of AUXCLASS(),POSSSUPERIORS(),
 >   and CLASSATTS() in MS-ADTS.  The updated section is attached.
 >   Please review it and let us know if you see any problem.

I don't think the new definiton of POSSSUPERIORS() makes any sense. In
particular, it defines POSSSUPERIORS(O) in terms of POSSSUPERIORS(O).

Recursion is fine, but only when you recurse with a different function
argument than the one you are defining! Otherwise the recursion would
never complete.

This is what you have:

  Let POSSSUPERIORS(O) be the union of
  O!systemPossSuperiors
  and O!possSuperiors
  and POSSSUPERIORS(C) for all C in SUPCLASSES(O)
  and {all C such that C!subClassOf is in POSSSUPERIORS(O)}

There are may ways to fix this. One that is simple is:

  Let _POSSSUPERIORS(O) be the union of
  O!systemPossSuperiors
  and O!possSuperiors
  and _POSSSUPERIORS(C) for all C in SUPCLASSES(O)

  Let POSSSUPERIORS(O) be the union of
  all C such that C!subClassOf is in _POSSSUPERIORS(O)

I'm sure you could find a neater way to handle this.

Cheers, Tridge


More information about the cifs-protocol mailing list