[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: Models: Matching Rule Uses



At 07:36 AM 2/14/2003, Kurt D. Zeilenga wrote:
>At 04:58 AM 2/14/2003, David Chadwick wrote:
>>I disagree. I think it is correct. For example, if you want to use
>>typeless matching you can specify just the matching rule without the
>>attribute type, and every AT that has that matching rule in its AT
>>description will be matched against.
>
>In an extensibleMatch search filter, if the type is absent and
>the rule present, then the value is matched against all attribute
>types which support that rule.   The set of attribute types may
>include attribute types which have no defined EQUALITY rule,
>attribute types which have a different EQUALITY rule, or attribute
>types which have the same EQUALITY rule.

This is more correct (in my opinion):
  The set of attribute types may include attribute types
  which have the same or different EQUALITY rule.  The set
  cannot contain any attribute type which has no EQUALITY
  rule as no comparisons whatsoever are possible without an
  EQUALITY rule (per X.501(93), 12.4.5).

The following, however, is still correct (in my opinion).

>Suitability of rule for use against values of an attribute type
>is NOT indicated in the specification of the attribute type
>description for the type, but by specification of a matching rule
>description for the rule.  This, I believe, is quite clear by
>the language used in RFC 2251 and 2252.  In particular,
>
>RFC 2252, 4.2, Attribute Types:
>   Note that the AttributeTypeDescription does not list the matching
>   rules which can can be used with that attribute type in an
>   extensibleMatch search filter.  This is done using the
>   matchingRuleUse attribute described in section 4.5.
>
>RFC 2252, 4.5, Matching Rules: 
>   Values of the matchingRuleUse list the attributes which are
>   suitable for use with an extensible matching rule.