Consider defining sort that ldap implementations can sort efficiently

[GoogleCodeExporter]

There is a mismatch between how SCIM sorting is specified and how LDAP sorting 
works, which makes it hard to implement SCIM over LDAP efficiently.

The SCIM specification says:

For all attribute types, if there is no data for the specified sortBy
value they are sorted via the 'sortOrder' parameter; i.e., they are
ordered first if ascending and last if descending.

The LDAP sort control specification says the opposite:

An entry that meets the search criteria may be missing one or more of
the sort keys. In that case, the entry is considered to have a value
of NULL for that key. This standard considers NULL to be a larger
value than all other valid values for that key. For example, if only
one key is specified, entries which meet the search criteria but do
not have that key collate after all the entries which do have that
key.

Could the SCIM specification be changed so that resources with no data for the 
sortBy value are ordered last if ascending and first if descending?

Original issue reported on code.google.com by treydr...@gmail.com on 27 Oct 2011 at 8:22

[GoogleCodeExporter]

Original comment by treydr...@gmail.com on 18 Nov 2011 at 10:22

• Changed state: Started

[GoogleCodeExporter]

Inverted sort logic:
For all attribute types if there is no data for the specified sortBy value they 
are sorted via the 'sortOrder' parameter; i.e., they are ordered last if 
ascending and first if descending.

Original comment by treydr...@gmail.com on 6 Dec 2011 at 7:31

[GoogleCodeExporter]

Original comment by treydr...@gmail.com on 7 Dec 2011 at 5:31

• Changed state: Done