Full path mapping not working for Taxonomy values

Question

Hi Friends,
I'm trying to update a main table field, which is of type Lookup [Taxonomy]. The taxonomy table has a 5-level hierarchy, with two columns, Code (Display Field 1) and Name (Display Field 2). There are no attributes maintained.
001, A   [Remote Key : 001] <ROOT NODE> {Level 1}
|---000, A1   [Remote Key : 000]  {Level 2}
      |---001, A2   [Remote Key : 001]  {Level 3}
            |---001000, A3   [Remote Key : 001000]  {Level 4}
            |---000102, SYST [Remote Key : 000102]  {Level 5}
            |---001001, A4   [Remote Key : 000101]  {Level 4}
            |---000102, SYST  {Level 5}
|---001, B   [Remote Key : 001]  {Level 2}
      |---001, B1   [Remote Key : 001]  {Level 3}
            |---001000, B2   [Remote Key : 001000]  {Level 4}
            |---000102, SYST  {Level 5}
            |---001001, B3   [Remote Key : 000101]  {Level 4}
            |---000102, SYST  {Level 5}
The figure, unfortunately doesn't come up right, so I've indicated the levels of the nodes.
The remote keys have been maintained at all the levels. I've also set my Configuration Options settings for Hierarchy Values Automap to Fullpath. However, when I try to do an "Automap", the values are being mapped to the wrong leaf node, taking only the leaf node remote key, rather than following the full path. The leaf node value (000102,  SYST) repeats multiple times, and the remote key could be maintained only for one of them. But instead of following the hierarchy, the mapping happens to the value for which remote key was maintained, here  001->000->001->001000->000102, no matter what the hierarchy is.
Am I missing something? Any help would be appreciated.
Regards,
Anil Madhavan
Edited by: Anil Madhavan on Jan 20, 2010 8:48 AM
Edited by: Anil Madhavan on Jan 20, 2010 9:25 AM

Answer 1

Hello Anil,
As for Hierarchy field using leaf nodes only for matching - this is the current design and from memory allocation point of view, I think it is reasonable (keeping the full path to the Node will use huge amounts of memory).
they way workaround this is by adding the code field to the matching step, this will allow you to make sure that the currect nodes are being mapped.
Hope this information helps,
Idan