Hello, Inoue-san, Sekhar,
Sorry for my late reply, I had to (and have still been) deal with my customer's trouble in the field, was away from work to go to hospital, etc. And I still don't understand psqlODBC well enough to keep up with Inoue-san.
From: Inoue, Hiroshi [mailto:h-inoue@dream.email.ne.jp]
I don't think the cause of the SQLSetPos failure reported is partition row movement.
Yes, the cause is simply that the current SQLSetPos() implementation does not handle table inheritance. For example, if a table called "parent" has some child tables like "child1" and "child2", and only the child tables have actual rows, SQLSetPos() against the result set from "parent" issues UPDATE statements on "parent". It should execute UPDATE on the child tables instead.
Regards
Takayuki Tsunakawa