Rereading the documentation part which I quoted in my post, I am inclined to conclude that you’ve stumbled upon a bug in either PostgreSQL or pgAdmin.

Maybe take it to the mailing list?

Here’s how a unique constraint shows in my version of pgAdmin:

Thanks for your interest in my old blog post,
Rowan

Comment on a very old post…

I have the opposite situation. From within pgAdmin, I can see a unique constraint, but *not* a unique index. However, it seems the constraint is backed by its own implicit index. Adding an explicit index to this column, however, yielded considerable query performance improvements.

I’m now more confused than I was before reading your post.

Any ideas (if you can remember back that far)?

–jon