On Sun, Aug 28, 2011 at 7:38 AM, Adam Kocoloski <[email protected]> wrote: > For what it's worth, I'd prefer to upgrade the _local tree to use > #full_doc_info records, store the security object as a document in the _local > tree (but cache it when the #db is first opened), and preserve the current > _security API with "clobbering/auto-rev-bump semantics".
Couch would also have to treat updates to that doc differently. Previously, the ddocs' validate_doc_update()s decide what's allowed but couch would overrule them in this case. I think that's what you're saying. To me, yes, that is an exception or edge case (what Paul called "dirty"); but the big CouchDB picture becomes more coherent. -- Iris Couch
