Github user chewbranca commented on the pull request:
https://github.com/apache/couchdb-couch-replicator/pull/26#issuecomment-188423675
@kxepal @eiri @rnewson IMO loading design docs from `ddoc_cache` is
obviously the right thing to do. The big question here is whether to support
non clustered databases in `ddoc_cache`.
One alternative approach to calling `ddoc_cache` directly in this scenario
is to use `couch_changes:open_ddoc/2` [1] which would work around the issue
we're seeing here. I think having to qualify loading from `ddoc_cache` is a
hacky solution we should strive to avoid, and which is why I didn't do it
initially, but it might be the pragmatic choice here until we eliminate this
current dichotomy of APIs.
[1]
https://github.com/apache/couchdb-couch/blob/master/src/couch_changes.erl#L341-L355
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
