Github user kaspersorensen commented on the issue:
https://github.com/apache/metamodel/pull/178
wrt 1) OK this is a sure improvement point. Do you think it's required at
this point, or can we use this module without it, and add this feature in a
later PR. I might just go ahead and do it, but I don't feel like committing to
it right now since I don't know when my time permits doing it.
wrt 2) Again I would argue that the same holds true for e.g. Cassandra.
Some of the databases we support are simply not always well fit for a SQL like
abstraction which MetaModel is. But OTOH there are also plenty of cases
(including a project that I'm on at work) where Kafka is used not because it's
massive volume, but because it has other compelling features (replication, load
balancing, auto purging). In such a case I really almost don't mind if I end up
scanning the complete stream to find the record that I'm looking for. It's such
a rare event and not critical in sense of performance, but giving me the
MetaModel module provides me with an easy abstraction to implement it.
---
