Chungmin Lee

Is this still the case? I can't find where the Spark LogicalPlan is serialized and stored with indexes.

Seems to be a fixed issue since v0.2.

I thought this was fixed in v0.2 because of the following lines in the release note: > In order to better support compatibility across Scala/Spark versions going forward, the team...

We don't have any design yet about how we should support arbitrary plan indexing, and the issue (as @sezruby described, not the original description) seems to be a trivial part...

@rapoth I think it would be an interesting feature to implement. WDYT?

I think a partial index or filtered index is a different thing, because materialization is not required for them. But as we only have covering indexes for now, there is...

Maybe I misunderstood the issue, but why is the order important? Why can't we simply XOR or even just add signatures? I'm asking this because I don't get why "false...

> > Maybe I misunderstood the issue, but why is the order important? Why can't we simply XOR or even just add signatures? I'm asking this because I don't get...

Yeah, we should do proper HEX encoding/decoding when JSON is involved. By the way, I think we should also avoid being locked in MD5 and should make it easy to...

I assume that the signatures are stored as JSON in users' data lake. Therefore, any change to the signature computation will make existing signatures won't work (even if we don't...