Unable to use Bigtable Query object as cache key

[rculbertson]

Hello!

We'd like to cache Bigtable query results with the Caffeine library, using Query as the cache key. That's difficult to do right now because the Query object is modified when the request is executed. During the call to toProto, the table name and profile ID are set on the ReadRowsRequest builder. This means the Query object has a different hash code after the request executes. So a subsequent request with the exact same query parameters will be a cache miss.

One solution would be to make a copy of the Query object before executing the request, and use that as the cache key. But I don't think it's currently possible to copy a Query object. I've submitted PR #5551 which allows users to copy Query objects.

[igorbernstein2]

Hey man, long time no speak :)

Sorry for the limitations of the current surface. The api was designed to be a dsl rather then a value object, but I can see the issues that this is causing you.
Although the approach in your PR fixes the issue, I would prefer to take a different approach: I think it would be better to remove the side effects of toProto() and make the Query cloneable instead.

I'll try to address this next week

[rculbertson]

Hi! Agreed, that sounds like a better fix. Thanks!