Persistent connection

[ruslandoga]

👋

Congratulations on https://auxten.com/chdb-is-joining-clickhouse/ 🎉

I wonder if there are any plans to allow keeping a persistent connection to ClickHouse in ChDB like #108

[ruslandoga]

My main use-case is bundling ChDB into Plausible and replacing our current HTTP client with the one based on ChDB to make self-hosting easier (making it a single container).

[ruslandoga]

ChDB would then be used to write into MergeTree tables and merge them (#107) and also allow parallel reads. I think both should be possible if a single server instance is started with background pools and query_stable or similar calls are dispatched to the pools like in clickhouse server

[ruslandoga]

I'd be happy to try and work on this (for real this time) :) But I have zero C++ experience.