Workspace reloaded by a second client

solargraph --version
0.39.17

Use case: I would like to run solargraph as a daemon and connect to it with different editors sharing the same local repos.

Example: After client1 connects with rootURI: "foo", client2 should be able to quickly connect with rootURI: "foo". The second connection is expected to be fast because the daemon has already traversed this root thanks to the first client.

Observed behavior: The client1 connection is slow as expected (this is a large project - "[ANY] Solargraph initialized (35.790505999699235 seconds"), but client2 triggers the same slow connection (~35s).

Naively, it seems we should be able to detect that client2 is requesting a workspace we've already setup. Please let me know if my intuition is wrong here!

If both clients are connecting to the same server, your intuition should be correct, but this is a use case I haven't explicitly tested. I'll need to try it and get back to you.