Lev Kokotov

You're describing query caching. I think this is a valuable and dangerous feature: just because the client issues the same query, doesn't mean it will return the same result both...

`immediately` is a very long time in terms of computing. Even if it seems like

> Isn't this at parity with what we have today in that the client has some responsibility for sending something to determine shard? This change is just trying to address...

Agreed! One small issue I ran into so far is ORMs, e.g. ActiveRecord, will start a transaction automatically if you create/save a model, but it won't put the comment into...

Less than a second: ``` time python main.py [200.] real 0m0.390s user 0m0.592s sys 0m0.878s ``` Code attached (sorry, long file because formatting). [main.py.txt](https://github.com/smartcorelib/smartcore/files/9550450/main.py.txt)

This is actually pretty common with poolers of any kind (e.g. PgBouncer). Some settings are competing with others, creating edge cases like this one. There is no way to resolve...

Yes, I've tested this on my end, it works :100:

I hit this on Linux (Ubuntu) as well. I ended up manually adding: ``` #[cfg(target_os = "linux")] let bindings = bindings .clang_arg(format!("-I/usr/include/c++/11")) .clang_arg(format!("-I/usr/include/x86_64-linux-gnu/c++/11")); ``` to my `build.rs`.

Rebase on master to get https://github.com/postgresml/postgresml/pull/1102 which should fix the tests.

> Can we merge this PR? Currently, we are blocked by #939 > > THANKS! You shouldn't be blocked if you set `pgml.venv` in postgresql.conf to point to your virtual...