Robert

Once again, the load test is not affected in any obvious way.

> Whoa that is awesome, a **61% decrease**! It's almost close to the same payload size! Being close the payload size almost seems a bit too good to be true......

I pushed an updated version with a milder dependency footprint. Memory results seem similar. In this version: - switch from text `unknown` to `json` binary encoding for the request body...

There are some options for making progress on this. The current best state is https://github.com/PostgREST/postgrest/pull/2349, which - avoids encodeAsCString copying by switching to `jsonBytes` encoding (so we have non-zero-terminated binary...

Also let's note some options that go beyond these "easy" wins: - We might be able to read the body into a strict bytestring directly. If we get a content-length...

> Does "streaming the body" mean the database connection is taken from the pool and blocked once a http request starts and is still ongoing? Indeed. It's a good point,...

> Just to clarify. The above means that even with the optimizations in the current libraries, we'll always need one extra copy of the request body right? Yes, with the...

@steve-chavez Where do you see that configurator-pg is broken on aarch64-darwin? I doubt that's the case, but if it is, could you file a bug report at https://github.com/robx/configurator-pg/issues. @AxelTLarsson The...

See also https://github.com/NixOS/nixpkgs/issues/137877. I have no idea why a cross-compiling gcc would be used, though.

I just ran into this myself. I saw that `nix flake update` updated my inputs, and it felt very natural to assume I'd be able to restrict that invocation to...