Sean Linsley

Yes, `nodes()` was a bare minimum implementation to get the test suite passing (copied from the pg_query Ruby gem). There isn't a foolproof way to know if we've covered all...

I think that's from a missing system dependency. `brew install cmake` might fix it. Once you do get it working, feel free to open a PR to update the documentation!

Since any solution to the stack overflow issue would involve pulling in dependencies that most users won't need, I decided to document this issue in the README so downstream applications...

However, JSON support seems to be much better. Maybe the documentation should point users to `jsonb[]` / `Vec` with examples. ```rs use sqlx::Row; use serde_json::{json, Value}; let a = vec![json!([1]),...

@jim-mlodgenski could you link to documentation showing how to use pg_tle with PL/Rust especially when running in an AWS managed database? I can't find any documentation (or even comments) suggesting...

In this dataset there's an average of 214 numbers per page (assuming page = file, the `Vec` being decompressed). These relatively small files are from timeseries data we collect. I'll...

I'm seeing around a 3.5x compression ratio versus standard compression methods. Compression may be similarly slow, but only decompression speed matters for this use case. FWIW I think that part...

This shouldn't be closed yet. The readme should at least be updated so you don't get more issues like this from confused users. But I'd argue this is a core...

Sorry for the delay, I was busy getting ready to open source our approach to using pco with Postgres: https://github.com/pganalyze/pco_store The `bucket_size.rs` benchmark shows the improved compression size and read/write...

Following up to confirm that your work has helped, thanks! Upgrading the pco_store benchmarks from pco 0.4.2 to 0.4.6 sped up decompression by ~2x for the "10 minute bucket" case...