[zombodb]

Author of `pgx` here.

We developed pgx so that we could rewrite "ZomboDB" (https://github.com/zombodb/zombodb) in Rust. ZDB is a custom "Index Access Method" for Postgres. Think btree index, but stored in Elasticsearch.

So that's definitely a thing.

Other ideas might be custom data analytics/transformation/processing engines that you'd prefer to run in the database instead of externally.

Custom data types to represent, for example, street addresses or genetic information.

The only limit is yourself! ;)

[boulos]

> The only limit is yourself! ;)

For folks who don't get the joke, please visit zombo.com before Flash is finally dead :).

[mumblemumble]

Alternatively, there's https://html5zombo.com

[manojlds]

For folks who still didn't get it, like me: https://en.wikipedia.org/wiki/Zombo.com

[Dowwie]

Have you evaluated Tantivy/Toshi as possible alternative to ES? I only ask because of your native use of Rust. PGX seems like a more robust offering than pg-extend-rs. This is a great addition to the community!

Also, v.0.0.3 doesn't create much confidence about stability! I'll keep an eye on the project and wish the team the best.

[zombodb]

re: v0.0.3 -- sure. I just published it last night.

We've been working on it since November last year, and have now fully ported ZomboDB to it.

It's proving out nicely, but keep in mind that Postgres' internals are infinitely complex. Getting safe wrappers around all its "things" is going to take a very very long time.

I'd rather get something that seems very stable now, and continue to iterate on it over time.

[ragerino]

I think the comment was less about your code but rather about your v0.0.3 version number. I guess a v0.99 would imply more stability.

That's why lots of companies have switched to date based versioning, similar to Ubuntu and others.

[zombodb]

I've looked into them. It seems they're designed to work within a single process, and it's not quite clear to me how sharing the underlying data files across postgres backends (even with proper Postgres locking) would work.

That's not say they aren't good frameworks. I'm sure they are. It just seems like they're designed for different use cases.

That said, I have other ideas on this front that I can't talk about today. ;)

[rukittenme]

> The only limit is yourself! ;)

You have no idea how significant of a limit that is... I'm kidding.

Thanks for the response! And zombodb looks very cool. I might bring it up since we have a couple of heavy analytics endpoints right now.