adamlouis

Thank you! Appreciate it! Glad you noticed some of the details.

I didn't intentionally model the Mongo query language. Some requirements were:

- JSON-based: so I did not have to parse the query

- Easy to extend: so I could add new operators later

- DB independent: so I could swap out SQLite with postgres, mongo, cockroach, duckdb, etc. IF I ever wanted to

- Expressive: so I could do aggregations / group-bys / etc

I considered using more popular standards like the ElasticSearch API but they weren't quite what I wanted. I liked how jsonlogic was basically just an AST. Its primitive structure makes code generation to other DB query languages straightforward.

I started this project with my own "JSON lisp" variant, but then found jsonlogic & used that since mine was _pretty_ close (only lists, no objects).

The top-level fields of my queries look like SQL (select, where, group_by, order_by) - I really like the Honeycomb query UI (https://www.honeycomb.io/query/) - so wanted to be able to support user experiences like that ... but also go further via arbitrary jsonlogic expressions.

Nice! The nested explorer you have there is REALLY slick. Great work!

This is something I had been thinking about - when dealing w/ normalized JSON documents, how might I allow exploration of referenced documents? & how might I encode those relationships is some way that's lightweight?

Thanks for sharing.

Thanks!

Yeah, jsonlogic does not seem THAT popular. A JSON-based solution that looks like an AST means you can skip query parsing & just do generation. I started this project with my own JSON Lisp DSL before going with jsonlogic.

In either case, I like the idea of being able to swap out arbitrary storage layers - sqlite, postgres, mongo, etc. - while keeping the JSON api.

Had not encountered odata - thanks for sharing!

Thanks for sharing! This is great work. Cool to see it's written in Nim!

Nice! Thank you!

By implementing the the ES API you may then be able to re-use parts of Kibana, Elastalert & other parts of the ES ecosystem ... could be cool!

Nice! Glad it resonated. Never quite sure how a project like this will land.

Thanks for sharing those - will check them out. Interested to see what happens as the size of the dataset grows.

I have not looked deeply, but Typesense[1] seems like another interesting project. Similar to ES or Algolia, easy to self-host, & with a seemingly efficient memory & disk footprint.

[1]https://github.com/typesense/typesense

Will look into it - thank you for the tip!

Thank you for checking it out! Glad to hear it might be useful.

Will take a cleanup pass soon - was excited just to ship it.

Hi! About 6 months ago, I built a Spotify integration to:

- save all plays

- save all “Discovery Weekly” playlists

- save top songs & artists of the last month, 6 months, and all-time & observe how each changes over time

This weekend, I wrote an article with some technical descriptions & learnings. Sharing my work here!

Thanks!

app link: https://spotify.suczewski.com

Thank you! Definitely a bit overboard.

`returnif` is fun, but there's certainly value in the simplicity of restricted, opinionated syntax (e.g Go).

Thanks! Agreed on the cost/benefit.

Was mostly a fun project to prove out & build a new idea. No plan ahead for this, but always new projects in the pipeline!