Discussing Issues and Improvements for Typesense

👋 hola! Here's some ideas/issues I ran into in my first week of messing around with Typesense, most of which i'm guessing have already been discussed so I'm more just looking for links...
• Nested objects and arrays of objects (even if they're not searchable/filterable).
• As an alternative to first-class nesting, maybe an option to format documents on the way out (i.e. store em flat, enhance/expand structure on the way out).
• Rate limiting? Key-specific or across the board.
• Google results for older versions? Love the version picker in the docs but it's easy to end up on an older version without noticing.
• Sorting on fixed length strings! (already found this, just noting that i'm excited about it 😁)

Thanks for the feedback! We are certainly going to address these gotchas.

> even if they're not searchable/filterable
You can already store non-indexed nested fields in Typesense. Typesense will just store any document you throw at it verbatim and use the schema to only index the fields mentioned in the schema.

Ah gtk, there's nothing special to do that? Do I leave them out of the defined schema or something?

Correct. Say you have a document with N fields. Just mention K of them (which must be flat) in the schema. Typesense will index those but won't care about what the other fields are -- but the document will be stored on disk and will be returned fully. I think we should make this clearer in the documentation.