Concept: Enhanced search modal

[JPPhoto]

Summary

Implements a more powerful search modal that we can expand over time to support additional fields (date, author, etc.)

Related Issues / Discussions

Discussed on Discord as one potential search upgrade implementation.

QA Instructions

Merge Plan

Checklist

• The PR has a short but descriptive title, suitable for a changelog
• Tests added / updated (if applicable)
• ❗Changes to a redux slice have a corresponding migration
• Documentation added / updated (if applicable)
• Updated What's New copy (if doing a release after this PR)

[joshistoast]

Thoughts on something like this?

Something that would be SUPER cool is a baked-in docs search

[JPPhoto]

Thoughts on something like this?

Yeah, but well beyond what my coding + an LLM can do. Feel free to rip out whatever parts of this you want, though! Or treat that as a future iteration of what's here?

[joshistoast]

Thoughts on something like this?

Yeah, but well beyond what my coding + an LLM can do. Feel free to rip out whatever parts of this you want, though! Or treat that as a future iteration of what's here?

Oh funny I didn't see the changes you made already before I posted.

I mean, all the search APIs are already in the UI, so making a modal to bring them all into one place shouldn't be too hard. Maybe one of these days I can work on a proof of concept, or even check out what you've done so far.

[JPPhoto]

I mean, all the search APIs are already in the UI, so making a modal to bring them all into one place shouldn't be too hard. Maybe one of these days I can work on a proof of concept, or even check out what you've done so far.

Please!

[lstein]

I'd add that this would be a great opportunity to do a "search by image similarity". I've been thinking of setting up a vector database in invoke and store embeddings from each output image. We could do a really fast image-based search using something like that.

We could also implement a vector search of the prompts. Wouldn't that be fun?

[joshistoast]

I'd add that this would be a great opportunity to do a "search by image similarity". I've been thinking of setting up a vector database in invoke and store embeddings from each output image. We could do a really fast image-based search using something like that.

We could also implement a vector search of the prompts. Wouldn't that be fun?

That would be sick!

[joshistoast]

This could be an action modal in addition to search, showing available actions for the current image selection. This way people don't need to hunt for buttons when they can just ctrlk and type in "send to" and you'd get options like send to canvas and send to upscale.

[JPPhoto]

Maybe we could iterate on this rather than getting it all in at once? 😅

[JPPhoto]

Just merged in main changes but might have messed up user isolation.