hi! i recently became aware of pathom and really appreciate the way resolvers compose and let the consumer worry less about how the data is connected / how to navigate to their desired data from what they already have. That said, I’m not a fulcro user and am only looking for read support, not mutations and some of the other things. I have some ideas of what the api for a more minimal / standalone resolver library might look like but i’m struggling with how to go about pathfinding between provided input and desired output. Is there any literature / tips on how to implement that? I’m familiar with basic graph theory but perhaps not how best to model this. could someone who is familiar with it discuss how pathom achieves this with me? I’m guessing a lot of it has to do with what the pathom index looks like.

hello @rutledgepaulv, welcome 🙂 the path finding of pathom was a pragmatic implementation, I didn't based on any formal literature (although I had past experience writing Dijkstra algorithms), and you guessed right about the index, it has a lot to do with that. There is a section in docs explaining the indexes, had you checked that? https://wilkerlucio.github.io/pathom/#_understanding_the_indexes

about the finding, the heart of it you can find this section in connect: https://github.com/wilkerlucio/pathom/blob/master/src/com/wsscode/pathom/connect.cljc#L619-L659

thanks! No i haven’t looked at those yet. I started trying to write some of the pathfinding stuff and realized I didn’t know how it was working lol

and that said, I like to just clarify that you don't need to use Fulcro to use Pathom, its a stand alone library, a parser in the end is a function, you can create it and call in any clojure or clojurescript program, no need to integrate with any visual thing. but if really wanna get on the core and write your own I'll be happy to talk about impl details 🙂

thanks. I wasn’t sure how connected (ha) it was to fulcro

also just curious about how to implement such a thing though

Fulcro and Pathom have a very good synergy, but are independent things

ok

if you can understand that snippet I sent, its pretty much all the work on the path finding, based on the index it goes from an input set to some output attribute, it returns all possible paths

perfect

I’ll look through those and perhaps follow up with more questions

@wilkerlucio Are you open to a PR for https://github.com/wilkerlucio/pathom/issues/86? I'd love to see a solution for it, and I'm happy to do some legwork to make it happen. If you are, it would be great to discuss what you'd like a solution to look like at a high level so I'm not working in a completely different direction. If not, no worries. 🙂