GraphQL is a specification for querying a slice of an application graph, retrieving a tree of data that perfectly matches a front-end view, regardless of where that data was pulled from. It covers tree-based read queries, mutations for updates, and subscriptions for live updates.
GRANDstack is devoted to facilitating GraphQl stacks using neo4j. I can't believe that subsriptions aren't supported. How can we implement them?
This is similar to this question answered earlier. Neo4j supports publishing changes via APOC Triggers. And you could still use GraphQL, but perhaps the subscription would be against a queue or topic in Kafka?
I have a demo setup using Kafka and subscriptions. Currently the only implementation is using Clojure, so the JVM. But the same kind of thing can be done with node. I'm still thinking about improving the current setup, and also want to experiment with other databases. Currently it's not really scalable, and has a capacity around 300 new subscriptions. Not sure what the total limit of subscriptions is. Repo is here, https://github.com/openweb-nl/kafka-graphql-examples
Hey @Gerard_Klijs. Is there any update about your Kafka and subscriptions. I'm trying to implement subscriptions with Grandstack and I'm stuck. I took a look at your Github repo but I couldn't wrap my head over how to do it with a Grandstack project.
I'm sorry haven't got to it. So many interesting things happening. I got to speak at Kafka Summit, but that took a lot of time, as upgrade of a Rust library was part of it.
I'm not familiar with the Grand Stack enough to know for sure how it should work. I would hope by adding the events from Kafka to Neo4J with the right properties it should 'just' work. And I really like to see it for myself still.
Can't promise anything though, already learning a lot of new things at the moment for a new assignment, so hard to learn additional things after work.
No worries Gerard, I understand that priorities come first and Plan B distracts from Plan A.
You have already done a great job at starting and experimenting with Kafka and subscriptions, soon someone else will come and take inspiration from your work and continue from there.
As for now, I wish you good luck with whatever you're doing, and thank you for replying .