I tried to see how the backend is implemented, but the link in the post doesn't work for me: https://hivehub.dev/polls
You are viewing a single comment's thread from:
I tried to see how the backend is implemented, but the link in the post doesn't work for me: https://hivehub.dev/polls
That link is a different frontend for viewing all the polls. The code repository is here: https://gitlab.com/peakd/hive-open-polls
It's an open protocol that hopefully maintains enough flexibility that the ecosystem can use instead of everyone developing their own solution that works for their app.
The current backend implementation is that it fetches data from HafSQL every minute. It uses plpgsql throughout (except one piece that uses node.js to fetch balances from a Hive API node).
Hardware requirements are extremely minimal, also no setup is required.
All feedback and suggestions are very welcome.
It's not a major issue, but if I understood correctly, the node.js is for getting balance data. We're planning on making balance_tracker a "standard" app for HAF soon (mostly just waiting for api and doc improvements) as it is already required to run the haf_block_explorer app. So in the future you could potentially replace that with direct calls to the balance_tracker api.
Would be great to have HafSQL part also incorporated, right now it is using mahdiyari's instance which is single point of failure from what we understand.
Yeah, would be great to have more HafSQL nodes.
Yes, I was waiting for the balance_tracker app to be ready. But also, we need to fetch Hive Engine balances (and maybe SPK network balances or other layers that become available in the future) so we'll likely have to keep the node.js part for that. Unless we can figure out a way to do it with just SQL.