hypercore-next
hypercore-next copied to clipboard
Hypercore 10 is an append only log with multi-writer primitives built in.
This issue tracks the remaining blockers for bundling and running Hypercore in a browser without shims. In no particular order, originally based on https://github.com/hypercore-protocol/hypercore-next/pull/52#issuecomment-954728654: - [x] #52 - [x] #61...
I think we need a "preupgrade" hook that, if set, runs when the user calls update, allowing them to ensure certain conditions around the upgrade. this would just be a...
snapshot.get(length + 1) should not give a SNAPSHOT_NOT_AVAILABLE error. that should only happen if the index is inconsistent.
It would be useful to have a method to destroy all data stored on the user's device related to a given hypercore, freeing up the user's local storage for data...
(not a bug, but also not needed)
Is there a WIP of hyperspace that uses this new version? Or are there still blockers for that?
Hi! I'm wondering if the hypercore-next protocol is documented somewhere? I'm especially curious about how multiwriter support is being implemented.
By corrupted I mean branched as in [The linear history requirement](https://github.com/dat-ecosystem-archive/DEPs/blob/master/proposals/0002-hypercore.md#the-linear-history-requirement): ``` ... otherwise they will generate branches and "corrupt" the hypercore. ``` Currently `_onfork` is called by default whenever...
Wondering if hypercore can or thinking about supporting middleware for use cases like auth. For example, a hook that checks a signature before accepting a mutation or returning a value....