Allow claims outside the claimtrie #398

Open
opened 2021-01-19 19:23:17 +01:00 by BrannonKing · 1 comment
BrannonKing commented 2021-01-19 19:23:17 +01:00 (Migrated from github.com)

The computation of the claim trie is the bottleneck on sync, and it will soon be the overall bottleneck on each new block. Many LBRY streams presently use this mechanism but have no need for claim competition, as they are grouped by channel.

We need to enable a mechanism to allow these streams-in-a-channel to migrate to a non-competitive registration. The proposed method is that we will modify the trie to ignore claims for the root node (aka, claims with an empty name will no longer compete). We will no longer allow someone to competitively claim the root node; there's no way to view it presently anyhow.

Subtasks

  • update LBRY spec to, among other things, remove the requirement that URLs have a name component. Add examples of URLs with only a claim ID.
  • we need a hard-fork plan for this.
  • the hard-fork will need to pull all the claims in root at present out of the trie.
  • should the RPC methods to query claims by name return the entire list of no-names?
  • proof RPC methods will need to not support empty claim names
The computation of the claim trie is the bottleneck on sync, and it will soon be the overall bottleneck on each new block. Many LBRY streams presently use this mechanism but have no need for claim competition, as they are grouped by channel. We need to enable a mechanism to allow these streams-in-a-channel to migrate to a non-competitive registration. The proposed method is that we will modify the trie to ignore claims for the root node (aka, claims with an empty name will no longer compete). We will no longer allow someone to competitively claim the root node; there's no way to view it presently anyhow. Subtasks - [ ] update LBRY spec to, among other things, remove the requirement that URLs have a name component. Add examples of URLs with only a claim ID. - [ ] we need a hard-fork plan for this. - [ ] the hard-fork will need to pull all the claims in root at present out of the trie. - [ ] should the RPC methods to query claims by name return the entire list of no-names? - [ ] proof RPC methods will need to not support empty claim names
lyoshenka commented 2021-01-19 21:07:36 +01:00 (Migrated from github.com)

Which places in the code currently assume that a claim has a name? Will any of those be hard to change?

Which places in the code currently assume that a claim has a name? Will any of those be hard to change?
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: LBRYCommunity/lbrycrd#398
No description provided.