Blockdb integration poc #1079
Blockdb integration poc #1079
Conversation
There was a problem hiding this comment.
I'm really glad you tried to match interfaces as much as possible. The fact that blockDb can only store one block at any height is pretty prohibitive right now - I think some of this would be resolved with SAE, but @StephenButtolph would have more details on this
| snaptree *snapshot.Tree | ||
| } | ||
|
|
||
| func ReadHeadBlock(db ethdb.Database, blockDb ethblockdb.Database) *types.Block { |
There was a problem hiding this comment.
This function feels out of place here
| if !bc.blockDb.HasBlock(blockNumber) { | ||
| bc.blockDb.WriteBlock(block) | ||
| } else { | ||
| // write block to the chaindb if we already stored a block at the same height |
There was a problem hiding this comment.
This seems really confusing, we only use blockDB to store 1 block at any given height? This seems fine after SAE but is pretty incompatible until then.
There was a problem hiding this comment.
Just a note: we usually use the prefix eth to mean that it was taken from upstream (i.e. go-ethereum). You should drop it from this folder name
|
replaced by #1189 |
2 participants
Why this should be merged
Shouldn't be merged in as is.
This is a POC for integrating a dedicated block database via
blockdband a starting point to start the discussion on how to best integratex/blockdbinto coreth.How this works
Updates coreth to use
blockdbfor block storage instead of the existing key-valueethdb.Database. Block related data like hash -> block number mapping, receipts are all still stored in the eth database. Only the raw block bytes (block header and body) are moved out into the blockdb.Verify.Rejectis called on them.How this was tested