Ensure empty withdrawal lists are set in BFT blocks when the protocol spec is shanghai or higher #6765
Conversation
… schedule is shanghai or higher Signed-off-by: Matthew Whitehead <[email protected]>
matthew1001
force-pushed
the
shanghai-qbft-fix
branch
from
b8df9c8 to
91c7f37
Compare
Signed-off-by: Matthew Whitehead <[email protected]>
Signed-off-by: Matthew Whitehead <[email protected]>
| final Block block = blockCreator.createBlock(headerTimeStampSeconds).getBlock(); | ||
|
|
||
| // Determine if we're at a protocol spec that requires withdrawals (even if empty) to be present | ||
| ProtocolSpec protocolSpec = |
There was a problem hiding this comment.
This is duplicated several times. It feels like we can do this in BftBlockCreator by overriding createBlock, checking the withdrawal validator, and switching on that. If the sequence number is truly needed then making a new createBlock method with both args. May need to also do it for PkuQbftBlockCreator since there is no shared parent classes.
There was a problem hiding this comment.
Good thought. I've done some refactoring as BftBlockCreator already has access to the parent header required to get the next block number. So it was a little easier than I thought. Now QbftRound and IbftRound can just continue to call createBlock(...) as they did before, and BftBlockCreator.createBlock(...) can decide which type of block to actually create. There's a little repetition in PkiQbftBlockCreator like you say, but it's definitely tidier now.
matthew1001
force-pushed
the
shanghai-qbft-fix
branch
5 times, most recently
from
036b7e2 to
41bf3e0
Compare
Signed-off-by: Matthew Whitehead <[email protected]>
matthew1001
force-pushed
the
shanghai-qbft-fix
branch
from
41bf3e0 to
082ba98
Compare
Signed-off-by: Matt Whitehead <[email protected]>
… spec is shanghai or higher (hyperledger#6765) * Ensure empty withdrawal lists are set in BFT blocks when the protocol schedule is shanghai or higher Signed-off-by: Matthew Whitehead <[email protected]> * Unit tests missing mock Signed-off-by: Matthew Whitehead <[email protected]> * Unit tests missing mock Signed-off-by: Matthew Whitehead <[email protected]> * Refactor and reduce code duplication Signed-off-by: Matthew Whitehead <[email protected]> --------- Signed-off-by: Matthew Whitehead <[email protected]> Signed-off-by: Matt Whitehead <[email protected]> Signed-off-by: amsmota <[email protected]>
… spec is shanghai or higher (hyperledger#6765) * Ensure empty withdrawal lists are set in BFT blocks when the protocol schedule is shanghai or higher Signed-off-by: Matthew Whitehead <[email protected]> * Unit tests missing mock Signed-off-by: Matthew Whitehead <[email protected]> * Unit tests missing mock Signed-off-by: Matthew Whitehead <[email protected]> * Refactor and reduce code duplication Signed-off-by: Matthew Whitehead <[email protected]> --------- Signed-off-by: Matthew Whitehead <[email protected]> Signed-off-by: Matt Whitehead <[email protected]> Signed-off-by: amsmota <[email protected]>
|
hello, |
|
+following |
… spec is shanghai or higher (hyperledger#6765) * Ensure empty withdrawal lists are set in BFT blocks when the protocol schedule is shanghai or higher Signed-off-by: Matthew Whitehead <[email protected]> * Unit tests missing mock Signed-off-by: Matthew Whitehead <[email protected]> * Unit tests missing mock Signed-off-by: Matthew Whitehead <[email protected]> * Refactor and reduce code duplication Signed-off-by: Matthew Whitehead <[email protected]> --------- Signed-off-by: Matthew Whitehead <[email protected]> Signed-off-by: Matt Whitehead <[email protected]>
PR description
If the protocol spec is as
shanghaior higher, a valid (if empty) withdrawals list must be present in the blocks proposed.Fixed Issue(s)
Fixes #6760