EigenDA Proxy + Historical Nitro Changes #11
Conversation
…ck in to see what breaks the build
| ) ([]byte, error) { | ||
| // we start from the 41st byte of sequencerMsg because bytes 0 - 40 are the header, and 40 - 41 is the eigenDA header flag | ||
| // we use the binary domain here because this is what we use in the derivation pipeline | ||
| return eigenda.RecoverPayloadFromEigenDABatch(ctx, sequencerMsg[41:], e.eigenDAReader, preimages, "binary") |
There was a problem hiding this comment.
Knit - domain could be enum
| if use4844 { | ||
| methodName = sequencerBatchPostWithBlobsMethodName | ||
| } | ||
| if useEigenDA { | ||
| methodName = sequencerBatchPostWithEigendaMethodName | ||
| } |
There was a problem hiding this comment.
knit - could be useful enforce that use488 and useEigenDA are never both true
There was a problem hiding this comment.
there are cases where we would want both flags to be true to be able to specify between using 4844 as fallback or calldata imo
| @@ -152,16 +152,17 @@ func (r *BlobPreimageReader) Initialize(ctx context.Context) error { | |||
|
|
|||
| // struct for recovering data from preimage, impl interface EigenDAReader | |||
|
|
|||
| func (dasReader *PreimageEigenDAReader) QueryBlob(ctx context.Context, ref *eigenda.EigenDARef) ([]byte, error) { | |||
| dataPointer, err := ref.Serialize() | |||
| func (dasReader *PreimageEigenDAReader) QueryBlob(ctx context.Context, cert *eigenda.EigenDABlobInfo, domain string) ([]byte, error) { | |||
There was a problem hiding this comment.
knit - we use blobInfo and cert for the same variable reference throughout. Might be useful to converge on a single naming scheme to disambiguate abstraction
There was a problem hiding this comment.
I prefer blobInfo because it's more meaningful in the context of EigenDA, but i know cert is the term used in the eigenda-proxy - would love @teddyknox to weigh in here
| data, err := daReader.QueryBlob(ctx, blobInfo, domain) | ||
| if err != nil { | ||
| log.Error("Failed to query data from EigenDA", "err", err) | ||
| return nil, err | ||
| } |
There was a problem hiding this comment.
should we be verifying the data after its been read?
There was a problem hiding this comment.
I believe theeigenda-proxy does the verification for us now, should we still include a second verification in when querying the blob?
|
|
||
| abi, err := abi.JSON(strings.NewReader(sequencerInboxABI)) | ||
| if err != nil { | ||
| panic(err) |
There was a problem hiding this comment.
lets remove the use of panic and return errors
| "github.com/ethereum/go-ethereum/log" | ||
| "github.com/ethereum/go-ethereum/rlp" | ||
| ) | ||
|
|
There was a problem hiding this comment.
can we not just import the existing proxy client?
There was a problem hiding this comment.
unfortunately we can't import the existing proxy client as there's a bit too much dependency hell for us to resolve because of differing go version requirements and conflicting go-ethereum versions across nitro, optimism, and the golang eigensdk
teddyknox
force-pushed
the
afk/new-eigenproxy-integration
branch
from
0a3cb87 to
c5d1db9
Compare
No description provided.