zoneconcierge: verifier for ProofEpochSubmitted
#246
Conversation
| bestSubmissionData := k.btccKeeper.GetSubmissionData(ctx, sk) | ||
| if bestSubmissionData == nil { | ||
| err := fmt.Errorf("the best submission key for epoch has no submission data") | ||
| panic(err) // this can only be a programming error |
There was a problem hiding this comment.
ProveEpochSubmitted is public function on zoneconcierge keeper and without further context, not having SubmissionData for any SubmissionKey it is not really an error. I would either return empty slice and have the caller to make the judgement about whether panic or not, or make it private helper as close to call site as possible.
There was a problem hiding this comment.
Good point! I have made ProveEpochSubmitted to return an error, then make the FinalizedChainInfo API to panic if ProveEpochSubmitted returns an error.
| // basic sanity check | ||
| if rawCkpt == nil { | ||
| return fmt.Errorf("rawCkpt is nil") | ||
| } else if txsInfo == nil { |
There was a problem hiding this comment.
if you check len of slices (like in next line len(txsInfo) != txformat.NumberOfParts) then this nil check is unnecessary as len of nil slice is 0
There was a problem hiding this comment.
Thanks, that's good to know. I thought len(nil) would lead to panic 😢
| } | ||
|
|
||
| // verify Merkle proofs for each tx info | ||
| powLimit := getPoWLimit(btcNetParams) |
There was a problem hiding this comment.
hmm I do not really like that this is panicking, maybe pass powLimit as argument and make the decision about panic closer to call site ?
Fixes BM-304
This PR provides the verifier function of
ProofEpochSubmitted. The verifier takes the expected raw checkpoint, the twoTransactionInfos, two BTC headers and some other metadata, then enforces the following verification rules:TransactionInfois valid w.r.t. the BTC headerTransactionInfos is same as the expected raw checkpoint.This verifier function will be a part of the verifier for the CZ timestamp. This PR also comes with a fuzz test.