Closed xJonathanLEI closed 2 years ago
Had a quick look at the diff, here's what might affect us:
class_hash
, change of signature
type, new acutal_fee
in various response objectsversion
field when signing a tx seems to be doing something now?Thanks for checking out! I took a look and it seems they added the concept of QUERY_VERSION
. I haven't seen them talking about this, but I think this has to do with solving the issue of having to sign a transaction to call estimate_fee
.
My guess is, the idea would be that account contracts will skip signature verification if version
is set to QUERY_VERSION
, such that transactions no longer need to be signed twice like it is right now. They're basically adding server-side support for this. I expect to see account contract implementers to add support soon.
If that's the case, I guess no action is needed from our side until it lands in account contract implementations. It's not illegal to not use QUERY_VERSION
for queries anyways (and we won't benefit from doing so until account contracts integrate it).
The get_state_update
endpoint seems to be down... No way to update test data for those.
It's inconsistency all over again.
class_hash
is added to DeployTransaction
, BUT only on get_transaction
responses. Transactions in get_block
have no such field...
I expect get_block
to eventually catch up and also serve class_hash
. So instead of splitting the type into two, I simply made the new field optional, as a temporary resolution.
It looks like get_block
does now have class_hash
for deploy transaction (see e.g. the sixth one in here), but it should still be marked as optional because old blocks like this one, first TX don't have it 😞
Wow that's a whole new level of inconsistency... Maybe we should report to StarkWare as bug? Need to test a bit more first though.
StarkNet 0.8.1 is about to release. Creating this issue to track integration progress.
See comments below. It looks like we only need to resolve API changes for this version.