When doing Ethereum signTypedData, and the primaryType="EIP712Domain",
we completely ignore the "message" part and only sign the domain.
According to the community, this is technically allowed by the spec,
and may be used by ETH smart contracts to save on gas.
Test case generated by @MetaMask/eth-sig-util's library.
See: https://ethereum-magicians.org/t/eip-712-standards-clarification-primarytype-as-domaintype/3286
This way AST parsers won't get stack exceeded (because an elif counts as
an else { if {}} to an AST parser apparently?) and so automated tools
will like tokens.py and friends.
Changes many fields to required -- as far as we were able to figure out,
signing would fail if these fields aren't provided anyway, so this
should not pose a compatibility problem.
Co-authored-by: matejcik <ja@matejcik.cz>
The JSON schema of the test cases has slightly changed.
The "curve" field has moved to the "testGroups" level and the X25519 algorithm has been renamed,
which requires adjustments in the parsing code.
A duplicate definition of "ecdh_vectors" has also been removed.
* fix fujicoin params
* add test_fujicoin/test_send_p2tr
* fix test
* fix serialized_tx.hex()
* Recreated based on a new transaction
* make gen, some fixes
Co-authored-by: matejcik <ja@matejcik.cz>
Previously we checked whether the current vendor header and
the new vendor header are the same by comparing the embedded keyset.
What originally looked like a good idea is not that good, because
this disallows us from ever changing the vendor header signing keys
without causing erasure of the storage during the version update.
This commit fixes that by changing the logic to comparing just the
vendor string.
Change of function names is purely cosmetic:
* vendor_keys_hash -> vendor_header_hash
* check_vendor_keys_lock -> check_vendor_header_lock
