Balance check at container.Put command will drop some invalid
transactions from the side chain network before inner ring
approval.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
In v0.8.0:
- containers were stored as stable marshaled binary,
In v0.9.0:
- containers are stored the same way as eACL,
- eACL structure has new `token []byte field`.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
With contract migration from v0.8.0 to v0.9.0 we need to replace
structures in contract memory. This produces native contract calls
that are not available in scope of `management.Update`. Therefore
we need to apply new scope first.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
There is a number of contracts which return only `true` value.
Also handling `FAULT` on the client is easier then also checking return
value.
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
Container now stored the same way as ExtendedACL: with
signature and session token. This is required for signature
checks when session token presented.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Signature check may fail if container has NFT attributes.
So these checks should be done in alphabet nodes and not
inside the contract.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Unified format uses transfer type as the first byte
and extra details next. List of transfer types used in
contracts defined in `transfer.go`. It includes:
- mint,
- burn,
- lock,
- unlock,
- container fee.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Delete initializing of R/W context in `init()`
and start using readonly storage context in
methods where it is possible.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Change []byte type to interop.*aliasName*
types in contracts where it is suitable to
improve the readability of the contracts.
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
This check was useful when there was a single
entry point in contract. But right now there is
no point in this.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Define InnerRingList function that calls "innerRingList" method of
particular smart contract. Define InnerRingListViaStorage function
that reads address of smart contract from the storage by key, and
calls InnerRingList with the result. Reuse these functions in all
contracts.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Replace getList function to common package and export it. Reuse the function
in Container and Reputation contracts.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Define IRNode structure in common package. Replace innerRingInvoker function
in common package and export it. Reuse this function in all contracts.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Replace vote/remoteVotes functions from all contracts (except alphabet) to
common package. Additionally replace setSerialized and bytesEqual. Create
InitVote function and use it in NeoFS and Netmap contracts.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Create common package. Define Ballot struct in common package. Use new type
in all contracts with ballots.
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
Remove old ballots first so there won't be any
interference with votes of the same tx in the
future after `blockDiff` blocks.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Basic income settlements depends on container estimation that
should be collected in P2P communication between storage nodes
and then stored in container contract. To synchronize these
actions there are two separate notification that inner ring
should produce in consensus.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
EpochProcess method should be invoked by inner ring nodes
and it removes old estimations from contract storage.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
Container contract checks signature of eACL iterating through keys
from neofsid contract.
Clients often ask eACL from storage nodes, but it's kinda hard for
them to check signature since they have no direct access to neofsid
contract. So the container contract can make one more iteration
through keys from neofsid contract and return suitable one.
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
