To address increasing complexity in code that handles signatures, this PR
discards all notion of "different" signature types at the library level. Both
the crypto and accounts package is reduced to only be able to produce plain
canonical secp256k1 signatures. This makes the crpyto APIs much cleaner,
simpler and harder to abuse.
This commit includes several API changes:
- The behavior of eth_sign is changed. It now accepts an arbitrary
message, prepends the well-known string
\x19Ethereum Signed Message:\n<length of message>
hashes the result using keccak256 and calculates the signature of
the hash. This breaks backwards compatability!
- personal_sign(hash, address [, password]) is added. It has the same
semantics as eth_sign but also accepts a password. The private key
used to sign the hash is temporarily unlocked in the scope of the
request.
- personal_recover(message, signature) is added and returns the
address for the account that created a signature.
The account manager was previously created by packge cmd/utils as part
of flag processing and then passed down into eth.Ethereum through its
config struct. Since we are starting to create nodes which do not have
eth.Ethereum as a registered service, the code was rearranged to
register the account manager as its own service. Making it a service is
ugly though and it doesn't really fix the root cause: creating nodes
without eth.Ethereum requires duplicating lots of code.
This commit splits utils.MakeSystemNode into three functions, making
creation of other node/service configurations easier. It also moves the
account manager into Node so it can be used by those configurations
without requiring package eth.
In order to avoid disk thrashing for Accounts and HasAccount,
address->key file mappings are now cached in memory. This makes it no
longer necessary to keep the key address in the file name. The address
of each key is derived from file content instead.
There are minor user-visible changes:
- "geth account list" now reports key file paths alongside the address.
- If multiple keys are present for an address, unlocking by address is
not possible. Users are directed to remove the duplicate files
instead. Unlocking by index is still possible.
- Key files are overwritten written in place when updating the password.
- Manager.Accounts no longer returns an error.
- Manager methods take Account instead of common.Address.
- All uses of Account with unkeyed fields are converted.
The account management API was originally implemented as a thin layer
around crypto.KeyStore, on the grounds that several kinds of key stores
would be implemented later on. It turns out that this won't happen so
KeyStore is a superflous abstraction.
In this commit crypto.KeyStore and everything related to it moves to
package accounts and is unexported.
* multiple passwords allowed in password file
* split on "\n", sideeffect: chop trailing slashes. fixes common mistake <(echo 'pass')
* remove accounts.Primary method
* do not fall back to primary account for mining
- cli: add passwordfile flag
- cli: change unlock flag only takes account
- cli: with unlock you are prompted for password or use passfile with password flag
- cli: unlockAccount used in normal client start (run) and accountExport
- cli: getPassword used in accountCreate and accountImport
- accounts: Manager.Import, Manager.Export
- crypto: SaveECDSA (to complement LoadECDSA) to save to file
- crypto: NewKeyFromECDSA added (used in accountImport and New = generated constructor)
There is no point to using time.Duration if the value is interpreted as
milliseconds. Callers should use the standard multiplication idiom to
choose the unit. In fact, the only caller outside of the tests already
does so.
Account is now always a non-pointer. This will be important once
the manager starts remembering accounts.
AccountManager is now always a pointer because it contains locks
and locks cannot be copied.
