EIP161, State Trie Clearing

[steviezhang]

Implementation of EIP161 passing most tests.

The following are the tests that have been skipped that are failing either due to a precondition not being implemented or require more looking into.

See the tracking issue #29 for more details.

    // EIP 684 Implementations
    skipTests["TransactionCollisionToEmptyButCode.json"] = "Not Implemented"
    skipTests["TransactionCollisionToEmpty.json"] = "Not Implemented"
    skipTests["TransactionCollisionToEmptyButNonce.json"] = "Not Implemented"
    skipTests["CreateCollisionToEmpty.json"] = "Not Implemented"
    skipTests["CreateHashCollision.json"] = "Not Implemented"
    skipTests["createJS_ExampleContract.json"] = "Not Implemented"

    // StaticCall implementations
    skipTests["staticcall_createfails.json"] = "STATICCALL Not Implemented"

    // Revert Implementations
    skipTests["FailedCreateRevertsDeletion.json"] = "REVERT Not Implemented"
    skipTests["CreateOOGafterInitCodeRevert.json"] = "REVERT Not Implemented"
    skipTests["CreateOOGafterInitCodeRevert2.json"] = "REVERT Not Implemented"

    // EIP 211 Implementations
    skipTests["CreateOOGafterInitCodeReturndataSize.json"] = "REVERT Not Implemented"
    skipTests["CreateOOGafterInitCodeReturndata2.json"] = "REVERT Not Implemented"

    // Random Test failures (REVISIT)
    skipTests["randomStatetest642.json"] = "random unimplemented"
    skipTests["randomStatetest644.json"] = "random unimplemented"
    skipTests["randomStatetest645.json"] = "random unimplemented"
    skipTests["Opcodes_TransactionInit.json/Byzantium/37"] = "random unimplemented"
    skipTests["Opcodes_TransactionInit.json/Byzantium/38"] = "random unimplemented"
    skipTests["Opcodes_TransactionInit.json/Byzantium/125"] = "random unimplemented"
    skipTests["Opcodes_TransactionInit.json/Byzantium/126"] = "random unimplemented"
    skipTests["CREATE_ContractRETURNBigOffset.json"] = "random unimplemented"`

[steviezhang]

@soc1c Is it fine that the circleCI tests are failing or does that indicate that I need to fix something, they seem to be failing on account of not being able to find the packages.

[soc1c]

failing tests are unrelated. someone should sit down and go through them

[GregTheGreek]

Please look into the failing errors

[austinabell]

Please look into the failing errors

I forgot to note, these errors have been around since we changed namespaces and switched to go modules.

TestHeader{1,2}ErrorLog is because the header was formed by trimming the gopath from the current filepath, but since we are using go modules, this test will fail. We can rewrite the header logic for the logger but it is out of the scope of this PR (unless we want to do with this one).

as for the bindings not working, I can look more into it but a comment in the Bind function defines the reason for it as:

// Bind generates a Go wrapper around a contract ABI. This wrapper isn't meant
// to be used as is in client code, but rather as an intermediate struct which
// enforces compile time type safety and naming convention opposed to having to
// manually maintain hard coded strings that break on runtime.

I can't find where the bindings are being generated (on my machine it is referencing ethereum/go-ethereum instead of our renamed namespace)