No changes yet :)
-
#547 requires your integration to understand the network
ByzantiumToConstantinopleAt5, which is analogous to existing pseudo networks such asFrontierToHomesteadAt5andEIP158ToByzantiumAt5. -
#557 changes the format of the rlp tests, byte strings are now prefixed with "0x"
- #547 added a test for the difficulty changes. This adds the network
ByzantiumToConstantinopleAt5, which is analogous to existing networks such asFrontierToHomesteadAt5andEIP158ToByzantiumAt5.
- #548 Added a test for EXTCODEHASH called on an account created in the same transaction
- #549 added a few more tests for EXTCODEHASH and EXTCODESIZE on accounts created in the same transaction
- #550 Checks that changes to an account's nonce, balance, or storage do not change the result of calling EXTCODEHASH
- #552 Added another EXTCODEHASH test, what if it's run on accounts which have self-destructed?
- #563 Another spectacularly detailed test case of EXTCODEHASH involving recursive calls and oog-induced revert
- #566 checks that EXTCODEHASH returns a different result when the code changes
- #568 adds a test for EXTCODEHASH; what if the contract being hashed self destructed while in the middle of a delegate call?
- #569 calls EXTCODEHASH on an empty contract
- #551 tests for a Constantinople bug the fuzzer found in geth
- #544 adds an RLP test, courtesy of the mana team
- #557 changes the format of the rlp tests, byte strings are now prefixed with "0x"
- #561 changed some of the tests. There were some changes to
testethwhich made the previous tests impossible to regenerate. They were refactored but not in a way which should break any client. - #567 adds a Constantinople test Mana found when they failed to sync with Ropsten
- Added initial test cases for
EXTCODEHASHEIP-1052, see PR #484 - More
EXTCODEHASHtests, see PR #544 - New
SSTOREstate tests and blockchain tests where an external call is overwriting/colliding with newSSTOREgas calculation rules, see PR #535
- New tests to cover cases where the result of an EVM opcode is written to a specified memory range and the result is shorter than the specified range, see PR #538
- Added
.ideato.gitignore, see PR #546
Test generation docs have been consolidated and integrated in the central ReadTheDocs testing documentation.
We also updated outdated parts on this doc section (see PR #539), so it should in principle now be possible to follow the guide and end up with a working test creation setup. There might still be some glitches, please let us know or submit a PR on ethereum/tests to if you stumble over something.
Other changes:
- New list with clients using the library in README, see PR #537
EIP-145(Bitwise shifting): Tests forSAR,SHLandSHRin GeneralStateTests/stShift directory, blockchain tests analogueEIP-1014(CREATE2): Various cases covered in GeneralStateTests/stCreate2 directory, blockchain tests analogueEIP-1052(EXTCODEHASH): Tests not merged yet, open PR #484EIP-1283(SSTORE): Dedicated tests in GeneralStateTests/stSStoreTest directory also covering Ropsten consensus issue cases, blockchain tests analogue, generally refilled state tests with newSSTOREgas metering rules in PR #511EIP-1234(difficulty): NewdifficultyConstantinople.jsonfile and regenerateddifficultyRopsten.jsonfiles in the BasicTests directory, see PR #518
Be aware that the format of BlockchainTests recently changed with the introduction of a new field sealEngine (values: NoProof | Ethash), see related JSON Schema change or BlockchainTest format docs for reference.
This means that you can faster-execute NoProof based tests skipping block validation. These tests nevertheless doesn't provide reliable values for PoW-based block header fields any more (mixHash, nonce), so make sure that you don't rely on the correctness of these values for the tests to pass.