feat(cli): evm_bytes rename + print asm

[marioevz]

🗒️ Description

Changes included:

• Module Renaming and Refactoring:

  • Renamed the evm_bytes_to_python module to evm_bytes and refactored the code to improve readability and functionality. (pyproject.toml, src/cli/evm_bytes.py, src/cli/evm_bytes_to_python.py) [1] [2] [3]

• Test Updates:

  • Renamed and updated tests to reflect the module name change from evm_bytes_to_python to evm_bytes. (src/cli/tests/test_evm_bytes.py) [1] [2]

• --assembly flag:

  • Added --assembly flag which allows printing the bytecode in assembly format (e.g. see format used in EIP-7002 or EIP-7251).

🔗 Related Issues

None

✅ Checklist

• All: Set appropriate labels for the changes.
• All: Considered squashing commits to improve commit history.
• All: Added an entry to CHANGELOG.md.
• All: Considered updating the online docs in the ./docs/ directory.
• Tests: All converted JSON/YML tests from ethereum/tests have been added to converted-ethereum-tests.txt.
• Tests: A PR with removal of converted JSON/YML tests from ethereum/tests have been opened.
• Tests: Included the type and version of evm t8n tool used to locally execute test cases: e.g., ref with commit hash or geth 1.13.1-stable-3f40e65.
• Tests: Ran mkdocs serve locally and verified the auto-generated docs for new tests in the Test Case Reference are correctly formatted.

[marioevz]

@danceratopz this is the output of the assembly version:

uv run evm_bytes binary-file ./src/ethereum_test_forks/forks/contracts/withdrawal_request.bin --assembly
caller
push20 0xfffffffffffffffffffffffffffffffffffffffe
eq
push1 0x90
jumpi
...

Don't really know where we could fit this example though.

[danceratopz]

Don't really know where we could fit this example though.

I asked our old collaborator gpt to put the example in the docstrings with google docstring style - then we get some nice online docs. The parameter table isn't quite right (from a cli user perspective, it is from an api perspective), but perhaps we can incrementally improve this over time. I do like the examples though.

[danceratopz]

@marioevz I broke tox -e docs in multiple ways, will get back to this!

[marioevz]

@marioevz I broke tox -e docs in multiple ways, will get back to this!

No worries, super simple fixes, loved the way the docstrings were formatted, thanks!

[danceratopz]

Nice addition! LGTM!

[raxhvl]

I'm trying to use evm_bytes to generate python source code from bytecode for gentest (#862). The square brackets in the CLI output is not valid python code in this context.

If the square brackets in the output is used only for visual representation, can we please replace them with parenthesis? I'd be happy to PR this change.

execution-spec-tests/src/cli/evm_bytes.py

Line 42 in 295879d

return f"Op.{self.opcode._name_}[{operands}]"

cc: @danceratopz @marioevz

[danceratopz]

I'm trying to use evm_bytes to generate python source code from bytecode for gentest (#862). The square brackets in the CLI output is not valid python code in this context.

If the square brackets in the output is used only for visual representation, can we please replace them with parenthesis? I'd be happy to PR this change.

Maybe I don't understand the question? It should be valid 🙂 It was added to help testing EOF. For example:

execution-spec-tests/tests/osaka/eip7692_eof_v1/eip7480_data_section/test_code_validation.py

Lines 73 to 81 in ac615f6

	Container(
	name="DATALOADN_middle",
	sections=[
	Section.Code(
	code=Op.DATALOADN[16] + Op.POP + Op.STOP,
	),
	Section.Data(data="1122334455667788" * 16),
	],
	),

This is achieved by implementing a __getitem__() method in Opcode, which is used to express the "data portion" of opcodes:

execution-spec-tests/src/ethereum_test_vm/opcode.py

Lines 149 to 152 in ac615f6

	def __getitem__(self, *args: "int | bytes | str | Iterable[int]") -> "Opcode":
	"""
	Initialize a new instance of the opcode with the data portion set, and also clear
	the data portion variables to avoid reusing them.

Ah, does evm_bytes generate opcodes with data portions for legacy (non-EOF) code? Can you provide an example?

[raxhvl]

For legacy opcodes, both approach seems to be equivalent.

In [1]: (Op.PUSH1[0x81] + Op.PUSH1[0x81]).hex()
Out[1]: '60816081'

In [2]: (Op.PUSH1(0x81) + Op.PUSH1(0x81)).hex()
Out[2]: '60816081'

When writing this manually, is there a convention to be followed?