EthAccount tests for execute_from_outside_v2

[remollemo]

Note

Low Risk
Primarily adds/updates tests, fixtures, and build/test tooling; no production contract logic changes, with the main risk being test brittleness due to hardcoded deterministic addresses/signatures.

Overview
Adds an end-to-end test suite for execute_from_outside_v2, covering success paths and expected reverts for time bounds, nonce replay, invalid signatures, domain-separator mismatches (EVM chain id, Starknet chain name, verifying contract), and caller restrictions.

Extends test utilities with EFO setup helpers (timestamp/chain id cheats), OutsideExecution builders, precomputed EIP-712 signatures, and an external DualCaseERC20Mock deployment to validate real approve/transfer multi-call behavior (including failure propagation via the safe dispatcher). Also introduces a Python signature generator (scripts/generate_test_signatures.py + requirements.txt) and updates Scarb config/lockfile to include starkware_utils and build the external mock contract for tests.

^{Written by Cursor Bugbot for commit 4484090. This will update automatically on new commits. Configure here.}

---

This change is 

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

[remollemo]

@remollemo resolved 1 discussion.
Reviewable status: 0 of 2 files reviewed, all discussions resolved.

[remollemo]

reopen!

[remollemo]

@remollemo reviewed all commit messages.
Reviewable status: 0 of 2 files reviewed, all discussions resolved.

[ItayRosenberg]

@ItayRosenberg made 2 comments.
Reviewable status: 0 of 2 files reviewed, 2 unresolved discussions (waiting on remollemo).

---

eth_712_account/src/test_utils.cairo line 74 at r3 (raw file):

/// ANY_CALLER constant from SNIP-9.
pub const ANY_CALLER: felt252 = 'ANY_CALLER';

r u testing the case where the caller is not any caller?

Code quote:

pub const ANY_CALLER: felt252 = 'ANY_CALLER';

---

eth_712_account/src/test.cairo line 157 at r3 (raw file):

    // Execute - this should succeed
    let _results = src9.execute_from_outside_v2(outside_execution, signature.span());

can u delete this?

Code quote:

 let _results =

[remollemo]

@remollemo made 1 comment and resolved 1 discussion.
Reviewable status: 0 of 6 files reviewed, 1 unresolved discussion (waiting on ItayRosenberg).

---

eth_712_account/src/test_utils.cairo line 74 at r3 (raw file):

Previously, ItayRosenberg wrote…

r u testing the case where the caller is not any caller?

I tested this manually, will add this here too.

[remollemo]

@remollemo resolved 1 discussion.
Reviewable status: 0 of 6 files reviewed, all discussions resolved (waiting on ItayRosenberg).

[ItayRosenberg]

@ItayRosenberg reviewed 5 files and made 9 comments.
Reviewable status: 5 of 6 files reviewed, 9 unresolved discussions (waiting on remollemo).

---

a discussion (no related file):
this what m agent wrote me

Python script nonce mismatch with test constants - The Python generator functions use hardcoded nonces (nonce=2 in generate_multi_call_test line 308, nonce=3 in generate_atomicity_test line 351) which don't match the Cairo constants (NONCE_MULTI_CALL=101, NONCE_ATOMICITY=102). The main() function overrides these after generation, so it works, but the intermediate generation is wasted. Consider using the nonce as a parameter from the start to avoid confusion.

---

a discussion (no related file):
re the duplication between the singatures in teh script and in the cairo
Two ways to improve:

1. Script writes directly to a generated file — instead of print(), have the script output a .cairo file (e.g., test_signatures.generated.cairo) that test_utils imports. No copy-paste needed.

2. Single source of truth — the Python script already has all the constants (EXPECTED_CONTRACT_ADDRESS, nonces, amounts). But test_utils.cairo also defines them (NONCE_SINGLE_CALL, MOCK_ERC20_INITIAL_SUPPLY, etc.). If they drift apart, signatures silently break. The script should ideally read from a shared config or at least assert the constants match.

---

a discussion (no related file):
here (it is not 100% related)

earn-contracts/eth_712_account/src/eth_712_utils.cairo

Lines 18 to 19 in 59c6462

	const OUTSIDE_EXECUTION_TYPE_HASH: u256 =
	0x57fbef2abe14202f3651b3935a8feddd357b8f83a862e046239d196ec76f281e_u256;

You ddint explain what is this (for the rest you did explain)

---

eth_712_account/src/test.cairo line 230 at r5 (raw file):

    let outside_execution = get_test_outside_execution();

    // Signature generated with EVM chain ID 2 - valid because contract uses chain_id from signature

im not sure that i understand this comment

Code quote:

 // Signature generated with EVM chain ID 2 - valid because contract uses chain_id from signature

---

eth_712_account/src/test.cairo line 280 at r5 (raw file):

// ================================
// execute_from_outside_v2 tests with actual calls

test_execute_from_outside_specific_caller_succeeds
and test_execute_from_outside_wrong_caller_reverts should be before this

Code quote:

// execute_from_outside_v2 tests with actual calls

---

eth_712_account/src/test.cairo line 370 at r5 (raw file):

    // Verify that the execution failed (error from second call propagates)
    assert!(result.is_err(), "Expected call to fail due to insufficient balance");
    // In production, Starknet's protocol guarantees all state changes are rolled back.

cant we veirfy it?
I thought that the problem is only with events but maybe im wrong

Code quote:

  // In production, Starknet's protocol guarantees all state changes are rolled back.

---

eth_712_account/src/test.cairo line 371 at r5 (raw file):

    assert!(result.is_err(), "Expected call to fail due to insufficient balance");
    // In production, Starknet's protocol guarantees all state changes are rolled back.
// snforge doesn't simulate this, so we cannot verify rollback here.

indent?

---

eth_712_account/src/test_utils.cairo line 93 at r5 (raw file):

pub fn setup_efo_test_with_erc20() -> (ISRC9_V2Dispatcher, ContractAddress, IERC20Dispatcher) {
    let src9 = setup_efo_test();
    let (token_address, token) = deploy_mock_erc20(src9.contract_address);

why not using
deploy_mock_erc20_contract_at

---

eth_712_account/src/test_utils.cairo line 164 at r5 (raw file):

        0x69b65ab69e218022a936846ea383df37, // s_low
        27, // v
        2 // hain_id: Not Ethereum

Suggestion:

chain_id: Not Ethereum

[remollemo]

@remollemo made 6 comments and resolved 2 discussions.
Reviewable status: 5 of 6 files reviewed, 7 unresolved discussions (waiting on ItayRosenberg).

---

a discussion (no related file):

Previously, ItayRosenberg wrote…

here (it is not 100% related)

earn-contracts/eth_712_account/src/eth_712_utils.cairo

Lines 18 to 19 in 59c6462

	const OUTSIDE_EXECUTION_TYPE_HASH: u256 =
	0x57fbef2abe14202f3651b3935a8feddd357b8f83a862e046239d196ec76f281e_u256;

You ddint explain what is this (for the rest you did explain)

what's to explain? that's the hash

---

eth_712_account/src/test.cairo line 230 at r5 (raw file):

Previously, ItayRosenberg wrote…

im not sure that i understand this comment

it's for the mismatching chain id test

---

eth_712_account/src/test.cairo line 280 at r5 (raw file):

Previously, ItayRosenberg wrote…

test_execute_from_outside_specific_caller_succeeds
and test_execute_from_outside_wrong_caller_reverts should be before this

i don't understand

---

eth_712_account/src/test.cairo line 370 at r5 (raw file):

Previously, ItayRosenberg wrote…

cant we veirfy it?
I thought that the problem is only with events but maybe im wrong

it doesn't work

---

eth_712_account/src/test_utils.cairo line 93 at r5 (raw file):

Previously, ItayRosenberg wrote…

why not using
deploy_mock_erc20_contract_at

what's wrong with this one?

---

eth_712_account/src/test_utils.cairo line 164 at r5 (raw file):

        0x69b65ab69e218022a936846ea383df37, // s_low
        27, // v
        2 // hain_id: Not Ethereum

:-)

[remollemo]

@remollemo made 2 comments and resolved 1 discussion.
Reviewable status: 5 of 6 files reviewed, 6 unresolved discussions (waiting on ItayRosenberg).

---

a discussion (no related file):

Previously, ItayRosenberg wrote…

this what m agent wrote me

Python script nonce mismatch with test constants - The Python generator functions use hardcoded nonces (nonce=2 in generate_multi_call_test line 308, nonce=3 in generate_atomicity_test line 351) which don't match the Cairo constants (NONCE_MULTI_CALL=101, NONCE_ATOMICITY=102). The main() function overrides these after generation, so it works, but the intermediate generation is wasted. Consider using the nonce as a parameter from the start to avoid confusion.

I don't care what your agent told. it's you agent.

---

a discussion (no related file):

Previously, ItayRosenberg wrote…

re the duplication between the singatures in teh script and in the cairo
Two ways to improve:

1. Script writes directly to a generated file — instead of print(), have the script output a .cairo file (e.g., test_signatures.generated.cairo) that test_utils imports. No copy-paste needed.

2. Single source of truth — the Python script already has all the constants (EXPECTED_CONTRACT_ADDRESS, nonces, amounts). But test_utils.cairo also defines them (NONCE_SINGLE_CALL, MOCK_ERC20_INITIAL_SUPPLY, etc.). If they drift apart, signatures silently break. The script should ideally read from a shared config or at least assert the constants match.

I will fix the python scripts etc. in next PR (the one with the normal tx support)

[remollemo]

@remollemo made 1 comment and resolved 5 discussions.
Reviewable status: 4 of 6 files reviewed, 1 unresolved discussion (waiting on ItayRosenberg).

---

eth_712_account/src/test.cairo line 280 at r5 (raw file):

Previously, remollemo wrote…

i don't understand

I moved it a bit.

[remollemo]

@remollemo resolved 1 discussion.
Reviewable status: 4 of 6 files reviewed, all discussions resolved (waiting on ItayRosenberg).

[remollemo]

@remollemo partially reviewed 1 file.
Reviewable status: 3 of 6 files reviewed, all discussions resolved (waiting on ItayRosenberg).

[ItayRosenberg]

@ItayRosenberg reviewed 3 files and all commit messages, and made 3 comments.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on remollemo).

---

a discussion (no related file):

Previously, remollemo wrote…

what's to explain? that's the hash

all the rest are with explanations

---

eth_712_account/src/test_utils.cairo line 93 at r5 (raw file):

Previously, remollemo wrote…

what's wrong with this one?

because it is already ready in the test_utils
so less code dup

---

a discussion (no related file):
missing test - INVALID_SIGNATURE_LENGTH

[remollemo]

@remollemo made 3 comments.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on ItayRosenberg).

---

a discussion (no related file):

Previously, ItayRosenberg wrote…

all the rest are with explanations

in #23

---

a discussion (no related file):

Previously, ItayRosenberg wrote…

missing test - INVALID_SIGNATURE_LENGTH

in #23

---

eth_712_account/src/test_utils.cairo line 93 at r5 (raw file):

Previously, ItayRosenberg wrote…

because it is already ready in the test_utils
so less code dup

changed (and using deploy_at)