test: pin that the factory forwards uncapped gas to initialize

[thedavidmeister]

Closes #120

ICloneableFactoryV4 states normatively that "the factory forwards whatever gas
the caller left it, and the deployer chooses that". The code that makes it true
is an absence: src/lib/LibICloneableFactoryV4.sol:184 calls the clone with no
{gas: …} option, so EIP-150's 63/64 rule governs and the callee's budget
tracks the caller's. Nothing in the suite read gas at all — grep -rn "gasleft\|gas:" test/
returned nothing — and mutation cannot reach an absent operand, so a
"defensive" cap would have left all 78 tests green while making every gas-heavy
implementation undeployable.

• test/concrete/TestCloneableGasRecorder.sol — new: records gasleft() on
  entry to initialize. No existing fixture exposes anything about the budget
  the call carried.
• testInitializeGasRisesWithCallerBudget — deploys through BOTH entry points
  at gasleft()/8 and then gasleft()/2 and asserts the observed figure rises.
  The budgets are fractions of what the test itself has left, so they are
  derived inside the same run rather than pinned, and both sit far above any cap
  a chain's own block gas limit would let a caller supply.

QA

• Discriminating tests: testInitializeGasRisesWithCallerBudget
  (test/src/lib/LibICloneableFactoryV4.cloneAndInitialize.t.sol).
• Mutations applied: 1 — child.call{gas: 30000000}(abi.encodeCall(...)) at
  src/lib/LibICloneableFactoryV4.sol:184, a plausible defensive cap. On main
  it SURVIVES: all 8 suites green, 78 passed / 0 failed. On this branch the
  cloneAndInitialize file goes 9 passed / 1 failed, and the one failure is
  testInitializeGasRisesWithCallerBudget on the open-salt leg. Unmutated, the
  whole suite is 79 passed / 0 failed.
• Oracle: src/interface/ICloneableFactoryV4.sol's own sentence about
  forwarding the caller's gas, and EIP-150 — 63/64 of a larger budget is larger,
  which is the only reason the strict inequality can hold without a pinned
  figure.
• Category check: the category is "a normative claim about the CALL OPTIONS
  the factory uses, invisible to mutation because the source carries no
  operand". The other options on that same call are value and callee/calldata.
  Value is pinned on both paths by …NoEthForwarded (namespaced today,
  open-salt in [F19] [LOW] Four shared-tail clauses are pinned only through cloneDeterministic; nothing asserts cloneDeterministicOpenSalt reaches them #123); callee and calldata are pinned by the recorder's selector
  and sData assertions. Gas was the only one with nothing behind it.

Touches test/src/lib/LibICloneableFactoryV4.cloneAndInitialize.t.sol, as does
the PR for #121 — both branch off main and append to the same file.

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

Next included review available in 29 minutes.

Check out review usage here.

View limit details

Limit details: You’ve used the included review currently available.

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

Learn how review limits work.

Review configuration:

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Advanced

Run ID: b3065e78-0031-4cea-9faa-47856375d04d

📥 Commits

Reviewing files that changed from the base of the PR and between f2d9e5a and 4114a87.

📒 Files selected for processing (2)

• test/concrete/TestCloneableGasRecorder.sol
• test/src/lib/LibICloneableFactoryV4.cloneAndInitialize.t.sol

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}