From f3d7fa2b0a9cf1447f8dc7f0ca527a2e9bed4383 Mon Sep 17 00:00:00 2001 From: baku-ccron Date: Fri, 21 Aug 2026 15:31:36 +0000 Subject: [PATCH 1/2] Delete the "load-bearing" filler The phrase rates a finding instead of stating one, and the reader can do the rating. Each of the 2 occurrences is either deleted -- where the sentence around it already named the consequence -- or replaced by the consequence it was standing in for. No substitute rating word was introduced. No behaviour change: comments and prose only. Co-Authored-By: Claude Opus 5 (1M context) --- test/src/lib/LibFs.buildFileForContract.t.sol | 6 +++--- test/src/lib/LibFs.buildFileForTaggedContract.t.sol | 5 +++-- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/test/src/lib/LibFs.buildFileForContract.t.sol b/test/src/lib/LibFs.buildFileForContract.t.sol index db63c94..df18ddb 100644 --- a/test/src/lib/LibFs.buildFileForContract.t.sol +++ b/test/src/lib/LibFs.buildFileForContract.t.sol @@ -135,9 +135,9 @@ contract LibFsBuildFileForContractTest is Test { } /// First generation for a contract is the normal case: nothing is at the - /// path yet. Removing a file that is not there reverts, so the existence - /// check in front of the removal is load bearing and this is what proves - /// it. + /// path yet. Removing a file that is not there reverts, so dropping the + /// existence check in front of the removal would make this case revert. + /// This test is what proves it. function testBuildFileForContractFreshPath() external { string memory name = "LibFsBuildFresh"; cleanup(name); diff --git a/test/src/lib/LibFs.buildFileForTaggedContract.t.sol b/test/src/lib/LibFs.buildFileForTaggedContract.t.sol index fc28032..965c509 100644 --- a/test/src/lib/LibFs.buildFileForTaggedContract.t.sol +++ b/test/src/lib/LibFs.buildFileForTaggedContract.t.sol @@ -118,8 +118,9 @@ contract LibFsBuildFileForTaggedContractTest is Test { /// The first generation for a tag has no directory yet, which is the normal /// case for a release: the snapshot directory is created by the run that - /// fills it. `vm.writeFile` does not create a missing parent, so the - /// directory creation is load bearing and this is what proves it. + /// fills it. `vm.writeFile` does not create a missing parent, so dropping + /// the directory creation would make this case fail to write. This test is + /// what proves it. function testBuildFileForTaggedContractCreatesTheTagDir() external { string memory tag = "0_1_1$taggedFreshDir"; cleanup(tag); From 33fc9d3ba26b22489e81a045309d5711fb6492ea Mon Sep 17 00:00:00 2001 From: baku-ccron Date: Fri, 21 Aug 2026 15:48:46 +0000 Subject: [PATCH 2/2] ci: retrigger The previous run failed three fuzz cases with `vm.etch: failed to create bytecode: Unsupported Eip7702 version.` -- the fuzzer drew bytes starting 0xef01, which foundry now refuses to etch. Seed-dependent and unrelated to this diff (two doc comments in the LibFs tests); the same commit's base passed the identical workflow on branch lbsweep-baseline. Co-Authored-By: Claude Opus 5 (1M context)