diff --git a/README.md b/README.md index 0e5bd34..db83f61 100644 --- a/README.md +++ b/README.md @@ -12,7 +12,7 @@ _**2021-01-07**_ A bug regarding zero-length slices was disclosed by @MrChico following an audit to the Optimism codebase. -The exact bug happened under the following conditions: if memory slots higher then the current free-memory pointer were tainted before calling the `slice` method with a desired length of `0`, the returned bytes array, instead of being a zero-length slice was an array of arbitrary length based on the values that previously populated that memory region. +The exact bug happened under the following conditions: if memory slots higher than the current free-memory pointer were tainted before calling the `slice` method with a desired length of `0`, the returned bytes array, instead of being a zero-length slice was an array of arbitrary length based on the values that previously populated that memory region. Overall, the usage of zero-length slices should be pretty unusual and, as such, hopefully, this bug does not have far-reaching implications. Nonetheless, *please update the library to the new version if you're using it in production*. diff --git a/test/TestBytesLib2.sol b/test/TestBytesLib2.sol index cb54ca3..3bea5cf 100755 --- a/test/TestBytesLib2.sol +++ b/test/TestBytesLib2.sol @@ -100,7 +100,7 @@ AssertBytes.equal(checkBytesZeroLength, checkBytesZeroLengthRight, "Sanity check assertEq(resultBytes, testBytes); AssertBytes.equal(resultBytes, testBytes, "Multiple (*2) of 32 bytes slice failed."); - // With v0.5.x we can now entirely replace the ThrowProxy patterns that was creating issues with the js-vm + // With v0.5.x we can now entirely replace the ThrowProxy patterns that were creating issues with the js-vm // and use an external call to our own contract with the function selector, since Solidity now gives us // access to those bool r;