How do you use `BeApproximately` with a value that changes in each iteration of a loop?

Description

• I’m iterating over 2 lists
• I want to compare the 2 values at each point in the list
• I want the value to be within 0.1% of each other

Complete minimal example reproducing the issue

var sheetRows = new List<float> { -24.0, 0.1, 10};
var loads= new List<float> { -24.0, 0.1, 10.05};
foreach ((var xl, var db) in sheetRows.Zip(loads, (n, w) => (n, w)))
{
   xl.Should().BeApproximately(db, db*0.001f);
}

Expected behavior:

To pass

Actual behavior:

Expected xl to approximate -24F +/- -0.024F, but -24F differed by 0F.
   at FluentAssertions.Execution.LateBoundTestFramework.Throw(String message)
   at FluentAssertions.Execution.TestFrameworkProvider.Throw(String message)
   at FluentAssertions.Execution.DefaultAssertionStrategy.HandleFailure(String message)
   at FluentAssertions.Execution.AssertionScope.FailWith(Func`1 failReasonFunc)
   at FluentAssertions.Execution.AssertionScope.FailWith(Func`1 failReasonFunc)
   at FluentAssertions.Execution.AssertionScope.FailWith(String message, Object[] args)
   at FluentAssertions.NumericAssertionsExtensions.FailIfDifferenceOutsidePrecision[T](Boolean differenceWithinPrecision, NumericAssertions`1 parent, T expectedValue, T precision, T actualDifference, String because, Object[] becauseArgs)
   at FluentAssertions.NumericAssertionsExtensions.BeApproximately(NumericAssertions`1 parent, Single expectedValue, Single precision, String because, Object[] becauseArgs)
   at RoutingTests.UnitResponseModuleTests.U010_CheckTranslationAsync(Int64 sId, String sheetName, Int32 expected, Int32 excelRowCount) in C:\Users\sogscx\source\repos\Float\Compute\Test\RoutingTests\UnitResponseModuleTests.cs:line 480
   at NUnit.Framework.Internal.TaskAwaitAdapter.GenericAdapter`1.BlockUntilCompleted()

Versions

• Which version of Fluent Assertions are you using? latest
• Which .NET runtime and version are you targeting? E.g. .NET framework 4.6.1 or .NET Core 2.1. .NET Core 3.1 running NUnit

[jnyrup]

@Smurf-IV The fix will be available in the upcoming 6.0 release.

I’ve added guard against negative values, but not zeroes. Approximating a value with a zero precision would of course be more clearly written as an exact comparison, but it works as expected.

[Smurf-IV]

I cannot re-open, but as per SO. There needs to be argument validation for the precision so that -ves (Or zero) are not used for the tolerance value.