Exact and signed numeric reports
Test precision, truncation, magnitude and optional known extrema.
Why does adding 0.1 and 0.2 miss an exact 0.3 report filter?
Use the supplied integer tick basis for an exact discrete total.
Why does forcing an integer counter to REAL lose one tick?
Keep exact integer accumulation within its declared range.
Why does a negative offset enter the wrong unit bucket?
Distinguish truncation toward zero from a lower interval boundary.
Why does a large negative deviation pass a tolerance report?
Apply the supplied limit to magnitude on both sides of zero.
Why does max(a,b) hide an available reading when the other is NULL?
Resolve optional inputs before taking the highest recorded value.