unnecessary_nan_comparison

A double can't equal 'double.nan', so the condition is always 'false'.

A double can't equal 'double.nan', so the condition is always 'true'.

The analyzer produces this diagnostic when a value is compared to double.nan using either == or !=.

Dart follows the IEEE 754 floating-point standard for the semantics of floating point operations, which states that, for any floating point value x (including NaN, positive infinity, and negative infinity),

• NaN == x is always false
• NaN != x is always true

As a result, comparing any value to NaN is pointless because the result is already known (based on the comparison operator being used).

The following code produces this diagnostic because d is being compared to double.nan:

bool isNaN(double d) => d == double.nan;

Use the getter double.isNaN instead:

bool isNaN(double d) => d.isNaN;