LibJS: Add a modulo() function to represent the "x modulo y" notation

This *not* being the same as C++ "%" in all cases is a massive footgun,
and hopefully one I came across for the last time.

1 files changed, 14 insertions, 0 deletions

diff --git a/Userland/Libraries/LibJS/Runtime/AbstractOperations.h b/Userland/Libraries/LibJS/Runtime/AbstractOperations.h
index a935932084..1f68721298 100644
--- a/Userland/Libraries/LibJS/Runtime/AbstractOperations.h
+++ b/Userland/Libraries/LibJS/Runtime/AbstractOperations.h
@@ -76,4 +76,18 @@ ThrowCompletionOr<T*> ordinary_create_from_constructor(GlobalObject& global_obje
     return global_object.heap().allocate<T>(global_object, forward<Args>(args)..., *prototype);
 }
 
+// x modulo y, https://tc39.es/ecma262/#eqn-modulo
+template<typename T>
+T modulo(T x, T y)
+{
+    // The notation “x modulo y” (y must be finite and non-zero) computes a value k of the same sign as y (or zero) such that abs(k) < abs(y) and x - k = q × y for some integer q.
+    VERIFY(y != 0);
+    if constexpr (IsFloatingPoint<T>) {
+        VERIFY(isfinite(y));
+        return fmod(fmod(x, y) + y, y);
+    } else {
+        return ((x % y) + y) % y;
+    }
+}
+
 }