This function rounds numerical value a to the nearest double value with b digits after the decimal separator. Since this function uses double values internally, rounding errors may occur. Use function "round(bigdecimal a, b)" to avoid this.

If a is not a numerical value, the function returns 0.0, see examples below.