I have a dataset tag and lookup expression on numeric entry component is taking data from this dataset. In dataset values are float like 1.8 but when I read this into a component it is rounded to 2.
Nope does not work, still rounds it to 2. I multiple the dataset values by 100 and then divide the lookup result by 100. It is an unneccessary way of doing it but only way I could figure it out.
The type of the value returned will always be coerced to be the same type as the noMatchValue.
With a bare 0, we have no way to know whether you're trying to specify an int, long, float, or double...so we default to integer, since it's most common. 0.0 will default to a floating point type (probably double, but I'm not sure off the top of my head). You could also do it explicitly, by wrapping your bare constant in a casting function, i.e. toDouble(0).