pine-script function 返回的值与预期不同

Question

I am struggling with the understanding of function executions in pine-script. Please see the short script below, it should compile as it is:

//@version=4
study("My Script")
a = 100.
d = 100.

d:=nz(d[1]) + 10

f_(_src) => nz(_src[1]) + 10
a := f_(a)

plot(a, "a", color.red, 5)
plot(d, "d", color.yellow, 5)

I would expect, that the value of a and d are the same, as the function "f_" executes the same code as in the assignment for d . But a has always a constant value of 110 from the 2nd cycle, when d increases each cycle linearly +10.

What is the reason it behaves like this? I would be very happy about some reference.

Thank you a lot for your help, Seb.

Answer 1

Your variables are initialized to 100. on every bar.

For variable a :

• The value passed to f_() is always the a variable which has just been initialized to 100. , as it's initialized on every bar, but once inside the function, that current value isn't used. Because you use nz(_src[1]) , the value used is the value of the parameter the last time the function was called, all this from the function's perspective—not the global scope's perpective.
• On the function's first call, the nz() call replaces the value with 0 as there is no previous value. The function returns 10 .
• On the second bar and all subsequent ones, you are still calling the function with the a=100 argument, but from thereon a previous value exists for the argument inside the function. It is always 100. because that's what you are calling the function with every time, so nz(_src[1]) always returns 100. , and the function always returns 110.

For variable d :

• On the first bar there is no previous value so nz(d[1]) returns 0 and the value 0 + 10 = 10 is assigned to d .
• On the second bar, nz(d[1]) returns 10 and the value 10 + 10 = 20 is assigned to d , and so on.
• In the case of d , the recently initialized value of 100. on each bar is, in fact, never used.