Do pure functions solve the "return const" problems?

[guslay]

Steven Schveighoffer Wrote:

> "Janice Caron" wrote
> >
> > Pure functions also happen to be thread-safe, but not all thread-safe
> > functions are pure.
> 
> That was my impression too, but read Bill's post closely.  What he is saying 
> is that pure functions do not need invariant parameters, which means they 
> are subject to thread issues.
> 
> If pure functions are not required to take invariant parameters, then they 
> cannot be guaranteed to be thread safe.
> 
> -Steve 
> 

Just to complement:

If f is pure, and a is invariant. f(a)+f(a) is equivalent to 2 * f(a).

If a is not invariant, a can be modified by f(). However, that is not a side effect: a is considered as an output of f(). f can still called be pure.

There is a dependency between the two f(a) that prevents reordering or caching or threading

first call : f(a) produces a1.
second call : f(a1) produces a2.

The thing with pure function is that just by looking at the nature of the parameters, you know how coupled they are.