Re: [boost] [futures] boost::futures

From: Hartmut Kaiser (hartmut.kaiser_at_[hidden])
Date: 2007-03-14 16:45:38

Braddock Gaskill wrote:

> > Instead of going with generalized future composition, I've 
> tried the 
> > route of
> > 'wait_all(<future-list>) ' and 'wait_any(<future-tuple>)'
> 
> This is exactly what I personally would like to see.  I don't 
> relish the thought of digging out a return type or exception 
> from a future<variant<float, string, tuple<int, double, bool> 
> > >.  I would just want something that wakes me up when 
> dinner is ready and let me figure out which of my futures are 
> valid or have failed, if I even care.

You don't need to use the return type, if you don't want to use it
afterwards. And in the case you're interested in the actula return value, a
function like above will have to return the variant<float, string,
tuple<int, double, bool> > as well. So what do you gain from using a
function? Operator overloading gives you variable future counts with far
less coding effort. With functions you have to spell out a specialization
for every possible number of futures you want to combine.

> wait(f1 || (f2 && f3));
> might be nice syntax, does actually add functionality, and is 
> simple enough as well.

Yeah, sure. The library in the vault has a futurize() function doing exactly
what you want, just pass a arbitrary complex future expression to it. No
need to know the returned data type.

Regards Hartmut