The issue is that comprehensions pre-allocate their storage and therefore
only work when the thing being iterated over has a length method that can
tell it how big an array to allocate. You can use the collect function
instead:
julia> collect(@task fib(4000000))
34-element Array{Any,1}:
0
1
1
2
3
5
8
13
21
34
55
89
144
233
377
610
987
1597
2584
4181
6765
10946
17711
28657
46368
75025
121393
196418
317811
514229
832040
1346269
2178309
3524578
julia> filter(iseven, ans)
12-element Array{Any,1}:
0
2
8
34
144
610
2584
10946
46368
196418
832040
3524578
julia> sum(ans)
4613732
But you can also just use the sum function, which will operate on anything
iterable:
julia> sum(filter(iseven, @task fib(4000000)))
4613732
A while ago I added a bunch of fairly efficient solutions to Project Euler
problems to the test suite, including #2, but don't click if you don't want
spoilers:
https://github.com/JuliaLang/julia/blob/7dd805feb0/test/euler.jl#L9-L20
On Sat, Dec 6, 2014 at 12:04 PM, Bill Allen <[email protected]> wrote:
> To learn julia, I'm working my way through Project Euler
> <https://projecteuler.net/>. For question #2,
> <https://projecteuler.net/problem=2> I ran into an issue that I haven't
> found a solution to.
>
> Given a function that computes fibonacci numbers:
>
> function fib(n)
> a,b = 0,1
> while(a < n)
> produce(a)
> a,b=b,a+b
> end
> end
>
> I can write a loop that computes the value:
>
> function fibsum(n)
> s = 0
> for x in Task(() -> fib(n))
> if x % 2 == 0
> s = s + x
> end
> end
> s
> end
>
> fibsum(4000000)
> 4613732
>
> Analogously, one can write:
>
> s = 0
> for x in [1,2,3]
> s = s + x
> end
> println(s)
> 6
>
> or more succinctly,
>
> sum([x for x in [1,2,3]])
>
> But, when I try that with the fib Task:
>
> sum([x for x in Task(()->fib(10))])
>
> I get
>
> ERROR: `length` has no method matching length(::Task)
> in anonymous at no file
>
> Which confuses me sense the task works nicely in an explicit loop.
>
> Regards,
> Bill
>
>
