Marking as resolved since the translation has been accepted :)

o_O

//    Manual  Notification    \\
\\ Please read the post above //

o_O

std::next is declared there.

hahahahaaaa Yeah Bro me too ... Hope you Enjoyed the kata ..

My bad, [[1,2],[0,2],[1,2]] should actually be [[1,2],[0,2],[0,1]]
That [[1,2],[0,2],[1,2]] is indeed connected, and if you ignore the node connected to itself, is also a tree.
Also, you need to test not just for connectedness, but also if the graph is a tree or not.
[[1,2],[0,2],[0,1]] is not a tree, because it contains the path 0->1->2->0

I hope that clarifies things.

Sir It's My honor translating my humble kata ... All my katas are yours ... Of course approved... regards Zizou

If B4B's explanation isn't clear enough, I can show more details:

1 2 x
? ? 2
a d 2
b 1 2 x
c e f 2
? ? ? 1

logic: 
There is 1 mine around 1(a,b,c,d,e,f must has 1 mine)
There is 1 mine around 2(d,e,f must has 1 mine)
So, there is no mine in a,b,c, you can open them.

This is a basic logic, very commonly used.
I think your code must implement this logic to look less clumsy :D

1 2 x
? ? 2
a ? 2
b 1 2 x
c ? ? 2
? ? ? 1

Your algo isn't smart enough: try to solve the tests on your own and you'll see that you can open a, b and c there. There are a lot of things to do by deduction in this kata, before going on the dark side of the force (combinations and so on).

Done! Thanks again for verifying the code. It's so difficult to notice these things when my solution code is part of the test code.

I believe it's fixed now.