A robotic arm, like a human arm, consists of an upper arm and forearm. The upper arm is attached to the shoulder joint and can rotate at the joint in any direction. The forearm is attached to the upper arm at the elbow joint. The forearm can turn with respect to the upper arm about the axis that passes through the elbow joint and is perpendicular to both upper arm and forearm. The angle at which the arm is bent at the elbow can't be acute. At the end of the forearm there is a claw consisting of two equal parts. The claw can open and close. When the claw is completely closed, it can hold a cylinder of a unit diameter. In this position the axis of the cylinder is parallel to the axis of the elbow joint.

There is a cylindrical rail of a unit diameter and infinite length near the robot. The robot wants to grip the rail. The claw is now open and in the end it must be completely closed around the rail. Determine the final position of the robot's arm.

In the first line you are given the length of the upper arm a and the length of the forearm b. In the second and the third lines there are the coordinates of two different points on the axis of the rail. The origin of the coordinate system is the shoulder joint. The rail does not pass through the shoulder. All the numbers are integer and their absolute values do not exceed 100; a, b ≥ 4. The thickness of the upper arm and forearm is negligible.

If there is no position of the robotic arm in which the claw can be closed on the rail, output “No solution.” Otherwise, output the coordinates of the elbow joint and the angle at which the arm is bent in radians with the maximal possible accuracy. The value of the angle must be in the range [π/2, π]. If there are several variants of the solution, output any one of them.