The most widely-used technique to generate wholebody motions on a humanoid robot accounting for various tasks and constraints is the inverse kinematics. Based on the taskfunction approach, this class of methods makes possible the coordination of the robot movements to execute several tasks in parallel and account for the sensor feedback, in real-time thanks to the low computation cost. To some extent, it also enables dealing with some of the robot constraints (e.g. joint limits or visibility) and managing the quasi-static balance of the robot. In order to fully use the whole range of possible motions, this paper proposes to extend the task-function approach to handle the full dynamics of the robot multi-body along with any constraint written as equality or inequality of the state and control variables. The definition of multiple objectives is made possible by ordering them inside a strict hierarchy. Several models of contact with the environment can be implemented in the framework. We propose a reduced formulation of the multiple rigid planar contact that keeps a low computation cost. The efficiency of this approach is illustrated by presenting several multi-contact dynamic motions in simulation and on the real HRP-2 robot.