Merging operators: Beyond the finite case

We extend the results of Konieczny and Pino Pérez (Journal of Logic and Computation, 2002) concerning merging operators in a finite logical framework to the infinite case (countably many propositional variables). The number of sources we consider remains finite. The main result is the representation theorem. In order to prove it, we state some results which are interesting for their own sake. Some postulates had to be restated in a new form. The new form is equivalent to the old one only in the finite case, but more appropriate to deal with the infinite case. The construction of merging operators starting from distances between valuations is also generalized. Indeed, we introduce a new kind of operators built upon the so-called Cantor distance.