Revision history [back]

I guess the number of idempotents should be basically the same as the number of direct summands -- is that right? If so, then maybe you're looking for the direct-sum decomposition of Z/n. Does the fundamental theorem of finitely generated abelian groups give you enough information?