What does context precision measure in a RAG retriever, and how does it differ from context recall?

What does context precision measure in a RAG retriever, and how does it differ from context recall?