>> Thus I suggest now to split the source code search pattern into >> four separate rules. > > Why? Does the Coccinelle software ensure that a variable like “r.ul” contains really useful data even if the expected branch of the SmPL disjunction was occasionally not matched? Regards, Markus
