I've just started using wave, but it occurs to me that there are some basic things I don't understand about it, and which aren't (I think) covered in the online docs. Does anyone know: 1 - presumably wave uses UTF-8? Will it accept arbitrary UTF-8 files? What if the source starts with a 3-byte UTF-8 BOM? 2 - how closely does the source have to conform to C/C++ code? I can understand that the source has to follow basic lexing conventions (use of apostrophes, and so on), but wave actually seems to carry out full C++ lexing. Why is this necessary - could it just process whitespace-delimited tokens (outside comments)? The reason I'm asking is that I'm pre-processing arbitrary non-C code. This seems to be Ok if I disable whitespace insertion for disambiguation, but I'd just like to be sure that I'm not missing something. Thanks - Paul