Sorry I forget to CC the list. Chris ---------- Forwarded message ---------- From: Christopher Li <sparse@xxxxxxxxxxx> Date: Sat, May 5, 2012 at 12:54 PM Subject: Re: dependency tee from c parser entities downto token To: Konrad Eisele <eiselekd@xxxxxxxxx> On Sat, May 5, 2012 at 9:59 AM, Konrad Eisele <eiselekd@xxxxxxxxx> wrote: >> >> I am not sure I understand your range representation yet. >> > > You need to view it with a fixed width font. Its not rocket science, > token lists (or arrays) are viewed as dotted lists. The token.pos > field is listed below each token as p[x] or as the file-location > in file-scope. I am not sure using range is a win because you need to find the token within a list of ranges. That is a lot of hassle. Just walk the replacement token in the macro expand hook and replace them all with new stream and line number seems easier. The look up is much easier, just an index into the array. > > I'll come up with a patch to implement this scheme when I have > time to and send it, it might take a while. Can you keep the change to the core sparse library (e.g. adding hook in parser and pre-processor) as a separate patch? You can send that part out for review earlier if other part of your dependency analyse is not ready yet. Chris -- To unsubscribe from this list: send the line "unsubscribe linux-sparse" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
