On Sat, Dec 22, 2018 at 07:22:07AM -0500, Steven Rostedt wrote: > On Sat, 22 Dec 2018 18:28:18 +0900 > Namhyung Kim <[email protected]> wrote: > > > > > > > for (i = 0; i < hist_data->attrs->n_actions; i++) { > > > str = hist_data->attrs->action_str[i]; > > > > > > - if (str_has_prefix(str, "onmatch(")) { > > > - char *action_str = str + sizeof("onmatch(") - 1; > > > + if ((len = str_has_prefix(str, "onmatch("))) { > > > + char *action_str = str + len; > > > > IMHO, returning (match) length might confuse people that it might > > support partial match and returns the length actually matched rather > > than full match. If I knew it always returns the length of prefix (or > > 0) why it matters? Using strlen() in the next line will have same > > effect of compiler optimization for the constant strings, no? > > > > if (str_has_prefix(str, "onmatch(")) { > > char *action_str = str + strlen("onmatch("); > > The reason to return the length was to get rid of the need for > duplicating the strlen("xxxx") because it's a burden to keep the two > the same. Not only that, a lot of places just do "str + 7" because it's > easier to type. > > Yes, it has the same effect on the compiler, but that's not what we are > trying to solve. We are trying to get rid of the duplication, because > that requires humans to get it right, and humans are not good at that.
Then I'm ok with that. :) Thanks, Namhyung
