Re: skip-chars-forward v. re-search-forward

[Kenichi Handa]

In article <E199FUs-0003kd-00@fencepost.gnu.org>, Richard Stallman 
<rms@gnu.org> writes:
>     It's still surprising that search-forward/backward is more
>     than twice faster.  I have no idea what kind of magic the
>     boyer-moore search is using.

> The comparison between search-forward and skip-chars-forward
> does not seem meaningful to me, because they do different jobs.
> The only case which both of these functions can do is
> (search-forward "a") and (skip-chars-forward "^a").

That is what I did in my benchmark test.
  (search-forward "%")
    vs (skip-chars-forward "^%")
    vs (re-search-forward "[%]")

> The time taken by search-forward is inversely proportional to the
> search string size in favorable cases, so if you tried a longer
> string, it was probably faster.

Yes, I know that.  I was surprised because search-forward
was twice faster even if the search string size is one.

---
Ken'ichi HANDA
handa@m17n.org