Geoff Hutchison writes:
>
> > I was not courageous enough to patch all the places in the
> >code that show
> > s = 0;
> > s.append(s, l);
>
> We may want a better construct for this as well. There are a lot of
> instances of clearing a string, then setting it. On the road to the
Since we now have the set method that does this, we just have to
replace everywhere.
> 3.1 release, someone ran htdig through gprof and found that a
> substantial portion of time was spent in String methods. I imagine
> this is even more true as we convert more variables and parameters
> from char *.
I'm not sure. Since more fct arguments are String, much more are const,
and more String methods are inline, less copy should occur. Of course we
won't know that for sure before doing a gprof. For instance when doing
Open(config["something"], arg), no copy occur at all because Open takes
a const String and config returns a String.
There should be a benchmark + gprof for all that.
Don't worry, though, I'm not going to do any more deep modifications
like that, I'll be very busy with htword this week.
Cheers,
--
Loic Dachary
ECILA
100 av. du Gal Leclerc
93500 Pantin - France
Tel: 33 1 56 96 09 80, Fax: 33 1 56 96 09 61
e-mail: [EMAIL PROTECTED] URL: http://www.senga.org/
------------------------------------
To unsubscribe from the htdig3-dev mailing list, send a message to
[EMAIL PROTECTED] containing the single word "unsubscribe" in
the SUBJECT of the message.
