KToolBar question

[Benjamin Meyer]

On Saturday 01 March 2003 2:51 am, Scott Wheeler wrote:
> On Saturday 01 March 2003 0:19, Benjamin Meyer wrote:
> > I modified KToolBar to only save settings if the settings have actually
> > changed.  From all my tests the code works just as it should without
>
> breaking
>
> > anything, but KToolBar is not the most comprehendable bit of code and so
> > I would like it is someone could take a look before I commit it.
>
> Just a note; please use 'diff -u' in the future.  I'm attaching my
> (recommended) .cvsrc.  If you have that you can just do a 'cvs diff' and it
> will show all of your changes in a format that we won't flame over.  :-)
>
> In this format, there's not context and it's difficult to read.

Thanks.  I will do that in the future.

> Now, moving on to the technical content -- what is the motivation for this?
> Has profiling shown that this is a bottleneck or a significant speed hit?
> Are you trying to speed up shut down time; is there a need for this?  Does
> this make the code more understandable?

More like it seemed like a bug that needed to be fixed.  Why should it save 
the settings if the settings didn't change?

-Benjamin Meyer

> Forget not:
>
> "We should forget about small efficiencies, say about 97% of the time:
> premature optimization is the root of all evil." -Donald Knuth
>
> Cheers,
>
> -Scott