Re: [Libreoffice] [REVIEW] request for review for patch for fdo#32742# regression

On 12/01/11 18:03, Norbert Thiebaud wrote:
> On Wed, Jan 12, 2011 at 10:35 AM, Octavio Alvarez
> <alvarezp@alvarezp.ods.org>  wrote:
> > On Wed, 12 Jan 2011 03:18:01 -0800, Noel Power<nopower@novell.com>  wrote:
> >
> > > Hi there
> > >
> > > fix for fdo#32742#, its a simple fix ( but was a bugger to find ) imo is
> > > riskless etc.
> > >
> > >
> > > http://cgit.freedesktop.org/libreoffice/base/commit/?id=0afe2266016b03f6e11008463042c7daacead0e1
> > > is ripe for signoff and cherrypicking, be grateful if someone could help
> > > with that
> > Is suggest to make ::std::sort on the next line use s_nCount also and
> > prevent a useless division operation.
> Actually the 'useless' division' is resolved at compile time. so it is free.
> on the other hand passing s_nCount instead of a constant _could_
> deprive the template code of some possible optimizations.
> (I don't know if std::sort is smart enough to take avantage of it..
> but it could, and eventually it will)
>
> Norbert
the other thing to remember is that this is some static initialisation, 
it is done just once, so even thinking about optimising that would seem 
to be overkill :-)