On Thu, Sep 29, 2011 at 10:33:50AM +0200, Maciej Rumianowski wrote:
> Hi David,
> > > @@ -53,7 +54,7 @@ TYPEINIT1_FACTORY( SvxClipboardFmtItem, SfxPoolItem ,
> > > new SvxClipboardFmtItem(0
> > > SvxClipboardFmtItem_Impl::SvxClipboardFmtItem_Impl(
> > > const SvxClipboardFmtItem_Impl& rCpy )
> > > {
> > > - aFmtIds.Insert( &rCpy.aFmtIds, 0 );
> > > + std::copy(rCpy.aFmtIds.begin(), rCpy.aFmtIds.end(), aFmtIds.begin());
> > > for( sal_uInt16 n = 0, nEnd = rCpy.aFmtNms.Count(); n < nEnd; ++n )
> > > {
> > > String* pStr = rCpy.aFmtNms[ n ];
> >
> > This is totally wrong! The original line _inserts_ all items from
> > rCpy.aFmtIds at the beginning of aFmtIds. std::copy _overwrites_ the
> > first n elements of aFmtIds by items from the given range--this requires
> > that aFmtIds has sufficient size. Since this is constructor, aFmtIds is
> > always empty, therefore the line is practically guaranteed to corrupt
> > memory (unless rCpy.aFmtIds is empty). I changed it to simple copy
> > construction of aFmtIds from rCpy.aFmtIds.
> Thanks, sorry for that. I thought if I am working on vector than i will
> automatically resize, but I missed in description that I have to
> manually resize.
Hi,
it will if you use vector fuctions. But standard algorithms work with
iterator ranges and know nothing about the structure behind them. E.g.,
std::copy will work just as well if the destination is plain old array.
This is nothing to be ashamed of--I suppose every one of us made the
same (or similar) invalid assumption at some time in the past :)
D.
_______________________________________________
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice
