Martin Vermeer wrote:
> 350 #warning CHECK! Are we comparing last_paste to the wrong list
> here? (Lgb)
> 351 if (boost::next(last_paste) != pars.end() &&
> 352 paste_the_end) {
>
> This remark looks valid to me...wrong comparison.
Not to me.
> What is doubleplusunnice about this, it that *usually*, by pure
> chance, the inequality will hold. So any malfunctioning due to this
> will be very sporadic.
If it's the wrong list, then the inequality would always hold (giving
crashes etc). I'd bet it's correct.
> Isn't the healthy thing to remove the inequality test and see if we
> can get some *clear* malfunctioning? And if nothing happens, leave it
> at that...
Not very healthy, as we have things like
boost::next(last_paste)->makeSameLayout etc. So I'd say the check it's
needed.
A few lines above we have:
pars.splice(past_pit, simple_cut_clone);
ParagraphList::iterator last_paste = boost::prior(past_pit);
so past_pit is an iterator of pars
so last_paste also
Alfredo
