Re: Performance: Loading a large plain data file takes very long - solution proposal

On 12/04/10 15:24, Bartosz wrote:
> Loading a large plain data file takes very long.
> This problem was described at:
> http://wiki.services.openoffice.org/wiki/Calc/Performance/The_Ou_case
>
> Look at file:
> http://svn.services.openoffice.org/opengrok/xref/DEV300_m94/sc/source/filter/xml/xmlsubti.cxx
>
> 38% of time spent in ScMyTables::NewColumn() because of replicated use of aTableVec[nTableCount - 1] (vector::operator[])
>
> What do you think about instead using:
>      aTableVec[nTableCount - 1]
> use instead:
>      aTableVec.back()
>
> I think it could increase an perfomance slightly.

Did you measure if that makes a real difference?
Eike's suggestion on the wiki page is to reduce the number of operator[] 
calls by obtaining a pointer or reference once. That will probably help 
more.

Niklas