Joaquin,
many thanks for your detailed answer. I decided to
handle this problem in a bit different manner. To split the entry class
in 2, where the first one with field x contains all to x related tuple
items as an indexed collection. That works fine and I can use
boost::transform_iterator and boost::filter_iterator to efficiently
access the content of the entry without copying the iterators into the
vector.
Many thanks again,
Ovanes
On Tue, Mar 4, 2008 at 11:06 PM, Joaquin M Lopez Munoz <
joaquin@tid.es> wrote:
>Hi Ovanes,