Why not updating in-place?

[peci1]

Is there a good reason for not having function

virtual bool update(T& data)=0;

in FilterBase?

It seems to me it might help getting the filter chain even more efficient.

Regarding backwards compatibility, it'd probably need to have a default implementation that calls the copying update(data_in, data_out), but filters that are capable of working in-place would get the chance to do so.

[tfoote]

Improving support for in place would be reasonable. I believe that many of them are coded so as to actually support that by passing the same reference to both arguments. But possibly setting a flag that can be used in in the tempalte logic for the FilterChain might be much more powerful. The availability of the single argument update is one such indicator that could be used.