Re: [boost] Boost.SQL, SOCI and beyond

From: Nicola Musatti (Nicola.Musatti_at_[hidden])
Date: 2006-10-09 04:33:12

Steve Hutton <shutton <at> featurecomplete.com> writes:
> [Maciej Sobczak wrote:]
> 
> The important thing to note about the interface is that the overloaded
> operators (this is the controversial part) are only syntax sugar over the more
> traditional interface, so that this:
> 
> int i;
> sql << "select ...", into(i);
> 
> is really just a thin wrapper over this:
> 
> int i;
> Statement st(sql);
> st.alloc();
> st.prepare("select ...");
> st.exchange(into(i));
> st.execute(true);
> 
> In other words, SOCI does not force anybody to use the most convenient
> interface ever invented and always provides the less convenient 
> alternatives. 
> More on this: http://soci.sourceforge.net/doc/reference.html#statement

I'm aware of that but this is not really an alternative: having to remember to
perform four function calls in the correct sequence is too error prone.
Moreover, for this kind of operation I consider it preferable to submit a single
interface to the Standard Committee.

Cheers,
Nicola Musatti