On Friday 27 September 2013 10:19:30 Derek Atkins wrote: > > I don't think it's inconsistent. We *know* which include files get > built vs which ones are part of the sources. So yes, we do treat them > differently. As I said, the alternative is doing what I suggested > above, using the full path to each header explicitly. The downside > of this alternate approach is that if you move a header (e.g., > business -> engine) then you need to change every usage of that > header. Moreover, it fails when you install the headers, unless you > keep the same hierarchy in the install-dir. > > > Geert > > -derek
Well, my knowledge on compilers is limited, so your explanation was enlightning. I'll stick to the scheme we have so far. Thanks, Geert _______________________________________________ gnucash-devel mailing list [email protected] https://lists.gnucash.org/mailman/listinfo/gnucash-devel
