On Sat, Apr 02, 2016 at 06:33:00PM -0500, Linus Torvalds wrote: > On Apr 2, 2016 18:14, "Dirk Hohndel" <[email protected]> wrote: > > > > So you add the dive pointer to the function signature, but you don't do > > anything with it, creating an unused parameter. Why? > > Next patch will do it. > > The dive pointer is necessary to look up the cylinder, but this patch tries > to avoid actually changing behavior. > > So all of this patch is basically prep. > > That said, it does change behavior in some ways: just by virtue of setting > the cylinder number this patch triggers behavior changes, because having a > cylinder index on the gas change will mean that it gets shown and saved. > > Similarly, if the gas change cannot actually be matched up to any cylinder, > it will be thrown away, again being technically a behavior change. > > So it's not entirely just infrastructure with no behavior change at all, > but this patch basically just does stricter checking and introduces > behavior that you *could* get before by specifying the cylinder index. > > The next patch that actually uses the cylinder gas mix information is a > more fundamental behavior change, in the sense that it actually introduces > behavior that you couldn't even force before. > > Does that explain the patch?
Yep, that works. I'll fix the whitespace bubu :-) /D _______________________________________________ subsurface mailing list [email protected] http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
