Re: Question on coding style in networking code

[george anzinger <george@mvista.com>]

chuckw@ieee.org wrote:
> 
>         Thank you for the reply.
> 
>         I absolutely agree that it is much easier to read and figure out what is
> going on.  You don't have to keep going back to the struct declaration to
> find out what the fields are.
> 
> Thanks again,
> Chuck
> 
> On Mon, Aug 20, 2001 at 12:56:53AM -0400, Alexander Viro wrote:
> >
> >
> > On Sun, 19 Aug 2001 chuckw@ieee.org wrote:
> >
> > >     struct x y = {
> > >             member1: x,
> > >             member2: y,
> > >             member3: z
> > >     };
> > >
> > > What is the deal with this?  Does the second way have any advantage over the previous?
> >
> > _Much_ easier to grep for. Less pain in the ass when fields are added/removed/
> > reordered.
> >
> > For anything with many fields (usually method tables) it's more convenient.
> > And no, it's not just networking - filesystem-related code, etc. uses it
> > all over the place.
> >
For large structures it is a pure joy to work with.  Check out sched.h
and the "task_struct" where many of the fields come and go with "CONFIG"
options.

George