On Mon, 2003-10-27 at 11:17, Bryan Henderson wrote:
> >> OK, so there appears to be agreement that the source tree should be
> >> rearranged to be more managable. How would we go about doing that?
> >
> >Well, we'd first decide what the rearrangement should be; there's not
> >much you can do to rearrange stuff if you don't know where you're going
> >to put it.
>
> I didn't think my question about the mechanics of moving files would
> depend on where we're moving them to. So let's just look hypothetically
> at the one thing everyone seems to agree on: moving the packet_*.c files
> from the top level directory into a packet/ subdirectory. This has been
> found to be a good change a couple of times, but didn't go anywhere. I'm
> asking what could _I_ do to move it along. I thought we could talk about
> the mechanics in parallel with talking about where we want the files to
> end up.
>
Fix the rule that creates register.c so that more than one source
location can be used. I'm not sure if this requires changing the two
possible scripts used to create register.c -- make-reg-dotc.py and
make-reg-dotc. You'll probably move the rule for creating register.c
from the top-level Makefile.am to epan/packet/Makefile.am. (I'm in favor
of putting the dissectors in a subdirectory of the epan directory).
Then the dissector sources can be moved one at a time, or in small
bunches, to the new location. When it's all done, the rule to create
register.c can be modified again so that it doesn't look in the
top-level directory for dissector sources.
--gilbert
