Re: RFC: Moving native code in java.net into architecture dependant files

[Andrew Haley <aph at redhat dot com>]

Michael Koch writes:
 > -----BEGIN PGP SIGNED MESSAGE-----
 > Hash: SHA1
 > 
 > Am Montag, 17. M?ärz 2003 09:47 schrieb Andrew Haley:
 > 
 > >  > > If not, the naming conventions seem to establish the
 > >  > > equivalence between the eCos platform and java.net
 > >  > > functionality being disabled. Is this confusing from a naming
 > >  > > standpoint? What if instead of natInetAddressEcos.cc (for
 > >  > > example), you had natInetAddressNoNet.cc and made another
 > >  > > variable called NET_PLATFORM which factored in not only
 > >  > > PLATFORM, but also whether enable_java_net was "no"?
 > >  >
 > >  > As I understood it Ecos is more used as a synomynon for a
 > >  > non-supported architecture. At least I think I read that on the
 > >  > lists (a comment from Tom I think).
 > >
 > > For the time being, yes.  But if someone submits a patch that
 > > properly enables networking on eCos it'll be welcome.
 > 
 > What do you think about Mohan's idea of introducing NET_PLATFORM ?

It seems rather complicated.  My view is that we should support
different OSs by means of target-specific files and we should avoid
the use of conditional compilation where possible.  The main exception
to this is trivial stuff like

#ifdef WIN32
       foo ();
#endif

which isn't worth factoring out.

I wonder if we definitely need enable_java_net.  I don't see any good
reason to omit net support on a target that doesn't require such
omission.

Andrew.