Re: Carbon port emacs-unicode-2 build problem under MacOSX

[Dan Nicolaescu <dann@ics.uci.edu>]

Jason Rumney <jasonr@gnu.org> writes:

  > Adrian Robert wrote:
  > > I was planning to ask about inclusion after finishing up the multi-tty
  > > integration.  But anyway since the issue has been raised, is autumn 2007 a good
  > > time to consider merging the Cocoa/GNUstep port into the unicode-2 branch, or
  > > would people prefer to wait until after unicode-2 -> trunk?  Based on my
  > > experiences, the latter approach would be much less headache, but don't know how
  > > far off the unicode merge is at the moment.
  > >   
  > 
  > 
  > I'd suggest the following timetable.
  > 
  > 1. Start the 22.2 pretest (hopefully a short pretest cycle).
  > 2. Release 22.2 without getting distracted by more merges.
  > 3. Merge unicode-2 to the trunk (I think it is ready now, so immediately
  > after 22.2 release would be good).
  > 4. Merge Cocoa port (if 22.2 pretest starts taking a long time though,
  > it might be better to merge to unicode-2 earlier).
  > 5. Enable new font-backend by default.
  > 6. Remove the old font code (which unless someone implements a backend
  > for Carbon, implies the end of the Carbon port)

This looks good, I would also like to add a step that can be done in
parallel: 

- Have the Cocoa port reviewed so that there are less questions about
  what it does and how it does stuff when it is included. Probably the
  changes to the common code are the most important.

Is the Cocoa port a complete replacement for the Carbon port from the
_user_ point of view? Does it support the same machines equally well?
If yes, then there seems to be little point in having both in the
tree, it might be better to remove the (already non-functional) Carbon
port as early as possible (now?).

Besides reducing complexity, it seems that there are not that many
people willing to work on the Mac platform, so having to deal with
less code might be a good idea.