On Mon, Jun 10, 2013 at 7:04 PM, Junio C Hamano [off-list ref] wrote:

Felipe Contreras [off-list ref] writes:

quoted

quoted

*1* ... which is a very reasonable thing to do.  But moving
    sequencer.o to builtin/sequencer.o is *not* the way to do this.

By now we all know what is the *CURRENT* way to do this; in other
words, the status quo, which is BTW all messed up, because builtin/*.o
objects depend on each other already.

builtin/*.o are allowed to depend on each other.  They are by
definition builtins, meant to be linked into a single binary.

That's not what John Keeping said[1]. I'm going to assume he was
wrong, but I don't think that's relevant for my point.

Either way, the meaning of builtin/ should probably be explained somewhere.

quoted

We are discussing the way it *SHOULD* be. Why aren't you leaning on that?

And I do not see the reason why builtin/*.o should not depend on
each other.  It is not messed up at all.  They are meant to be
linked into a single binary---that is what being "built-in" is.

A good way forward, the way it *SHOULD* be, is to slim the builtin/*.o
by moving parts that do not have to be in the single "git" binary
but are also usable in standalone binaries out of them.

And that is what I just suggested.

But init_copy_notes_for_rewrite() can *not* be used by anything other
than git builtins. Standalone binaries will never use such a function,
therefore it doesn't belong in libgit.a. Another example is
alias_lookup(). They belong in builtin/lib.a.

[1] http://article.gmane.org/gmane.comp.version-control.git/227017

-- 
Felipe Contreras