Re: [RFC] Submodules in GIT

Andy Parkins wrote:

On Thursday 2006 November 30 11:57, sf wrote:

quoted

quoted

Worse, if you allow that to happen, the supermodule can commit a state
that cannot be retrieved from the submodule's repository.  The ONLY thing
a supermodule can record about a submodule is a commit.

So what? You have a submodule commit that only exists in the
supermodule. I fail to see the problem. The changes you made to the
submodule _in the supermodule_ can later be pulled from wherever you want.

Eh?  The files aren't stored in the supermodule, they're stored in the 
submodule.  The ONLY thing in the supermodule is the commit hash.  The 
objects for the submodule are still /in/ the submodule.

But you have got the submodule on your local disk anyway. So just setup 
alternates and the supermodule contains all of the submodule.

It sounds like you're suggesting that the supermodule commit includes files 
from the submodule?  How can that work?   The two aren't separate entities 
then, it's just one big repository. 

It works as it always works in git: The supermodule commit contains the 
submodule commit, the submodule commit contains the submodule files, so 
the supermodule contains the submodule (at least the part of the 
submodule that is visible). It _must_ be one repository but it need not 
be big (once more, use alternates).

I mean, what would this supermodule commit look like?  Would it include a 
commit message?  Which module should that commit message be about?  Should 
the commit's parents be stored?  Which parents, the submodule HEAD or the 
supermodule HEAD?  Which tree object should it link to?  The one in the 
submodule doesn't exist, so it'll have to be a freshly made up one for the 
supermodule - except now you've put submodule paths in the supermodule.  
Nope.  That's never going to work.

Again I do not see the problem. Probably I have a much simpler picture 
of submodules: They are just commits in the supermodule's tree. 
Everything else follows naturally from how git currently behaves.

Of course it works. It is simple, it is the git way.

Am I missing the point?

Regards

Stephan

-- 
b.i.t.
beratungsgesellschaft für informations-technologie mbh
Stephan Feder
elisabethenstr. 62   fon: +49(0)6151/827575
64283 darmstadt      fax: +49(0)6151/827576
mailto:sf@b-i-t.de   www: http://www.b-i-t.de