[RFC] Checkout uses branch tree as a fallback accelerator

Aaron Bentley aaron.bentley at utoronto.ca
Thu Dec 20 17:20:58 GMT 2007


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

This patch allows checkout to use a branch tree as a fallback
accelerator tree.  But breaks this test:

    def test_no_upgrade_recommendation_from_bzrdir(self):
        # we should only get a recommendation to upgrade when we're
accessing
        # the actual workingtree, not when we only open a bzrdir that
contains
        # an old workngtree
        self.run_bzr('init --knit a')
        out, err = self.run_bzr('checkout a b')
        if err.find('upgrade') > -1:
            self.fail("message shouldn't suggest upgrade:\n%s" % err)

What is The Right Thing here?

Is it right to start warning about old trees when branching from them?
That would mean changing the test.

Should we suppress such warnings?  That would require code changes going
down fairly deep, or else ugly hacks.

I think I've said before that I would prefer if we only warned about
these things when we take out write locks.  That way, we know the user
*can* upgrade it.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHaqR50F+nu1YWqI0RAkH3AJ42LRa3teH8VgPggavcQk3EhL1azQCcCcvJ
6oGWAHZxi+qGa2YALx+LpY0=
=c8KG
-----END PGP SIGNATURE-----
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bzr-mail-t1i4JL.patch
Type: text/x-patch
Size: 5749 bytes
Desc: not available
Url : https://lists.ubuntu.com/archives/bazaar/attachments/20071220/ef08fea4/attachment.bin 


More information about the bazaar mailing list