[MERGE][#291046] Fix pushing an unstackable branch + stackable repo when there is a default stacking policy.
John Arbash Meinel
john at arbash-meinel.com
Tue Dec 16 18:37:09 GMT 2008
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Aaron Bentley wrote:
> Andrew Bennetts wrote:
>> The correct fix I think is to behave as if the command-line had said --stacked
>> --stacked-on=foo, i.e. upgrade the choice of branch format we create if stacking
>> is requested. So this patch does this.
>
> I don't think that is the correct fix. Default stacking policies are
> supposed to be non-intrusive. I would say the correct fix is not to
> stack in that circumstance.
>
> Aaron
>
>
You can only get this if you already upgraded the repository to a
stackable format. And I think it is just the fact that:
bzr upgrade --1.6 repo
Doesn't go through and upgrade all of the containing branches. I would
go as far to say that people actually meant to upgrade all of their
branches because they wanted stacking.
So I think it is a reasonable fix. I certainly understand the argument
for for not stacking at all, but I think it is a collusion of
limitations that is getting them into the situation in the first place.
Not to mention the code structuring is such that it makes it difficult
to not stack. (Branches know about repos, but repos don't really know
about branches. And while repositories are the thing that is stacked,
the branch is the one telling the repo it is stacked, which is a bit of
an inversion anyway.)
In the end, I think both are reasonable solutions, and I think the one
Andrew picked is easier, and better to have *a* solution than *no* solution.
John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iEYEARECAAYFAklH9VUACgkQJdeBCYSNAAO20QCeKOrHbPBhybiyh/I261gICyvQ
PvMAnj21zJP7p7AVjIXXjfV0Q+sthnoW
=B++H
-----END PGP SIGNATURE-----
More information about the bazaar
mailing list