bzr-svn 0.3 - Other weird errors

[Aaron Bentley]

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

John Arbash Meinel wrote:

> I've talked with Jelmer a bit on this. And he seems to really want to
> avoid having a bzr revision id stored. I'm not 100% sure why, but I do
> know it is something he is explicitly avoiding.

This is up to him, of course.  But I don't think bzr-svn should 'fetch'
to svn if it actually is not capable of preserving revision ids.

For the specific case of bound branches, it can be fixed by committing
remotely first, as he notes.  (though this doesn't address commit --local)

> In my mind, if the data comes from bzr, then you have commit access to
> svn, and it is fine to set some properties. (We already have to set
> file-ids for reasons like this).
> 
> As long as it is clear what to do when those cannot be set (because you
> have a readonly svn source). Which I think we have working already.

Agreed.  To support write from bzr, you have to stick bzr's properties
SOMEWHERE.  The repository seems like a good choice from a
discoverability perspective.  Perhaps a standalone 'property supplement'
store would also work.

Aaron
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFFsPOJ0F+nu1YWqI0RAsuIAJ93yS2EbIn2o8/Z9spjpWCH4Jk48wCeMyoP
R9YPMPtdSp2NfiBFSu539Sc=
=a5eV
-----END PGP SIGNATURE-----