Merge the initial codes

[马旋（SuperMMX）]

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

Hi, Martin Pool <mbp at canonical.com> :

On Mon, 22 Dec 2008 14:28:57 +1100
Martin Pool <mbp at canonical.com> wrote:

> On 18 Dec 2008, 马旋（SuperMMX） <supermmx at gmail.com> wrote:
> > 
> > Hi, Bazaar guys:
> > 
> > bzr 1.9 on win32 (XP)
> > 
> > I branch feature from the empty trunk, and merge it back, but when showing status,
> > it displays "working tree is out of date, run 'bzr update'".
> > 
> > It seems that you must have commits in the trunk, otherwise merging will not success.
> 
> That is probably true.  bzr can't really represent the concept of the
> first commit being a merge into that branch.  Revisions have zero or
> more parent revisions.  Normally one parent for a regular commit, or two
> for a merge.  It's the fact that there are multiple parents that makes a
> revision a merge.  If there's no mainline parent, then the first
> revision can't be a merge.
> 
> To avoid this, just make a commit on the trunk to start it before doing
> anything else.

Right, I understand that, should this be noted somewhere in documents?
So users can avoid such situation even there are wordarounds for this.
(e.g, commit in trunk, and merge -r xx.xx feature)

Thanks for you explanation.

- -- 
A. Because it makes the logic of the discussion difficult to follow.
Q. Why shoudn't I top post?
A. No.
Q Should I top post?

A: Because it destroys the flow of the conversation
Q: Why is it bad?
A: No, it's bad.
Q: Should I top post in replies to mailing lists? 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (MingW32)

iEYEARECAAYFAklPDVoACgkQYjhzyV/TMxsxWgCcDca1Uow0kH8nqzYjzv/49IhC
dOIAn1whvOiOIQlpiE+IqF6LkYgxP3qZ
=Efez
-----END PGP SIGNATURE-----