branches, bugs, and landings for 2.0 and beyond

[John Arbash Meinel]

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

Martin Pool wrote:
> 2009/9/10 Matthew D. Fuller <fullermd at over-yonder.net>:
>> On Thu, Sep 10, 2009 at 03:04:09PM +1000 I heard the voice of
>> Martin Pool, and lo! it spake thus:
>>>  2.0.0 aimed at bzr 2.0final
>> This should definitely be '... aimed at bzr 2.0.0', as your later mail
>> suggests.
> 
> True, I think this shows it is clearer that way.
> 
>>>  2.0 aimed at 2.0.1
>> Phrase this rather as "aimed at 2.0.x"; 2.0.1 will come from a 2.0.1
>> branch branched from 2.0 when we get to rc'ing it.
> 
> Right.

If we are going to do this, then I propose this patch for bzr:

=== modified file 'bzrlib/__init__.py'
- --- bzrlib/__init__.py  2009-08-30 21:34:42 +0000
+++ bzrlib/__init__.py  2009-09-10 15:42:01 +0000
@@ -81,7 +81,7 @@
     ...
     ValueError: version_info (1, 4, 0, 'wibble', 0) not valid
     """
- -    if len(version_info) == 2 or version_info[2] == 0:
+    if len(version_info) == 2:
         main_version = '%d.%d' % version_info[:2]
     else:
         main_version = '%d.%d.%d' % version_info[:3]

This changes 'bzr --version' to print:
  bzr 2.1.0dev

rather than
  bzr 2.1dev

It also will tell my win32 build script to build:
  bzr-2.0.0rc2

Rather than
  bzr-2.0rc2

Thoughts?

John
=:->
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkqpHpEACgkQJdeBCYSNAAMVngCfYN6jzF/eRkK+ptHoJZU7hEwT
o20AoKGwDppDZhuZQrAhOwWaW/uTfDXH
=kOwF
-----END PGP SIGNATURE-----