current key TODO's [my opinion only)

[Aaron Bentley]

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

John Arbash Meinel wrote:
> Aaron Bentley wrote:
> 
>>Robert Collins wrote:
>>| * +x bit preservation on windows [will need working tree format bump,
>>| thus blocked on bzr-dir branch]
>>
>>There is already x-bit preservation on Windows.  There's no UI for
>>*changing* the x-bit though.

> Actually, no there isn't. The problem is that the working tree inventory
> gets sanitized when it is written out.

Oh.  Odd.  Anyhow, it's certainly supported by tree transform.

> When I looked into it, I found that it wasn't easy to fix. The problem
> was that the code that knew what was going on during merging (whether
> the bit was going to be changed or not) only 'recorded' that information
> by modifying the file. Which is where you and I discussed having the
> apply_changeset code actually modify a Tree object, rather than doing
> the modifications directly to the filesystem only.

That would be my preference, yes.  I did it the ugly way in Tree
Transform, since there isn't a corresponding function on WorkingTree to
set the executible bit.

Aaron

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

iD8DBQFD4Mjs0F+nu1YWqI0RAjHFAJ9c+bds+tmMtpOy5AEHxyWtwkCvSwCfVUJK
Pf2DZPBDn0bqi9CQhm4gsS0=
=El2q
-----END PGP SIGNATURE-----