[MERGE] bzr ls --debris

[Aaron Bentley]

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

John Arbash Meinel wrote:

> I thought 'self.runbzr' was mostly deprecated. Especially since you have
> to pass the ugly 'backtick=True'.

For blackbox testing, I prefer ('ls --debris') to ('ls', '--debris').
It reads more like what I would type.  I don't remember hearing anything
about it being deprecated.

> What about using:
> 
> output = self.run_bzr('ls', '--debris')[0]

Sure, I can do that.


> Should debris_classifier be part of WorkingTree? It doesn't seem like
> something WorkingTree is using directly. And seems more like we should
> have maybe a separate file with customized classifiers.

Martin said he thought it was a good idea to have the 'is_detritus' test
be on WorkingTree, so putting the classifier there seemd like the next
logical step.

> ....
> 
> If we are switching to classifiers, then Tree.list_files also needs to
> be fixed.

Thanks for catching that.  Done.

> At the very least, we need to add a test which runs 'bzr ls --debris -r
> 1' so that this code path is tested.

Okay, I'm testing unit testing Tree.list_files, EmptyTree.list_files and
WorkingTree.list_files, plus 'ls --debris -r -1'.

> Other than needing to fix Tree.py, adding a test case, and deciding
> whether or not 'debris_classifier' and 'is_debris_filename' should be
> part of WorkingTree, +1.

Okay, so the only remaining issue is where to keep the debris :-)

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

iD8DBQFEkdIb0F+nu1YWqI0RAh8CAJ9LP829aKQdcAcQTUu32qH1nGVSBgCdHNQ7
5uzblai3DstgmCyY7AGWRIQ=
=+G6X
-----END PGP SIGNATURE-----