DESCRIPTION

Decorate folder as dirty if subtree contains untracked files

The new decorator implementation did not decorate folders as dirty if
untracked files existed in the subtree.

Bug: 345099

Change-Id: Ibee5d0dad240e64ad366ab4ed33a008fb1d22c50
Signed-off-by: Marlie Beasley xxx@xxx.xxx


COMMENTS

author: Marlie Beasley
date: 2012-01-13 12:19:14.000000000

Uploaded patch set 2.

-------------------------------------
author: Brenden Conley
date: 2012-01-13 12:34:37.000000000

Patch Set 2:

Build Started https://hudson.eclipse.org/sandbox/job/egit.gerrit/2313/ 

-------------------------------------
author: Brenden Conley
date: 2012-01-13 12:50:56.000000000

Patch Set 2: Build Successful

Build Successful 
 
https://hudson.eclipse.org/sandbox/job/egit.gerrit/2313/ : SUCCESS

-------------------------------------
author: Sebastian Porter
date: 2012-02-06 09:34:28.000000000

Patch Set 2: I would prefer that you didn't submit this

The patch improves things but there are two missing pieces:
- Files should also get the '>' decoration to be consistent 
  with modified files. This would also be consistent to the
  CVS team provider.
- I know folders aren't directly tracked by Git but since
  an empty folder has the '?' decoration it should it 
  should also propagate the '>' upwards and have the '>'.

-------------------------------------
author: Marlie Beasley
date: 2012-02-06 13:29:52.000000000

Patch Set 2:

I'm not sure on this:

1. I would prefer if I do not get a dirty marker on project level if there is an empty folder inside the project. I think only important information should bubble upwards.

2. This fix brings back the features of the old decorator implementation.
Decorating untracked files with > would be a change that needs discussion on the mailing list.

-------------------------------------
author: Sebastian Porter
date: 2012-02-06 13:46:38.000000000

Patch Set 2: No score

re 1: '>' stands for "dirty" and for me, as soon as there is any diff compared to what's in the repository, such a project is dirty. Otherwise you have to rename the label for '>' to "interesting changes". Also, the fact that EGit shows a '?' indicates that it is not completely unimportant.

re 2: I see that the information is redundant but keep in mind that for disabled people, the icon is not "readable", hence '>' is an important information for them.

-------------------------------------
author: Sebastian Porter
date: 2012-02-06 13:47:05.000000000

Patch Set 2: I would prefer that you didn't submit this



-------------------------------------
author: Marlie Beasley
date: 2012-02-08 12:06:18.000000000

Patch Set 2:

I propose the following:

1. Let's accept this change as is (restores the old decoration behavior)

2. Start a discussion on the mailing list about the described decoration changes

-------------------------------------
author: Sebastian Porter
date: 2012-02-22 13:50:05.000000000

Patch Set 2: Looks good to me, but someone else must approve

>I propose the following:
>
>1. Let's accept this change as is (restores the old >decoration behavior)
>
>2. Start a discussion on the mailing list about the >described decoration changes

The current patch already helps a lot, hence I think it's best to proceed as you suggest.

-------------------------------------
