DESCRIPTION

tmf: Add Guava dependency

This allows usage of some Guava goodies, namely additional
Collections (MultiMap etc.), the convenience methods (Lists.*),
and the ImmutableCollection versions.

Change-Id: I85c4fb32e7d5db368c35f17109563caedb2c9d73
Signed-off-by: Braxton Mccarthy xxx@xxx.xxx


COMMENTS

author: Braxton Mccarthy
date: 2014-04-08 19:51:52.000000000

Uploaded patch set 3.

-------------------------------------
author: Braxton Mccarthy
date: 2014-04-08 19:58:23.000000000

Patch Set 3:

Fixed the comments. Also depended on version 12.0 explicitely in the targets, because the CQ only covers v12.

-------------------------------------
author: Brenden Conley
date: 2014-04-08 20:21:55.000000000

Patch Set 3:

Build Started https://hudson.eclipse.org/linuxtools/job/linuxtools-gerrit/6300/

-------------------------------------
author: Brenden Conley
date: 2014-04-08 21:06:23.000000000

Patch Set 3: Verified+1

Build Successful 

https://hudson.eclipse.org/linuxtools/job/linuxtools-gerrit/6300/ : SUCCESS

-------------------------------------
author: Kaeden Tapia
date: 2014-04-08 23:16:00.000000000

Patch Set 3: Verified+1

Looks good. Let discuss this with others before merging it though.

-------------------------------------
author: Annalise Jimenez
date: 2014-04-09 00:55:01.000000000

Patch Set 3: Code-Review+1

I'm all for it!

-------------------------------------
author: Annalise Jimenez
date: 2014-04-09 00:58:28.000000000

Patch Set 3: Code-Review-1

Though what do you mean with CQ only covers V12? 

We've had problems with target that specify an exact version and we will in time with this: the package gets deleted and people get errors they don't understand when installing the development environment. I'd prefer the 0.0.0 version

My +1 becomes a -1 until that is clarified.

-------------------------------------
author: Braxton Mccarthy
date: 2014-04-09 01:20:07.000000000

Patch Set 3:

Actually, I'm not really sure if additional CQs are needed to simply use a newer versions, or if it's to bring that new version in Orbit if it is not already.

In any case, I think it's safer to use specific version numbers in general. We specify the exact update site, so we are guaranteed that this particular version will exist in that site (it won't break magically 5 years from now). By using 0.0.0, there's more chance of things breaking because a newer version might break their API.

Also, there's a bug right now in the target editor, where 0.0.0 doesn't always pick up the latest version available. Marc-André can tell you more about it ;)

-------------------------------------
author: Annalise Jimenez
date: 2014-04-09 01:35:53.000000000

Patch Set 3: Code-Review+1

I've debugged a few students where the version in the repository was not the same as the one mentioned in the target file so using the hard-coded way is not full bullet proof either...

I guess there's no total solution. Anyway, you probably know better about that kind of stuff than I do, so I'll go back to +1 ;-)

-------------------------------------
author: Ivy Mitchell
date: 2014-04-09 14:22:35.000000000

Patch Set 3: Code-Review-2

Adding a -2 until the cq is done so we don't accidentally push it. If only we had an IPClean field.

-------------------------------------
author: Kaeden Tapia
date: 2014-04-09 14:53:29.000000000

Patch Set 3:

"I've debugged a few students where the version in the repository was not the same as the one mentioned in the target file so using the hard-coded way is not full bullet proof either..."

You are thinking of the swtbot problem, right? For orbit packages, that's not a problem, all versions they are always kept.

BTW, there is a patch to fix the 0.0.0 not always getting the latest versions (see https://git.eclipse.org/r/#/c/24677/).

-------------------------------------
author: Ivy Mitchell
date: 2014-04-09 15:42:25.000000000

Patch Set 3: Code-Review+2

CQ was approved.

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