Add failing provider notice and fix #1173333

Add failing provider notice and fix #1173333

- When a provider is known to be failing add a warning container and show the
provider icon. 
- Drive by fix for #1173333 to make sure that we can 'deactive' a selected
charm in the editorial and search views.

https://code.launchpad.net/~rharding/juju-gui/provider-display/+merge/161267

(do not edit description out of merge proposal)

[rharding, 2013-04-27 01:39:50]

Please take a look.

[rharding, 2013-04-27 13:31:17]

Please take a look.

[rharding, 2013-04-27 13:48:55]

Please take a look.

[gary.poster, 2013-04-27 17:06:02]

code looks good.  will qa later today.

https://codereview.appspot.com/8651046/diff/6001/test/test_browser_editorial.js
File test/test_browser_editorial.js (right):

https://codereview.appspot.com/8651046/diff/6001/test/test_browser_editorial....
test/test_browser_editorial.js:164: it('unsetting the activeID will remove the
active', function() {
the active...duck? the active flagellum? :-)

removes the active marker?

[rharding, 2013-04-29 10:42:22]

Please take a look.

[rharding, 2013-04-29 10:51:26]

Originally the mysql and cinder charms were failing on ec2 and usable for QA. On
my ec2 instance of charmworld the test results aren't loading so it's not able
to be used for qa'ing. Hopefully staging or manage. will come up to enable test
results for it.

[j.c.sackett, 2013-04-29 15:06:21]

https://codereview.appspot.com/8651046/diff/1017/app/models/charm.js
File app/models/charm.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/models/charm.js#newcode415
app/models/charm.js:415: failing.push('hp');
Does this mean that HP cannot independently fail? Or just that openstack and HP
tests for our purposes are always the same?

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/ed...
File app/subapps/browser/views/editorial.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/ed...
app/subapps/browser/views/editorial.js:113: // Watch for changse to the activeID
so that we can mark/unmark active
s/active/activeID/

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/se...
File app/subapps/browser/views/search.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/se...
app/subapps/browser/views/search.js:118: },
Can we flag this and the other instance of essentially the same code in
editorial with a todo note about sharing this code? We probably need a
'hasActiveCharm' extension or something.

[rharding, 2013-04-29 15:09:51]

Feedback below.

https://codereview.appspot.com/8651046/diff/1017/app/models/charm.js
File app/models/charm.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/models/charm.js#newcode415
app/models/charm.js:415: failing.push('hp');
On 2013/04/29 15:06:22, j.c.sackett wrote:
> Does this mean that HP cannot independently fail? Or just that openstack and
HP
> tests for our purposes are always the same?

Correct. According to the juju test people: the openstack test is to run the
charm tests on HP cloud. If that tests fail, both are considered failing.

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/ed...
File app/subapps/browser/views/editorial.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/ed...
app/subapps/browser/views/editorial.js:113: // Watch for changse to the activeID
so that we can mark/unmark active
On 2013/04/29 15:06:22, j.c.sackett wrote:
> s/active/activeID/

This is watching the activeID ATTR in order to determine if we need to unset the
'active' class set.

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/se...
File app/subapps/browser/views/search.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/se...
app/subapps/browser/views/search.js:118: },
On 2013/04/29 15:06:22, j.c.sackett wrote:
> Can we flag this and the other instance of essentially the same code in
> editorial with a todo note about sharing this code? We probably need a
> 'hasActiveCharm' extension or something.

Filed as #1174358

[j.c.sackett, 2013-04-29 15:12:58]

Thanks for responses; LGTM.

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/ed...
File app/subapps/browser/views/editorial.js (right):

https://codereview.appspot.com/8651046/diff/1017/app/subapps/browser/views/ed...
app/subapps/browser/views/editorial.js:113: // Watch for changse to the activeID
so that we can mark/unmark active
On 2013/04/29 15:09:52, rharding wrote:
> On 2013/04/29 15:06:22, j.c.sackett wrote:
> > s/active/activeID/
> 
> This is watching the activeID ATTR in order to determine if we need to unset
the
> 'active' class set

Ah, ok, misread it.

[rharding, 2013-04-29 19:34:02]

*** Submitted:

Add failing provider notice and fix #1173333

- When a provider is known to be failing add a warning container and show the
provider icon. 
- Drive by fix for #1173333 to make sure that we can 'deactive' a selected
charm in the editorial and search views.

R=gary.poster, j.c.sackett
CC=
https://codereview.appspot.com/8651046