Merge lp:~jimmiebtlr/juju-core/fix-no-peers-charm into lp:~go-bot/juju-core/trunk
Proposed by
Jimmie Butler
Status: | Rejected |
---|---|
Rejected by: | William Reade |
Proposed branch: | lp:~jimmiebtlr/juju-core/fix-no-peers-charm |
Merge into: | lp:~go-bot/juju-core/trunk |
Diff against target: |
219 lines (+59/-24) 5 files modified
charm/meta.go (+9/-9) charm/meta_test.go (+15/-6) schema/schema.go (+20/-0) schema/schema_test.go (+9/-9) testing/repo/quantal/empty-fields/metadata.yaml (+6/-0) |
To merge this branch: | bzr merge lp:~jimmiebtlr/juju-core/fix-no-peers-charm |
Related bugs: |
Reviewer | Review Type | Date Requested | Status |
---|---|---|---|
Juju Engineering | Pending | ||
Review via email: mp+218195@code.launchpad.net |
Description of the change
Allow charm with empty peers, provides, requires.
In the case that a null was read in for the maps, use
an empty map instead of throwing error. Added test case
to show error. Bug #1313793
To post a comment you must log in.
Unmerged revisions
- 2713. By Jimmie Butler
-
Go fmt.
- 2712. By Jimmie Butler
-
Change test order, put error check first.
- 2711. By Jimmie Butler
-
Change test order to check error first.
- 2710. By Jimmie Butler
-
Merge with most recent master.
- 2709. By Jimmie Butler
-
Merge in most recent master
- 2708. By Jimmie Butler
-
Remove unneeded statement in parseRelation.
- 2707. By Jimmie Butler
-
Merge master chanegs
- 2706. By Jimmie Butler
-
Add back commented empty charm fields test.
- 2705. By Jimmie Butler
-
Go fmt.
- 2704. By Jimmie Butler
-
Merge in most recent master
31 @@ -353,6 +359,7 @@ k.Interface( )) Coerce( rv.MapIndex( k).Interface( ), vpath)
32 vpath[len(vpath)-1] = fmt.Sprint(
33 newv, err := c.value.
34 if err != nil {
35 +
36 return nil, err
37 }
38 out[newk.(string)] = newv
I think this is a spurious change.
I'm a bit surprised we don't have a direct test of schema/* that would fit this change. I believe you have it covered, but only at the charm level.
Is there a test in schema that you could add that would be a bit more of a direct test?
Otherwise, LGTM.