#1975 NORM Untriag: sugar-native-factory always reports success

[Zarro Boogs per Child]

#1975: sugar-native-factory always reports success
---------------------+------------------------------------------------------
  Reporter:  bert    |       Owner:  danw     
      Type:  defect  |      Status:  reopened 
  Priority:  normal  |   Milestone:  Untriaged
 Component:  sugar   |     Version:           
Resolution:          |    Keywords:           
  Verified:  0       |  
---------------------+------------------------------------------------------

Comment(by bert):

 Replying to [comment:11 danw]:
 > I've reverted this change; the bug is actually in sugar-native-factory

 IMHO the shell should be robust even if the activity factory misbehaves.
 Since activities are allowed to bring their own factory we have to
 anticipate failure.

 The original bug title was "Sugar shell does not detect failure to launch
 activities" which still applies.

 "sugar-native-factory always reports success" would be a separate bug
 then, in which you assert that the create() call should actually wait
 until there is an instance opened and only then reply success? This would
 complicate the activity factory considerably, it would have to deal with
 X11 business. Maybe if you changed the return code to be just non-zero to
 mean success? The returned XID isn't used anyways. Besides, there is no
 proper spec what the create() call should do in a failure case.

-- 
Ticket URL: <https://dev.laptop.org/ticket/1975#comment:14>
One Laptop Per Child <https://dev.laptop.org>
OLPC bug tracking system