Ticket #196 (accepted planned)
sane shutdown behaviour of the CommandRegistry
| Reported by: | ichthyo | Owned by: | ichthyo |
|---|---|---|---|
| Priority: | lesser | Milestone: | 1alpha |
| Component: | lumieraProc | Keywords: | |
| Cc: | Blocked By: | #295 | |
| Blocking: | #195 |
Description (last modified by ichthyo) (diff)
reminder @self
- can it happen that CommandImpl deleter functions are invoked after the registry is gone??
- how can we prevent new commands to be issued after shutting down the registry??
Change History
comment:1 Changed 3 years ago by ichthyo
- Status changed from new to accepted
- Description modified (diff)
comment:2 Changed 2 years ago by ichthyo
- Owner ichthyo deleted
- Status changed from accepted to assigned
comment:3 Changed 2 years ago by ichthyo
- Owner set to ichthyo
- Status changed from assigned to accepted
this was just a Trac/Ticket test: you can indeed just remove the owner when re-assigning
comment:4 Changed 2 years ago by ichthyo
When we enter the destructor-phase of application shutdown, we can care within ~CommandRegistry to clean the indices. Now problems can only arise if someone still holds a Command (handle) at that point. As the Command objects usage is quite libearl (and the only central facility is the CommandRegistry), the only chance lies within the Commands themselves...
Note: See
TracTickets for help on using
tickets.

obviously my job...