-
Notifications
You must be signed in to change notification settings - Fork 14
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
scheduler_msgs: revise Request.status labels #60
Comments
New transition graph documentation: |
Looks good Jack. I actually don't have anything to discuss here :) |
Sure. I am not yet ready to commit this change, but wanted to put the ideas in writing where they will be handy when the time is right. Maybe something more will come up after further development. If it does, let's add it here. |
remains backwards compatible with earlier scheduler_msgs/Request uses CANCEL states in place of RELEASED, RELEASING avoids deprecated labels
I closed pull request #64 so we can consider additional changes. I want to only make one update for this go-around. |
rename terminal status label CLOSED add reason field to supplement status
The previous commit renames the terminal status to CLOSED. It also adds a |
scheduler_msgs: Request message revisions (#60)
scheduler_msgs: backport Request revisions to Hydro (#60)
I tried to take out the backwards compatibility logic in my code (utexas-bwi/rocon_scheduler_requests#21). The pre-release tests build dependencies from source, so they work. But the devel tests install binary dependency packages, so they fail on the build farm. So, I need a new hydro branch release to avoid having to put back the compatibility logic. Would that be OK? |
Sure. Just send a pull request and I'll bump a new release anytime. I have scripts which mean it takes all of 5mins for me to do. |
The updates are already merged from #67, I just need a Hydro release. |
This issue is a forum for discussing scheduler request status updates. The proposed changes will be made in two steps:
The current proposal is:
With this proposal, CANCELED becomes the only terminal state. Once acknowledged, the canceled request will be deleted.
As best I can tell, the abort() and reject() events can handled via preempt(), simplifying the state transitions considerably.
The text was updated successfully, but these errors were encountered: