-
Notifications
You must be signed in to change notification settings - Fork 27
Use matched count to determine server_is_available #185
Conversation
b3em
commented
Aug 23, 2017
- change the signature of internal server_is_available, removed the count_xxx function pointer
- the server_is_available function now use publication_matched and subscription_matched functions
- DCPSPublication and DCPSSubscription listeners add partition name for discovered topics
- change the signature of internal server_is_available, removed the count_xxx function pointer - the server_is_available function now use publication_matched and subscription_matched functions - DCPSPublication and DCPSSubscription listeners add partition name for discovered topics
LGTM. Do you want me to trigger a CI build for this? Otherwise please feel free to merge this. |
Yes please I like a CI build, tested locally and build fine just like to make sure I don't push anything broken into master |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
lgtm, I wasn't aware of get_publication_matched_status()
, I thought we would need to have a listener and keep track of it ourselves. We should look at using this function in our other rmw implementations.
I agree, the current listener implementation adds the topic_names (incl partition) to a list and the count functions return the count of publishers/subscribers based on the topic_name match with the list. This check is incomplete, QoS also needs to be compatible in order to count matches.
I suggest adding new functions (and maybe even remove the previous count functions)
with these function you could then get the actual number of matching publishers/subscribers. Extra information available in the get_publication_matched_status() could also be added. |
It would indeed be good to add this to the I am not sure if you would like to contribute this addition to the |
I can, but first need to make sure that the beta3-release fixes for opensplice are ready. |
Think I should have rebased instead |
Absolutely. Before you look at the |