arpadboda commented on a change in pull request #735: MINIFICPP-1158 - Event
driven processors can starve each other
URL: https://github.com/apache/nifi-minifi-cpp/pull/735#discussion_r385079077
##########
File path: libminifi/include/ThreadedSchedulingAgent.h
##########
@@ -73,6 +73,8 @@ class ThreadedSchedulingAgent : public SchedulingAgent {
ThreadedSchedulingAgent(const ThreadedSchedulingAgent &parent);
ThreadedSchedulingAgent &operator=(const ThreadedSchedulingAgent &parent);
std::shared_ptr<logging::Logger> logger_;
+
+ std::set<std::string> processors_running_;
Review comment:
For sake of simplicity: the most straight forward container in case you have
to search by value.
Performance doesn't matter here as this code piece is only executed during
flow change (startup, shutdown, update).
Because of the above I find this point of the guideline a bit
overprotective. In case it's a core of a low latency software, yes, justify
your container selection (and hopefully measure it, too!), but in our case,
it's not that kind of software and definitely not a part of the code that has
any effect on throughput. In this case I would suggest choosing the container
that results in the most straight forward code.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
