fix #862 - unsub of a message ID that is already unsubbed

[CDKnightNASA]

Closes #862

Describe the contribution
After the changes implemented in #101, there may be routing table entries with no subscribers (RoutePtr->ListHeadPtr would be NULL.) This could cause a seg-fault. Also, even if there are entries in the routing table, there will be no event generated if the unsubscribe does not find a matching route entry.

Testing performed
Ran unit test (with updated event count.)

Expected behavior changes
Repeated unsubscriptions should function fine, generating an informational event if there is not a current subscription.

System(s) tested on
Debian 10.5

Contributor Info - All information REQUIRED for consideration of pull request
Christopher.D.Knight@nasa.gov

[skliper]

I appreciate that in the last couple of weeks we've seen three different loop structures to accomplish the same logic. While loop with an incrementer, for loop with a break, and now a for loop with dual exit condition. Way to keep things interesting!

[astrogeco]

I appreciate that in the last couple of weeks we've seen three different loop structures to accomplish the same logic. While loop with an incrementer, for loop with a break, and now a for loop with dual exit condition. Way to keep things interesting!

I don't know if you're genuinely excited or not-so-subtly indicating that we should choose one way and stick to it as much as possible :)

[CDKnightNASA]

I feel like we keep going around and around on this...

(Sorry, I just couldn't resist!)

[yammajamma]

@CDKnightNASA is this ready for a CCB review to get more inputs?

[CDKnightNASA]

@CDKnightNASA is this ready for a CCB review to get more inputs?

yes thx

[astrogeco]

CCB 2020-08-09 APPROVED, How do we ensure we test and catch this? Open a new issue for creating a test.

[skliper]

I opened #873 for the added test case.

[skliper]

Added the test case in #875

[skliper]

As long as it passes test w/ #875 I'm good with this.