Cleanup gmock Test Dependencies

[bjin2364]

Changes

• something that's slightly been bothering me are the gmock test deps, when none of the unit tests are using gmock. This PR updates the dependencies so they're gtest

Testing

• passing unit tests

Notes

• This PR is stacked on top of Update ThreadSafeBox Tests #459

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 85.33%. Comparing base (69bd9a3) to head (1beb22f).
⚠️ Report is 7 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #460      +/-   ##
==========================================
+ Coverage   85.25%   85.33%   +0.08%     
==========================================
  Files          17       17              
  Lines        1397     1398       +1     
  Branches      132      132              
==========================================
+ Hits         1191     1193       +2     
+ Misses        120      119       -1     
  Partials       86       86              

Flag	Coverage Δ
unittests	85.33% <100.00%> (+0.08%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
realtime_tools/test/lock_free_queue_tests.cpp	94.04% <ø> (+0.59%)	⬆️
realtime_tools/test/realtime_buffer_tests.cpp	100.00% <ø> (ø)
realtime_tools/test/realtime_mutex_tests.cpp	100.00% <ø> (ø)
realtime_tools/test/realtime_publisher_tests.cpp	92.85% <ø> (ø)
...e_tools/test/realtime_server_goal_handle_tests.cpp	92.80% <ø> (ø)
...time_tools/test/realtime_thread_safe_box_tests.cpp	98.82% <100.00%> (+0.01%)	⬆️
...ealtime_tools/test/test_async_function_handler.cpp	100.00% <ø> (ø)
realtime_tools/test/thread_priority_tests.cpp	100.00% <ø> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[christophfroehlich]

we use gmock per default in all our repositories. I haven't compared it, but the argument of @bmagyar is that the failure messages are more descriptive.

[bjin2364]

the argument of @bmagyar is that the failure messages are more descriptive.

Interesting... for my own education, what would be an example of this case? My understanding is unless we're making calls to gmock-specific API, then the outcome is exactly the same as if we depended directly on gtest.

[christophfroehlich]

maybe this was only a decision for simplicity to have the same all over the place, and use gmock matchers where they make sense like testing::AllOf

[bjin2364]

From my experience so far, the issue with defaulting to gmock is:

• including gmock doesn't guarantee inclusion of all gtest features
  • example: https://github.com/ros-controls/realtime_tools/pull/459/files#diff-e1370ac741dbdaaaff7506efc1b7c8231b66a7c6c98704be1befad0158159416R31 explicitly needs gtest for the typed test features.
• the implicit gtest dependency from including gmock made the code more confusing to read

I think developers can always include gmock later when they need a feature from it, but I can close this PR if this goes against a preferred convention

[bmagyar]

Thanks for highlighting the fact that there are features from gtest that are not available when one includes the gmock headers. We've been using gmock across the board in our projects as we tend to rely on matchers.

That being said, I'd also prefer we stay one step closer to the advanced matcher functionality than typed tests, even though some of our testing don't use either of those 🤷