Refactor GfxReplayTest::testSimulatorIntegration

[0mdc]

Motivation and Context

The GfxReplayTest::testSimulatorIntegration test is currently using the same simulator instance to record and play a playback file. This doesn't model real use cases and may hide issues.

This changeset refactors the test so that separate simulator instances record and play.

How Has This Been Tested

Tests are successful locally.

Types of changes

• Docs change / refactoring / dependency upgrade
• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)

Checklist

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have completed my CLA (see CONTRIBUTING)
• I have added tests to cover my changes.
• All new and existing tests passed.

[aclegg3]

Refactor looks good to me, but curious about the log output here for CI testing.

@mosra I see the following in the CI log, but number 5 should be testSimulatorIntegration and instead displays as getNumberOfChildrenOfRoot which is a helper function called within the latter test. Is this a bug or am I missing something?

Test command: /home/circleci/project/habitat-sim/build/tests/GfxReplayTest
9: Environment variables: 
9:  HABITAT_SIM_LOG=quiet
9:  MAGNUM_LOG=QUIET
9: Test timeout computed to be: 10000000
9: Starting GfxReplayTest with 5 test cases...
9:     OK [1] testRecorder()
9:     OK [2] testPlayer()
9: [15:38:08:271680]:[Gfx] Player.cpp(37)::readKeyframesFromFile : File file_that_does_not_exist.json not found.
9:     OK [3] testPlayerReadMissingFile()
9: [15:38:08:271933]:[IO] Json.cpp(70)::parseJsonFile : Parse error reading /home/circleci/project/habitat-sim/src/../data/./gfx_replay_test.json Error code 4 at 1
9: [15:38:08:272044]:[Gfx] Player.cpp(44)::readKeyframesFromFile : Failed to parse keyframes from /home/circleci/project/habitat-sim/src/../data/./gfx_replay_test.json .
9:     OK [4] testPlayerReadInvalidFile()
9: [15:38:08:276886]:[Core] ManagedContainer.h(143)::registerObject : <Scene Instance> : No valid handle specified to register this Scene Instance managed object. Aborting.
9: [15:38:08:276960]:[Core] ManagedContainer.h(143)::registerObject : <Scene Instance> : No valid handle specified to register this Scene Instance managed object. Aborting.
9:     OK [5] getNumberOfChildrenOfRoot()
9: Finished GfxReplayTest with 0 errors out of 79 checks.
 9/24 Test  #9: GfxReplayTest ....................   Passed    0.32 sec

[mosra]

Not a bug, but a consequence of how the test function name is captured via the CORRADE_VERIFY() / CORRADE_COMPARE() macros -- the first time the macro gets used in a test case, it captures the name of the surrounding function. You have two options:

1. Remove the CORRADE_VERIFY(lastRootChild); from getNumberOfChildrenOfRoot(), or change it to e.g. CORRADE_INTERNAL_ASSERT(lastRootChild). That sounds like the more robust solution to me -- the helper doesn't really verify anything, just returns a value (that gets later verified), so it shouldn't need to use the verification macros in my opinion.

2. Make sure some macro gets called from within testSimulatorIntegration() before getNumberOfChildrenOfRoot() is called, or just add something like

   CORRADE_VERIFY(true); /* capture correct function name */

   at the very top of the testSimulatorIntegration() test case. This is rather prone to breaking again in the future, tho, especially when getNumberOfChildrenOfRoot() gets used in more test cases.

[aclegg3]

Looks good, let's merge it.