[camera_android] Avoid thread race conditions when closing the camera.

[davidmartos96]

Fixes the second race condition described on flutter/flutter#165092

Pre-Review Checklist

• I read the Contributor Guide and followed the process outlined there for submitting PRs.
• I read the Tree Hygiene page, which explains my responsibilities.
• I read and followed the relevant style guides and ran the auto-formatter.
• I signed the CLA.
• The title of the PR starts with the name of the package surrounded by square brackets, e.g. [shared_preferences]
• I linked to at least one issue that this PR fixes in the description above.
• I updated pubspec.yaml with an appropriate new version according to the pub versioning philosophy, or I have commented below to indicate which version change exemption this PR falls under¹.
• I updated CHANGELOG.md to add a description of the change, following repository CHANGELOG style, or I have commented below to indicate which CHANGELOG exemption this PR falls under¹.
• I updated/added any relevant documentation (doc comments with ///).
• I added new tests to check the change I am making, or I have commented below to indicate which test exemption this PR falls under¹.
• All existing and new tests are passing.

If you need help, consider asking for advice on the #hackers-new channel on Discord.

cc @camsim99

Footnotes

1. Regular contributors who have demonstrated familiarity with the repository guidelines only need to comment if the PR is not auto-exempted by repo tooling. ↩ ↩² ↩³

[flutter-dashboard]

It looks like this pull request may not have tests. Please make sure to add tests or get an explicit test exemption before merging.

If you are not sure if you need tests, consider this rule of thumb: the purpose of a test is to make sure someone doesn't accidentally revert the fix. Ask yourself, is there anything in your PR that you feel it is important we not accidentally revert back to how it was before your fix?

Reviewers: Read the Tree Hygiene page and make sure this patch meets those guidelines before LGTMing.If you believe this PR qualifies for a test exemption, contact "@test-exemption-reviewer" in the #hackers channel in Discord (don't just cc them here, they won't see it!). The test exemption team is a small volunteer group, so all reviewers should feel empowered to ask for tests, without delegating that responsibility entirely to the test exemption group.

[camsim99]

LGTM!

[stuartmorgan-g]

Per the bot comment, this would need tests.

@camsim99 please make sure to check for sufficient testing as part of code review.

[stuartmorgan-g]

How does this fix the race condition? If I'm understanding correctly, the premise is that we are calling this method simultaneously from two different threads. If that's true, then this version will call close twice on the same session, which will almost certainly just replace a null pointer exception with state error exception.

If this class is trying to interact with the same objects from multiple threads at the same time, that needs to actually be fixed in some robust way (most likely either locking or cross-thread redispatch). Changing individual variable access is almost never the right way to handle threading issues, and that seems to apply here as well.

[davidmartos96]

@stuartmorgan-g Thank you for reviewing.

I also thought about it possibly being replaced by a State error from the SDK, but after implementing the fix on our production app with heavy camera usage, the NullPointerException that was filling up our Crashlytics events stopped appearing and it didn't get replaced by a different error.

I'm not really sure how to really test this, because of its threading nature. I agree a more thread safe implementation could be made. But the original idea of this PR was a simple way to stop obtaining the NullPointerException on the "legacy" camera_android package, which I assume will be replaced by the camerax one at some point in the future.

[stuartmorgan-g]

I also thought about it possibly being replaced by a State error from the SDK, but after implementing the fix on our production app with heavy camera usage, the NullPointerException that was filling up our Crashlytics events stopped appearing and it didn't get replaced by a different error.

Apparently close is special-cased.

However, if close is being called on the wrong thread in some cases, I don't see how we have any guarantee that the close call couldn't be interleaved with other non-close calls on the main thread, so it still seems like this is papering over one symptom of a threading bug rather than fixing the threading bug. Do we have some guarantee that this case is unique?

I'm not really sure how to really test this, because of its threading nature.

It depends on what the fix ultimately is. For instance, if we fix it by ensuring that all access happens on a single thread, then it's testable by synthetically calling the offending callback from a different thread in a test, and ensuring that the (mock) camera object is still called on the main thread. We have tests like that in several plugins that receives callbacks on background threads.

I agree a more thread safe implementation could be made. But the original idea of this PR was a simple way to stop obtaining the NullPointerException

When we have evidence of a bug (a threading issue, in this case), the goal should be fixing the bug, rather than hiding the symptom that alerted us to the bug.

on the "legacy" camera_android package, which I assume will be replaced by the camerax one at some point in the future.

I'm not sure I follow. Is the argument that it doesn't matter if this is fixed correctly because the primary development focus is on the CameraX version? If this is issue is worth making, reviewing, and landing a PR for, then it matters enough for our usual development standards to apply.