Cellular: Refactor cellular variable visibilities

[kivaisan]

Summary of changes

Cellular: Refactor cellular variable visibilities

• Earlier some variables were public even though used only internally
• Also refactored variables to the end of class definitions
• Removed duplicate _property_array from CellularDevice
• Changed _impl methods as protected

Impact of changes

This can potentially be a breaking change but at least currently none of the existing mbed-os cellular targets required any changes.

Migration actions required

If customer wants to use some cellular variable marked as private, they need to inform @ARMmbed/mbed-os-wan and we can review the case.

Documentation

---

Pull request type

[] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[] Feature update (New feature / Functionality change / New API)
[X] Major update (Breaking change E.g. Return code change / API behaviour change)

---

Test results

[] No Tests required for this change (E.g docs only update)
[X] Covered by existing mbed-os tests (Greentea or Unittest)
[] Tests / results supplied as part of this PR

---

Reviewers

@ARMmbed/mbed-os-wan

---

[ciarmcom]

@kivaisan, thank you for your changes.
@ARMmbed/mbed-os-wan @ARMmbed/mbed-os-maintainers please review.

[adbridge]

@bulislaw please review

[adbridge]

@kivaisan What migration actions might people have to take ? Also presumably there will need to be some documentation changes to accompany this ? If so please comment accordingly in the documentation section.

[adbridge]

Looks reasonable to me

[adbridge]

Going to start ci while waiting for @bulislaw review

[mbed-ci]

Test run: FAILED

Summary: 1 of 11 test jobs failed
Build number : 1
Build artifacts

Failed test jobs:

• jenkins-ci/mbed-os-ci_greentea-test

[kivaisan]

@kivaisan What migration actions might people have to take ? Also presumably there will need to be some documentation changes to accompany this ? If so please comment accordingly in the documentation section.

Like mentioned in description, this didn't require any changes to any of our existing cellular targets. If someone uses variable which was originally intented to be internal but was incorrectly set as public earlier, they just need to inform us and we can evaluate whether tre visibility of the variable can be changed or is there something wrong with their targets design. This is how we've been handling these earlier.

[kivaisan]

CI failures does not seem to relate this PR. I rebased this PR with the latest mbed-os and at least NRF52840_DK + BG96 (cellular) and K64F (ethernet) DNS tests are OK.

So could you please restart the CI?

[mbed-ci]

Test run: FAILED

Summary: 1 of 4 test jobs failed
Build number : 2
Build artifacts

Failed test jobs:

• jenkins-ci/mbed-os-ci_unittests

[kivaisan]

I found this UT failure but this PR does not cause it. I also verified the branch locally and UTs are OK

13:09:08  /builds/ws/mbed-os-ci_unittests@3/unitTest-4755/mbed-os/UNITTESTS/events/equeue/test_equeue.cpp:680: Failure
13:09:08  Value of: touched > 1
13:09:08    Actual: false
13:09:08  Expected: true

[kjbracey]

Known failure - proposed fix is in #12126. Will retrigger.

[mbed-ci]

Test run: FAILED

Summary: 1 of 11 test jobs failed
Build number : 3
Build artifacts

Failed test jobs:

• jenkins-ci/mbed-os-ci_greentea-test

[kivaisan]

Again the latest failure is not this PR related.

[adbridge]

@kivaisan looks like we need a rebase

[kivaisan]

Rebased

[0xc0170]

CI started

[mbed-ci]

Test run: FAILED

Summary: 1 of 11 test jobs failed
Build number : 4
Build artifacts

Failed test jobs:

• jenkins-ci/mbed-os-ci_dynamic-memory-usage

[0xc0170]

I restarted dynamic memory usage. There was a failure in the filesystem one, will check other PRs and master