Purpose

If a push message triggers a sync for an account that already has an ongoing sync, there's a risk of losing remote changes made near the end of the existing sync. To address this, we can always append an additional sync worker when triggered by a push request, but restrict it to at most one pending/blocked worker.

Short description

When sync request comes from push:

• use synchronized block to avoid race conditions for the check, that ...
• ... there is not already blocked or enqueued work for same unique work (given pair of account + data type)
• if there is not, then enqueue with ExistingWorkPolicy.APPEND_OR_REPLACE
• access enqueue operation result object (listenable future), to force waiting until work is enqueued

Checklist

• The PR has a proper title, description and label.
• I have self-reviewed the PR.
• I have added documentation to complex functions and functions that can be used by other modules.
• I have added reasonable tests or consciously decided to not add tests.