[fadi-george]

Description

One Line Summary

Experiment: send sessions API requests on their own small queue instead of through OperationRepo.

Details

Motivation

Compare against the #2777 / #2780 / #2781 stack. A failing sessions request in OperationRepo blocks user and subscription operations, and the stack needs about 300 lines of shared OperationRepo changes. This keeps sessions fully separate.

Size (excluding #2774 / #2775):

	Code	Tests	OperationRepo changes
Stack	+734 / -90	+1234 / -136	+300
This PR	+553 / -79	+484 / -102	0

Scope

• One persisted record per session: the create, plus the latest unsent update. A newer update replaces an unsent one, and an end is final.
• Idempotency keys are reused on retry, including after a restart.
• Backoff 5s doubling to 300s with jitter. Retry-After is honored (capped at 1h). A request is dropped after 5 answered failures or a non-retryable 4xx. Network errors don't count.
• Retries on foreground, and on network available while foregrounded.
• Waits for backend user and subscription IDs. Records with a local ID that is no longer current are dropped. This replaces the pinned IDs on SessionModel.
• Ending a session in the background job sends the end right away, limited to 30s.
• Behind the sessions v2 flag. No change when it is off.

OPTIONAL - Other

• Includes feat: [SDK-5369] session model and sessions v2 feature flag #2774 and feat: [SDK-5374] add sessions API backend client #2775 as base commits, since they only exist on new-session-events.
• The KMP submodule points at KMP new-session-events for the flag, so this can't merge until that is on KMP main.
• Not included: direct_attribution_id, periodic duration updates.

Testing

Unit testing

• SessionsApiQueueTests: merging, retries and keys, drop rules, backoff sequence, Retry-After, cap, local ID handling.
• SessionServiceTests: create, background end + flush, end time after kill, flag off, cold start ordering.
• Full core suite, spotless, detekt and assembleRelease pass.

Manual testing

Not tested on a device yet.

Affected code checklist

• Notifications
  • Display
  • Open
  • Push Processing
  • Confirm Deliveries
• Outcomes
• Sessions
• In-App Messaging
• REST API requests
• Public API changes

Checklist

Overview

• I have filled out all REQUIRED sections above
• PR does one thing
• Any Public API changes are explained in the PR details and conform to existing APIs

Testing

• I have included test coverage for these changes, or explained why they are not needed
• All automated tests pass, or I explained why that is not possible
• I have personally tested this on my device, or explained why that is not possible

Final pass

• Code is as readable as possible.
• I have reviewed this PR myself, ensuring it meets each checklist item

🤖 Generated with Claude Code

[github-actions]

📊 Diff Coverage Report

Diff Coverage Report (Changed Lines Only)

Gate: aggregate coverage on changed executable lines must be ≥ 80% (JaCoCo line data for lines touched in the diff).

Changed Files Coverage

• Time.kt: 1/1 touched executable lines (100.0%) (2 touched lines in diff)
• OneSignalImp.kt: 1/1 touched executable lines (100.0%) (2 touched lines in diff)
• ⚠️ SessionModule.kt: Not in coverage report (may not be compiled/tested)
• SessionModel.kt: 9/9 touched executable lines (100.0%) (32 touched lines in diff)
• ISessionsBackendService.kt: 15/15 touched executable lines (100.0%) (48 touched lines in diff)
• SessionsBackendService.kt: 35/35 touched executable lines (100.0%) (71 touched lines in diff)
• SessionService.kt: 38/39 touched executable lines (97.4%) (70 touched lines in diff)
• ISessionsApiQueue.kt: 2/2 touched executable lines (100.0%) (29 touched lines in diff)
• SessionRecord.kt: 24/24 touched executable lines (100.0%) (63 touched lines in diff)
• SessionsApiBackoff.kt: 16/16 touched executable lines (100.0%) (55 touched lines in diff)
• SessionsApiQueue.kt: 81/84 touched executable lines (96.4%) (177 touched lines in diff)
• SessionsApiRetryTrigger.kt: 15/18 touched executable lines (83.3%) (43 touched lines in diff)
• SessionsApiSender.kt: 69/71 touched executable lines (97.2%) (139 touched lines in diff)
• AppIdResolution.kt: 5/5 touched executable lines (100.0%) (15 touched lines in diff)
• ⚠️ MockHelper.kt: Not in coverage report (may not be compiled/tested)

Overall (aggregate gate)

311/320 touched executable lines covered (97.2% — requires ≥ 80%)

📥 View workflow run

[fadi-george]

Closing in favor of fixing OperationRepo retry for all ops (SDK-5535).