fix: merge main into workstation/swift-migration and apply pending test fixes

[nickolas-dimitrakas]

Draft. This is now a real git merge main into workstation/swift-migration, not a hand-authored commit reproducing main's changes. Previous revision of this PR did the latter; this replaces it.

Why a merge instead of a port

workstation/swift-migration diverged from main before #867 and #870 landed, so this branch never got the CI hang fixes or the workspace-switch test fixes. The first version of this PR re-implemented that content as a fresh commit. That works for a one-time sync, but it leaves no shared history — a future real merge of main into this branch would see none of it as already-applied, and would either duplicate the work or conflict against it for no reason.

This version merges main's actual tip into this branch instead, so #867, #870, and everything else main has picked up since (CODEOWNERS additions, an AGENTS.md trim, dependency bumps) land with their real commit history intact.

What merged cleanly, and what didn't - and a correction on attribution

Everything auto-merged except UnitTests/ObjCTests/MParticleTests.m, where both branches had independently rewritten the same five workspace-switch test bodies. Correction to an earlier version of this description: only one of those five, testSwitchWorkspaceKitsWithStop, was actually converted from dispatch_after to MPWaitForCondition by main's #870. The other four are converted in commit 35fe1c82 on the still-open #878/#916 stack, not on main. I said "main removes dispatch_after for MPWaitForCondition" for the whole block; that overstated it for four of the five methods, and I want the record accurate rather than let it stand.

Git's line-based merge did not raise a conflict marker on this file, but it silently interleaved hunks from the different rewrites each side had made: 10 dispatch_after calls survived alongside 3 MPWaitForCondition calls, split unevenly across the five methods. A clean merge report is not proof the result is correct - I checked, and this one wasn't. Rather than reconcile only the one method main had actually converted and leave the other four on old-style fixed delays, I carried over the same conversion from #878/#916 for all five, since this branch needs the same fix and the merge conflict was the natural point to apply it. Assertions are this branch's own (resetRegistry, kitContainer_PRIVATE, registeredKits.anyObject) throughout - only the wait mechanism changed. The one actual conflict marker, in the MParticle () class-extension additions, was just two property declarations that both belonged (kitContainer_PRIVATE from this branch, initialized from main).

testActiveKitsRegistryThreadSafety in MPKitContainerTests.m merged without any of this trouble - main's #870 change there only reworked how the test tracks failures internally (an NSLock-guarded exception instead of an unsynchronized __block BOOL), in lines this branch hadn't independently touched.

MPRoktTests.m diverged too far structurally for the merge to locate matching context at all - main's changes there (async timeout raises, teardown drain, a reject-site fix, the mock teardown ordering) simply didn't apply. The teardown-ordering piece is re-applied directly in the second commit here, since it's correct regardless of file structure; the rest doesn't apply to this branch's version of the file and isn't part of this PR.

A known weak wait, tracked upstream rather than fixed here

Cursor flagged that testSwitchWorkspaceOptions and testSwitchWorkspaceSideloadedKits wait on signals that fire before startup actually finishes (instance.options != nil is set synchronously near the top of startWithOptions:, well before the async identify call and configureWithOptions: that initialized waits for). Confirmed by reading mParticle.m directly - it's a real gap, not a false positive.

Both tests came from #878/#916 (see the correction above), which has the identical wait and the identical gap. Fixing it here would diverge from that PR rather than match it, and the next real merge would just reintroduce the weaker version. Left as a follow-up on #878 - once fixed there, it reaches this branch on the next sync.

Deliberately still not ported

The kitsRegistry lock fix does not apply here. The migration moved the registry into MPKitContainer.swift, where registryLock and registry are both static and every accessor takes the lock already - this branch's registry locking is sounder than main's pre-fix version. Main's separate wrapperInstance race fix (#916) is the same story: it's fixing a nonatomic property race in MPKitContainerExecutionAdapter.m's Objective-C teardown path, and this branch's Swift-side locking doesn't have the same shape of problem. Neither is in scope here.

Validation

• No duplicate method names introduced by the merge (checked directly, since that's the failure mode a bad Objective-C merge would produce).
• build-for-testing succeeds.
• MParticleTests + MPKitContainerTests: 147/147 passing, including all five workspace-switch tests (0.1-0.4s each, versus 20-30s of fixed delay before) and the thread-safety stress test.
• trunk check clean.
• Local runs, Xcode 26.3 / iOS 26 simulator. CI not yet re-run against this revision.

🤖 Generated with Claude Code

[github-actions]

🐦 Swift Migration Progress

Production implementation code at 6322ca6d47fa compared with edc6f6c82bcd.

Area	Goal	Progress	Base	This PR	Swift SLOC	Objective-C remaining	Change
Core SDK	Short term — in scope	████▌░░░░░	44.61%	44.61%	6,630	8,232	➖ 0.00 pp
Core SDK	Long term — all Objective-C	███▎░░░░░░	32.09%	32.09%	6,630	14,033	➖ 0.00 pp
SDK kit infrastructure	Short term — in scope	██████░░░░	60.28%	60.28%	1,645	1,084	➖ 0.00 pp
SDK kit infrastructure	Long term — all Objective-C	███▍░░░░░░	34.33%	34.33%	1,645	3,147	➖ 0.00 pp
Standalone kits	Short term — in scope	▏░░░░░░░░░	1.06%	1.06%	152	14,218	➖ 0.00 pp
Standalone kits	Long term — all Objective-C	▏░░░░░░░░░	1.06%	1.06%	152	14,218	➖ 0.00 pp

Objective-C retained by design: Core SDK 5,801 · SDK kit infrastructure 2,063 · Standalone kits 0.

This PR's code movement

Area	Swift lines added	Objective-C lines removed
Core SDK	0	0
SDK kit infrastructure	0	0
Standalone kits	0	0

How this is measured

• Current composition uses production source lines of code (SLOC) from cloc; comments and blank lines are excluded.
• Short term — in scope excludes the Objective-C the migration will not delete, so 100% is the end of this project: every in-scope implementation gone.
• Long term — all Objective-C keeps the full denominator. Reaching 100% there means the public API itself becomes Swift, which is a breaking change reserved for a future major release.
• The gap between the two rows is the retained public/kit contract, runtime-identity, and boundary-glue surface listed in Tools/swift-migration-retained-objc.txt.
• Retained wrappers keep their Objective-C interface but still shed logic to Swift. That thinning moves the long-term row and the retained figure, not the short-term row.
• Both revisions are measured with the manifest from the head revision, so a manifest edit does not by itself move the reported change. A retained file this pull request renamed or deleted still counts as retained at the base.
• Pull request movement uses physical additions/deletions from git diff base...head --numstat; it counts retained files too and is intentionally separate from SLOC totals.
• Core excludes SDK kit infrastructure and vendored libraries. Standalone kits include only files below Kits/**/Sources.
• Tests, examples, headers, build outputs, vendored libraries, and the MParticle/Sources Swift overlay are excluded.
• Objective-C++ (.mm) is included in the Objective-C figures and removed counts.

Generated with cloc 2.10. This report is informational and does not gate migration direction.

[cursor]

PR Summary

Low Risk
Changes are CI, documentation, lockfile, and test harness only; no production SDK behavior is modified.

Overview
Ports CI reliability fixes and ObjC unit-test timing/thread-safety improvements onto the swift-migration branch, plus slimmer agent docs and per-kit CODEOWNERS.

CI: native-tests waits for simulator boot (simctl bootstatus), turns off parallel test clones, and enables per-test timeouts. build-kits adds job/step timeouts and replaces bare SPM resolve with git low-speed settings, an 8-minute watchdog, child-process cleanup, and two retry attempts. Java setup moves to actions/setup-java@v6 in cross-platform and integration workflows.

Tests: Workspace-switch cases in MParticleTests drop nested dispatch_after delays in favor of MPWaitForCondition (main-run-loop polling until initialized, kit registry state, or a new shared instance). MPRoktTests stops the partial mock before releasing the kit under test. The testActiveKitsRegistryThreadSafety stress path records failures under an NSLock and asserts on the main queue instead of calling XCTest APIs from worker threads.

Repo hygiene: AGENTS.md is rewritten as a short, repo-specific agent guide (commands, traps, PR/CI gotchas). Each kit under Kits/ gets a .github/CODEOWNERS (Rokt SDK+ → @ROKT/sdk-engineering). RNExample/package-lock.json picks up routine transitive npm bumps.

^{Reviewed by Cursor Bugbot for commit 6322ca6. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

📦 SDK Size Impact Report

Measures how much the SDK adds to an app's size (with-SDK minus without-SDK).

Metric	Target Branch	This PR	Change
App Bundle Impact	2.38 MB	2.38 MB	+N/A
Executable Impact	848 bytes	848 bytes	+N/A
XCFramework Size	6.84 MB	6.84 MB	+N/A

➡️ SDK size impact change is minimal.

Raw measurements

Target branch (workstation/swift-migration):

{"baseline_app_size_kb":84,"baseline_executable_size_bytes":75464,"with_sdk_app_size_kb":2520,"with_sdk_executable_size_bytes":76312,"sdk_impact_kb":2436,"sdk_executable_impact_bytes":848,"xcframework_size_kb":7000}

This PR:

{"baseline_app_size_kb":84,"baseline_executable_size_bytes":75464,"with_sdk_app_size_kb":2520,"with_sdk_executable_size_bytes":76312,"sdk_impact_kb":2436,"sdk_executable_impact_bytes":848,"xcframework_size_kb":7000}

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, have a team admin enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 6322ca6. Configure here.}