[ps2]

Bumps Loop and LoopKit on next-dev for the new Settings › Display screen.

• Chart now line, fixed glucose scale and label spacing LoopKit#667: chart support for a now line, a fixed glucose scale and x-axis label spacing (all off by default).
• Settings › Display: now line, glucose scale and time window Loop#2613: Settings › Display with:
  • Now Line: a vertical line at the current time on each chart.
  • Glucose Scale: Auto, or Fixed 40–400 mg/dL (2–22 mmol/L).
  • Time Window: Full Forecast (today's behavior), 3 Hours, 6 Hours or 12 Hours. On wider screens the hour options keep a phone's zoom and show more history, up to 2×.

Defaults match today's behavior. Built with the LoopWorkspace scheme on Xcode 27.1 for the simulator (19 plugins).

[marionbarker]

Test

✅ Successful test of feature
❌ Failed tests

• multiple tests failed in class LoopDataManagerTests

Configuration

Configure LoopWorspace for bump/next-dev-display-options.

Test Details

Build onto SE 3rd gen running iOS 27.0.1.

• pump is simulator
• cgm is Nightscout

Settings has new Display Row.

• test enable now line - works nicely
• test fixed 40 to 400 scaling
• test the various history/forecast display selections

[ps2]

CircleCI has failed twice on this PR, each time in different tests, all timing-sensitive and outside what this bump touches:

• 1310: CriticalEventLogExportManagerTests.testExport (10 s timeout) and testExportPartial (temp file missing)
• 1311: CarbStoreHKQueryTests.testHKQueryAnchorPersistence (3 s timeout) and GlucoseStoreTestSharingUndetermined.testHealthKitQueryAnchorPersistence (Core Data store not open); the hosted test app also restarted partway through

The tests that failed in one run passed in the other. The bump only changes chart code, PredictedGlucoseChartTests, the status screen, Settings and LoopConstants.

Local run of the same command CI uses (xcodebuild -scheme LoopWorkspace ... test) at bd69f5d on a fresh iPhone 17 simulator (iOS 27.0): ** TEST SUCCEEDED **, all bundles passed. The four tests above each took 0.03–0.05 s.

My guess is that a slow or overloaded CI machine is hitting the timeouts and setup races in these tests, and the bump isn't the cause.