Edge to edge fix

[domonkosadam]

No description provided.

[github-actions]

🧪 Unit Test Results

---

📊 Summary

• Total Tests: 0
• Failed: 0
• Skipped: 0
• Status: ⚠️ No test results found

Last updated: Wed, 01 Apr 2026 11:53:02 GMT

[claude]

Review Summary

This PR swaps the IME (keyboard) inset handling in EdgeToEdgeScaffold from a contentWindowInsets-based approach to a Modifier.imePadding() approach. The intent is clear and the change is small, but it carries a meaningful behavioral difference worth documenting.

---

Issues

• Unused import (EdgeToEdgeScaffold.kt, line 19) — import androidx.compose.foundation.layout.WindowInsets is no longer referenced after removing the .union(WindowInsets.ime) call. The IDE/lint will flag this as an unused import.

• Behavioral change: imePadding() on the Scaffold modifier affects the entire Scaffold, not just the content area (EdgeToEdgeScaffold.kt, line 54)

  The two approaches behave differently:

  	Old (contentWindowInsets.union(WindowInsets.ime))	New (modifier.imePadding())
  Content bottom padding when keyboard open	Grows by IME height	Unchanged (Scaffold already shrunk)
  bottomBar position when keyboard open	Fixed — can be hidden behind keyboard	Floats above keyboard
  Top bar when keyboard open	Unaffected	Unaffected (correct)
  IME insets consumed?	No (only declared in contentWindowInsets)	Yes — inner imePadding() calls become no-ops

  If the motivation is to keep a non-empty bottomBar visible above the keyboard, the new approach achieves that correctly. But callers that expect paddingValues.calculateBottomPadding() (passed in the content lambda) to include IME height will need to be aware it no longer does — the Scaffold container itself has already shrunk.

  Also, since imePadding() consumes the IME insets, any composable inside content that also applies imePadding() will silently receive zero insets. This is correct (prevents double-padding) but can be surprising. A code comment noting this intent would help future maintainers.

---

Positive Notes

• The change is minimal and easy to reason about.
• Removing the union(WindowInsets.ime) from contentWindowInsets is consistent with the new modifier-based approach — no double-handling of IME insets within the Scaffold's content slot.
• Clean import cleanup (removing ime and union).

[github-actions]

📊 Code Coverage Report

✅ Student

• PR Coverage: 42.65%
• Master Coverage: 42.65%
• Delta: +0.00%

✅ Teacher

• PR Coverage: 25.37%
• Master Coverage: 25.37%
• Delta: +0.00%

✅ Pandautils

• PR Coverage: 23.67%
• Master Coverage: 23.67%
• Delta: +0.00%

---

📈 Overall Average

• PR Coverage: 30.56%
• Master Coverage: 30.56%
• Delta: +0.00%