The empty-orders branch ("Nenhum pedido encontrado.") had no
.frame(maxWidth: .infinity), unlike the loading/error branches -
ScrollView sizes to its content's intrinsic width, so the whole
container (and its background) shrank to the text's width, leaving
the real black window visible on both sides. Forces the VStack (and
ScrollView) to always fill width regardless of which branch renders.
Cascades .buttonStyle(.plain) down from each LCENavigationView
instance instead of waiting on the LCEssentials package fix
(committed separately, unpublished: LC_Essentials commit 8b0f650) to
be released and this repo's Package.resolved bumped. Overrides
iOS 26's automatic Liquid Glass chrome on LCENavigationView's
internal nav buttons - including the invisible backfilled mirror
button on whichever side has no real action - without affecting any
button that already sets its own explicit style closer to itself.
setNavigationBarBackgroundColor was never called on the 15 screens
adopted in 3456184, so it stayed at the default .clear and the real
window background (black) showed through. Set it to
AppColors.backgroundLight on all of them, matching each screen's own
actual content background. Also wraps CartView's title in
LCENavigationView for consistency with the rest of the app - no
setLeftButton since Cart is a tab root, no back button needed.
The other reported bug (empty right-button slot rendering a visible
glass circle) is fixed in LCEssentials itself (separate repo,
commit 8b0f650 on fix/navigation/liquid-glass-button-chrome) but
won't take visual effect here until that's published and this
repo's Package.resolved is bumped to point at it.
Standardizes the back button to a single shared component
(AppBackButtonIcon, 32x32/14pt) instead of ~16 hand-duplicated
52x52/24pt copies, and wires each screen through LCENavigationView
per this project's mandatory-usage rule. Home and Store Detail
intentionally excluded - both have bespoke header designs (collapsing
header, hero-image overlay) this change would visually disrupt.
ProfileLoggedOutView conflict: kept this branch's simpler
enterAuth()-closure version over main's inline-embedded LoginView
(fix/profile/logged-out-auth-intro, predates the login-entry-
animation-bug fix). That bug was the actual reason the separate-
screen approach looked broken; it's already fixed here
(decisions/2026-08-06-login-entry-animation-bug.md), so the simpler
pattern works without re-threading a root binding through
MainTabView/ProfileLoggedOutView, which this branch deliberately
removed in favor of the closure. Ported the location-usage-
description Portuguese copy and the entitlements additions
(aps-environment, location.push) from main.
Resolves the App Attest env/entitlements changes onto the flattened
PediFoods.xcodeproj structure (Darwin/PediFoods.xcodeproj no longer
exists on this branch). Also restores Release signing to Manual /
Apple Distribution / AppStore provisioning profile - the merged-in
fix had switched it to Automatic / Apple Development, which breaks
headless CI archiving.
FiltersModalView.swift was 0% covered. Reaching it via guest mode hit
the backend's App Attest simulator-bypass rejection (403
APP_ATTEST_VERIFICATION_FAILED on /api/public/session, confirmed via
curl - a pre-existing, out-of-repo backend issue already documented in
decisions/2026-08-06-ui-test-account-and-app-attest-bypass.md), which
kept the guest address picker stuck showing a load-error sub-screen.
While chasing that, found a real bug in ContentView.swift:
.sheet(item: $appState.activeModal) had no onDismiss, so an
interactive swipe-to-dismiss never reset the bound item to nil -
anyone who swipes the address picker away without picking a location
leaves appState.activeModal stuck non-nil, silently breaking the next
modal presentation app-wide. Fixed with
onDismiss: { appState.activeModal = nil }.
Rewrote HomeFiltersFlowTests to use the authenticated QA account
instead of guest mode, routing around the backend-blocked guest wall
entirely. Also fixed a separate, real XCUITest issue: a plain .tap()
on the filter icon (inside Home's .offset()-transformed collapsing
header) reliably produced "Computed hit point {-1, -1}" - fixed via
coordinate(withNormalizedOffset:).tap().
Verified: HomeFiltersFlowTests passed clean across 2 consecutive
isolated runs. Full-suite regression run could not be completed this
round - the local environment stopped completing any instrumented
test launch afterward (reproduced across background/foreground,
clean DerivedData, simulator reboot, decoupled build/test, process
cleanup), unrelated to these changes. Documented in
decisions/2026-08-11-coverage-push-to-85-percent-status.md, with a
next-session TODO to run the full suite once the environment recovers.
