THROWAWAY: does the fix survive OverlayBackdrop once the panel is fully open #174
Loading…
Reference in a new issue
No description provided.
Delete branch "chore/task-233-backdrop-experiment"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Not for merging. Reviewer of PR #173 correctly flagged that OverlayBackdrop (position: fixed; inset: 0; z-index 1299; pointer-events: auto once visible) covers the whole page above the map once the add-place panel is open, and my SecondClickX=450 fix only avoided the panel's own footprint, not the backdrop's. This branch adds one scratch test that waits 600ms (double the 300ms deferral window) before the second click, so the panel and its backdrop are unambiguously, fully open, then clicks the map at x=450. Will be closed once CI reports.
Confirmed: even 600ms after the first click (double the 300ms deferral window), a click on the page's own map at x=450 is intercepted by
div.overlay-backdrop(overlay-backdrop-visible), not the panel's own content.OverlayBackdropisposition: fixed; inset: 0withpointer-events: autoonce visible, so it covers the entire viewport the instant the panel opens — no X/Y on the page map is safe once that happens. This means PR #173's coordinate change has no effect on the actual failure rate; it only relocates which element is named as the interceptor. Closing — not for merging; deleting the branch now.Closing — this branch's purpose is served.
It was a throwaway experiment for task 233, forcing the second map click 600ms after the first (double the deferral window, panel unambiguously open) to test whether moving
SecondClickXto 450 genuinely removed the click-vs-panel race.It disproved the fix. The click is still intercepted, now by
div.overlay-backdrop(position: fixed; inset: 0; z-index: 1299; pointer-events: auto) rather than by the panel's own content. Since the backdrop covers the whole viewport once the panel is open, no coordinate change can help — it only relocates which element is named in the failure.That result is recorded on PR #173 (parked) and on Vikunja task 233. Branch deleted.
Pull request closed