Commit 5794d14
committed
fix(tater): actually put sprite behind the plan
The previous commit diagnosed this wrong. The real regression came from
PR #509 (Zed-style overlay scrollbars), which wraps <main> in
overlayscrollbars. The library's CSS applies
`position: relative; z-index: 0` to its `[data-overlayscrollbars-padding]`
and `[data-overlayscrollbars-viewport]` wrappers (see
overlayscrollbars.css:206/214), which creates a stacking context INSIDE
<main>. That traps the plan's `z-10`/`z-50` inside it: from the outer
content wrapper's perspective, the entire plan subtree paints at
effective `z=0`, so the tater at `z-5` ends up on top of it.
Bumping planAreaRef doesn't help — it's already trapped inside the
library's z-0 stacking context, so its value only changes layering
WITHIN that bubble. What has to change is the layer the whole bubble
paints at in the outer context, i.e. the <main> host itself.
Give the OverlayScrollArea <main> `relative z-20`, which makes it a
stacking context at z=20 in the outer wrapper. Keep the sprite at z-15
so it still paints above the z-10 sidebar rail / resize handles but
below <main>. Revert planAreaRef back to `relative z-10` — its bump
was a no-op given the trap.
https://claude.ai/code/session_019Q3pp1qUN52e76PqpHY6mg1 parent 9e5cb46 commit 5794d14
2 files changed
Lines changed: 3 additions & 3 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
1525 | 1525 | | |
1526 | 1526 | | |
1527 | 1527 | | |
1528 | | - | |
| 1528 | + | |
1529 | 1529 | | |
1530 | 1530 | | |
1531 | 1531 | | |
| |||
1539 | 1539 | | |
1540 | 1540 | | |
1541 | 1541 | | |
1542 | | - | |
| 1542 | + | |
1543 | 1543 | | |
1544 | 1544 | | |
1545 | 1545 | | |
| |||
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
25 | 25 | | |
26 | 26 | | |
27 | 27 | | |
28 | | - | |
| 28 | + | |
29 | 29 | | |
30 | 30 | | |
31 | 31 | | |
| |||
0 commit comments