Commit 5e83c60
committed
#20259 - VirtualizingStackPanel: window-independent extent, opt-in container virtualization
Three separable things, all in VirtualizingStackPanel and the ItemsControl
plumbing behind it.
1. The extent estimate no longer depends on which items happen to be realized.
A persistent per-item size record replaces the average over the realized
window, so revisiting an offset reports the same extent, and once every item
has been measured the extent is the exact total. The running sum is maintained
incrementally at the single upsert site with Neumaier compensation, so a
measure pass stays O(realized window) as in stock. Every fork-added tuning
constant that used to damp the oscillation goes with it: EMA smoothing, extent
freezing and boundary clamping, the layout cycle breaker, estimate caching, the
sub-pixel resize threshold, warmup head sampling.
2. Reset preservation now requires every realized element to still validate,
rather than a bare majority. A mid-list insert or remove coalesced into a
single Reset used to preserve a stale mapping, so shifted items rendered under
the wrong containers.
3. Container-level virtualization, opt-in per template. A template that hands
out a recycle key through IVirtualizingDataTemplate, which means
EnableVirtualization in XAML or RecycleKeySelector on a FuncDataTemplate, has
its container and its child recycled as one unit instead of the child subtree
being torn down and rebuilt on every scroll. Over heterogeneous complex rows
that is ~3.4x faster scrolling and 83% less allocated, at identical container
prepare counts. A template that does not opt in keeps stock behaviour, and
ContainerVirtualization.IsEnabled is a kill switch rather than the opt-in.
Supporting changes: constant-free anchor compensation in ValidateStartU,
RetainMatchingContainers for disjunct scrolls, opt-in warmup that grows the
pool off encountered keys, a guard against 0x0 effective viewports, and
ContentPresenter batch updates so Content and ContentTemplate land together.
VirtualizingPanel.Refresh becomes virtual so a template or theme change reaches
already-realized containers.
Note that an opted-in container is hidden and kept in Panel.Children rather
than torn down, so Loaded/Unloaded and AttachedToVisualTree/
DetachedFromVisualTree fire once for the first item a container displayed and
not per item. Templates that need per-item work should use DataContextChanged.
Tests: VirtualizingStackPanelTests grows to 153 methods over 310 cases, with
new ContainerVirtualizationTests and ContentPresenterTests_BatchUpdate.
Benchmarks in tests/Avalonia.Benchmarks/Controls cover scroll and jump
timings, deterministic container counts, and what the size record costs in
memory and per pass. Sample: ListBoxComplexLayoutPage in ControlCatalog.1 parent 6ef2d0b commit 5e83c60
29 files changed
Lines changed: 10623 additions & 1130 deletions
File tree
- samples/ControlCatalog
- Converter
- Pages
- ViewModels
- src
- Avalonia.Controls
- Presenters
- Templates
- Utils
- Markup/Avalonia.Markup.Xaml/Templates
- tests
- Avalonia.Benchmarks
- Controls
- Avalonia.Controls.UnitTests
- Presenters
Lines changed: 178 additions & 0 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
| 1 | + | |
| 2 | + | |
| 3 | + | |
| 4 | + | |
| 5 | + | |
| 6 | + | |
| 7 | + | |
| 8 | + | |
| 9 | + | |
| 10 | + | |
| 11 | + | |
| 12 | + | |
| 13 | + | |
| 14 | + | |
| 15 | + | |
| 16 | + | |
| 17 | + | |
| 18 | + | |
| 19 | + | |
| 20 | + | |
| 21 | + | |
| 22 | + | |
| 23 | + | |
| 24 | + | |
| 25 | + | |
| 26 | + | |
| 27 | + | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
| 33 | + | |
| 34 | + | |
| 35 | + | |
| 36 | + | |
| 37 | + | |
| 38 | + | |
| 39 | + | |
| 40 | + | |
| 41 | + | |
| 42 | + | |
| 43 | + | |
| 44 | + | |
| 45 | + | |
| 46 | + | |
| 47 | + | |
| 48 | + | |
| 49 | + | |
| 50 | + | |
| 51 | + | |
| 52 | + | |
| 53 | + | |
| 54 | + | |
| 55 | + | |
| 56 | + | |
| 57 | + | |
| 58 | + | |
| 59 | + | |
| 60 | + | |
| 61 | + | |
| 62 | + | |
| 63 | + | |
| 64 | + | |
| 65 | + | |
| 66 | + | |
| 67 | + | |
| 68 | + | |
| 69 | + | |
| 70 | + | |
| 71 | + | |
| 72 | + | |
| 73 | + | |
| 74 | + | |
| 75 | + | |
| 76 | + | |
| 77 | + | |
| 78 | + | |
| 79 | + | |
| 80 | + | |
| 81 | + | |
| 82 | + | |
| 83 | + | |
| 84 | + | |
| 85 | + | |
| 86 | + | |
| 87 | + | |
| 88 | + | |
| 89 | + | |
| 90 | + | |
| 91 | + | |
| 92 | + | |
| 93 | + | |
| 94 | + | |
| 95 | + | |
| 96 | + | |
| 97 | + | |
| 98 | + | |
| 99 | + | |
| 100 | + | |
| 101 | + | |
| 102 | + | |
| 103 | + | |
| 104 | + | |
| 105 | + | |
| 106 | + | |
| 107 | + | |
| 108 | + | |
| 109 | + | |
| 110 | + | |
| 111 | + | |
| 112 | + | |
| 113 | + | |
| 114 | + | |
| 115 | + | |
| 116 | + | |
| 117 | + | |
| 118 | + | |
| 119 | + | |
| 120 | + | |
| 121 | + | |
| 122 | + | |
| 123 | + | |
| 124 | + | |
| 125 | + | |
| 126 | + | |
| 127 | + | |
| 128 | + | |
| 129 | + | |
| 130 | + | |
| 131 | + | |
| 132 | + | |
| 133 | + | |
| 134 | + | |
| 135 | + | |
| 136 | + | |
| 137 | + | |
| 138 | + | |
| 139 | + | |
| 140 | + | |
| 141 | + | |
| 142 | + | |
| 143 | + | |
| 144 | + | |
| 145 | + | |
| 146 | + | |
| 147 | + | |
| 148 | + | |
| 149 | + | |
| 150 | + | |
| 151 | + | |
| 152 | + | |
| 153 | + | |
| 154 | + | |
| 155 | + | |
| 156 | + | |
| 157 | + | |
| 158 | + | |
| 159 | + | |
| 160 | + | |
| 161 | + | |
| 162 | + | |
| 163 | + | |
| 164 | + | |
| 165 | + | |
| 166 | + | |
| 167 | + | |
| 168 | + | |
| 169 | + | |
| 170 | + | |
| 171 | + | |
| 172 | + | |
| 173 | + | |
| 174 | + | |
| 175 | + | |
| 176 | + | |
| 177 | + | |
| 178 | + | |
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
| 1 | + | |
| 2 | + | |
| 3 | + | |
| 4 | + | |
| 5 | + | |
| 6 | + | |
| 7 | + | |
| 8 | + | |
| 9 | + | |
| 10 | + | |
| 11 | + | |
| 12 | + | |
| 13 | + | |
| 14 | + | |
| 15 | + | |
| 16 | + | |
| 17 | + | |
| 18 | + | |
| 19 | + | |
| 20 | + | |
| 21 | + | |
| 22 | + | |
| 23 | + | |
| 24 | + | |
| 25 | + | |
| 26 | + | |
| 27 | + | |
| 28 | + | |
| 29 | + | |
| 30 | + | |
| 31 | + | |
| 32 | + | |
| 33 | + | |
| 34 | + | |
| 35 | + | |
| 36 | + | |
| 37 | + | |
| 38 | + | |
| 39 | + | |
| 40 | + | |
| 41 | + | |
| 42 | + | |
| 43 | + | |
| 44 | + | |
| 45 | + | |
| 46 | + | |
| 47 | + | |
| 48 | + | |
| 49 | + | |
| 50 | + | |
| 51 | + | |
| 52 | + | |
| 53 | + | |
| 54 | + | |
| 55 | + | |
| 56 | + | |
| 57 | + | |
| 58 | + | |
| 59 | + | |
| 60 | + | |
| 61 | + | |
| 62 | + | |
| 63 | + | |
| 64 | + | |
| 65 | + | |
| 66 | + | |
| 67 | + | |
| 68 | + | |
| 69 | + | |
| 70 | + | |
| 71 | + | |
| 72 | + | |
| 73 | + | |
| 74 | + | |
| 75 | + | |
| 76 | + | |
| 77 | + | |
| 78 | + | |
| 79 | + | |
| 80 | + | |
| 81 | + | |
| 82 | + | |
| 83 | + | |
| 84 | + | |
| 85 | + | |
| 86 | + | |
| 87 | + | |
| 88 | + | |
| 89 | + | |
| 90 | + | |
| 91 | + | |
| 92 | + | |
| 93 | + | |
| 94 | + | |
| 95 | + | |
| 96 | + | |
| 97 | + | |
| 98 | + | |
| 99 | + | |
| 100 | + | |
| 101 | + | |
| 102 | + | |
| 103 | + | |
| 104 | + | |
| 105 | + | |
| 106 | + | |
0 commit comments