Vitor PamplonaandClaude Opus 5
f54194f70d
Icon(imageVector = …) calls rememberVectorPainter, and a VectorPainter rasterises its paths into a cached graphics layer per instance, so the feed re-rasterised the same glyphs once per card. A font glyph is a blit from the shared text atlas, shared across every call site for free. tools/icon-font/build_icon_font.py converts the Kotlin ImageVector DSL to SVG paths and builds a TTF with fontTools. Font metrics mirror the bundled Material Symbols font (upem 960, ascent 1056, descent -96, advance 960) so glyphs align with existing call sites; generated outlines land within a few units of Google's own. Measured on the uniform-corpus feed benchmark (SM-T220, three arms A/B/A, 0.2% identical-arm noise floor, gate 18/18/18 cards): frame duration P90 -10.7% frame overrun P90 -17.4% DrawReactions 114.8 -> 76.7 ms/iteration For reference, ablating the reaction icons entirely gives frame P90 -13.5%, so this captures ~84% of the available headroom. It supersedes the shared-VectorPainter approach (-8.2%), which needed CompositionLocal plumbing and hand-scoping to avoid cross-size cache thrashing; glyphs are atlas-shared automatically. Artwork is unchanged: this converts Amethyst's existing vectors rather than substituting Google's glyphs. Verified on device by pixel comparison -- unconverted icons are 0-diff, and the converted ones differ only by sub-pixel antialiasing between the text and vector rasterisers. Stroked icons are deliberately NOT converted. A glyph outline can only be filled, so converting Zap (strokeLineWidth 1.2) turned a thin outline bolt into a solid one; the build script now detects a stroke and skips the icon, leaving Following, Zap and ZapSplit on their ImageVectors. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01V1CzYQvWyHfipSW7x3j4Yo
Amethyst icon font
Builds amethyst_icons.ttf from the Kotlin ImageVector icons in
commons/src/commonMain/kotlin/com/vitorpamplona/amethyst/commons/icons/.
Why
Icon(imageVector = …) calls rememberVectorPainter, and a VectorPainter
rasterises its paths into a cached graphics layer per instance. A feed therefore
re-rasterised the same handful of glyphs once for every card scrolled in. A font glyph
is a blit from the shared text atlas instead, shared across every call site in the app
for free — no CompositionLocal plumbing, no per-screen scoping.
Measured on the uniform-corpus macrobenchmark (SM-T220, three arms, 0.2% noise floor):
| approach | frame P90 | overrun P90 | artwork |
|---|---|---|---|
one shared VectorPainter per icon |
−8.2% | −14.2% | unchanged |
| MaterialSymbols glyph substitutes | −10.4% | −16.0% | changes |
| this font | −10.7% | −17.4% | unchanged |
| ceiling: draw no icons at all | −12.7% | −22.9% | n/a |
Usage
pip install fonttools
python3 tools/icon-font/build_icon_font.py <icons-dir> <out.ttf> <out.kt>
See the "Amethyst's own icons are also a font" section of .claude/CLAUDE.md for the
mandatory regeneration step and why both outputs must be committed together.
How it works
The ImageVector builder DSL maps 1:1 onto SVG path commands (moveTo → M,
curveToRelative → c, …; none of the icons use arcTo), so the script extracts the
path data, emits an SVG d string, and draws it into a TrueType glyph via fontTools —
converting cubics to quadratics and flipping the y axis, since SVG grows downward and
font outlines grow upward from the baseline.
Font metrics deliberately mirror the bundled material_symbols_outlined.ttf
(unitsPerEm 960, ascent 1056, descent −96, advance 960) so the glyphs align with
existing MaterialSymbols call sites and Icon() sizing. Generated outlines land within
a few units of Google's own: our Like spans (78,94)–(882,851), their heart
(80,120)–(880,854).
An icon whose path data the parser cannot reach is reported and skipped rather than
silently emitted empty; it must keep using its ImageVector.