neuwirth
/
epiphany
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
epiphany/crates/epiphany-render-svg/DECISIONS.md

5.5 KiB
Raw Blame History

epiphany-render-svg — decisions and Pass 12 candidates

This file records (a) the Phase-2 QUICKSTART decisions Agent I made for the renderer, and (b) ambiguities batched as Pass 12 candidates (spec/PASS12_BATCH.md) rather than improvised in code.

Scope and phase status

epiphany-render-svg is one renderer behind the Chapter 7 RenderIR interface: it turns a ResolvedLayoutIR into well-formed SVG 1.1, drawing each glyph as a genuine Bravura SMuFL outline <path>. Per the QUICKSTART development pattern it is built and golden-locked against the stub solver's output first, before the real engraving solver and the score→real-notation engraving pass land. The stub returns the constrained IR's geometry verbatim — a structural projection, not yet real notation (each layout object becomes one arbitrary glyph in a row) — so this phase proves the renderer is correct and faithful (genuine outlines, provenance preserved, output XML-valid and deterministic), independent of engraving quality. The renderer already consumes any solver's ResolvedLayoutIR, so when the real epiphany-engrave solver lands, the visible result improves with no renderer change (the demo binary's --solver=stub|real flag exercises both).

The non-overreach rule (Chapter 7 / QUICKSTART, Agent I)

The renderer makes SVG-encoding choices only and no engraving-semantic choices. Every emitted element traces to a ResolvedGlyph (and thus a score-graph source, via data-prov/data-source-kind) or to a declared renderer wrapper (the <svg> root, the metadata comment, the y-flip <g>, a per-layer <g>). A glyph with no bundled outline is surfaced as a diagnostic and drawn as a visible bounding-box fallback <rect> — never silently dropped and never invented. The acceptance harness asserts one drawn element per glyph and one provenance trace per drawn element.

Implementation decisions (QUICKSTART "Decisions you'll need to make")

  1. Spelling / 2. Solver architecture — N/A here (Agent H; epiphany-engrave).
  2. Renderer SVG dialect — SVG 1.1 + inline presentation attributes. Maximum portability for what is effectively a viewer/print artifact; no CSS, no SVG 2 features. (CSS styling can come later if needed.) Output validates under the system xmllint (libxml2) as well as the in-crate well-formedness checker.
  3. Catalog / 5. Binary Format versioning — N/A (Agents K, J).

Local decisions

  • Glyph rendering — inline genuine Bravura outline <path>s (GlyphMode::PathOutline), the default and only mode this phase. The QUICKSTART's recommendation: path outlines make the SVG self-contained (it renders in any browser, image tool, or print pipeline with no font installed), at the cost of file size. An embedded-@font-face mode is a documented future option, intentionally not stubbed so the interface does not lie about a capability that is absent.
  • Outline source — the official OFL Bravura.otf, extracted reproducibly. tools/extract_bravura_outlines.py fetches the font + SMuFL glyphnames.json and emits src/outlines_generated.rs. The font is not vendored; only the generated Rust is committed (tools/OFL.txt carries the SIL Open Font License 1.1 under which the outlines are redistributed). Exactly the glyph set the v0 pipeline can name (layout-ir's BRAVURA_METRICS) is bundled; a test asserts every pipeline glyph has an outline (so the table cannot silently fall behind).
  • Coordinate system — staff spaces, y-up, one global flip. Outlines are extracted in staff-space units (SMuFL em = 4 staff spaces, Bravura unitsPerEm = 1000 ⇒ 1 staff space = 250 font units), y-up (musical convention). The whole document is wrapped in one translate(-min_x, max_y) scale(1, -1) group, so every glyph is placed with a plain translate(x, y) and the viewBox is 0 0 W H in staff spaces; width/height carry the px display scale. The renderer never bakes the flip into per-glyph data.
  • Determinism — fixed 4-decimal number formatting, -0 normalised to 0. Outline d data is fixed bundled text; every computed coordinate is formatted identically, so identical input yields byte-identical SVG. The acceptance harness golden-locks the full SVG and a machine snapshot (object/glyph/path/ provenance/layer/per-class/hard-constraint counts + XML validity).
  • No external XML dependency. The workspace is deliberately dependency-light; xml::check_well_formed is a hand-rolled validator for the subset the renderer emits, and the acceptance test additionally cross-checks with xmllint when present so the "XML-validates" claim rests on a real parser too.

Pass 12 candidates

See spec/PASS12_BATCH.md (rows P12-I1, P12-I2, P12-I3). Most relevant here:

  • P12-I1 — the constrained IR is a structural placeholder, so the rendered stub output is not yet recognizable notation. The QUICKSTART's human-review visual-acceptance gate ("the SVG visually parses as standard music notation") is therefore a next-phase gate, met once real engraving lands; this phase's gate is renderer correctness/faithfulness. Recorded so the visual gate is not mistaken for already-met.
  • P12-I2 — stable layout-object id derivation (MUSCLOID, Pass-11 item 2.6, deferred to Agent I) is still unwired: the determinism crate exposes no MUSCLOID tag and is frozen. The renderer traces provenance by the existing provisional stable_id; wiring the ratified derivation is Track A work.