aiongg
279336c445
Ported from VectorCraft's start-up fixes (#502, #545, #651, the display-GPU choice, #806, #964) and reconciled with the backend fallback of #297: - The app picks the window's adapter itself (eframe's native_adapter_selector, apps/designcraft/src/gpu.rs): adapters that can't present never, then WGPU_ADAPTER_NAME, hardware before software, native backends before OpenGL, the GPU that drives the primary display (Windows EnumDisplayDevices through winsafe, Linux /sys/class/drm), the power preference (WGPU_POWER_PREF; otherwise power saving on Windows and macOS, the system's order elsewhere), then DX12 before Vulkan. Every adapter and the choice are logged. - wgpu's indirect-call validation is off (WGPU_VALIDATION_INDIRECT_CALL=1 turns it back on): its compute shader failed to compile on an Intel Mac and lost the Metal device (#334, first error). - run_native runs under catch_unwind. A wgpu error or a panic in wgpu, egui-wgpu or naga during the first frames, or a first frame that doesn't reach the screen in 20 s, starts the app again without that adapter (DESIGNCRAFT_GPU_SKIP; exec on Unix, a new process on Windows). When no adapter of the start's backends is left, it starts again with the next backend. Each restart leaves out one more adapter or backend, so they end. Adapters without PCI ids (Metal) are keyed by name. - The backend fallback (gpu/backend.rs, gpu.json) stays for driver faults that take the process down. Windows tries DirectX 12, then OpenGL, then Vulkan, each in its own instance: Vulkan last because creating its instance loads every installed Vulkan driver, and OpenGL kept out of the DX12 instance because creating it crashed in AMD's atio6axx.dll (#167). - main returns an ExitCode and logs why the app stopped to standard error and the log file, which is all a Windows start without a console leaves (#273). README, architecture, gaps, hardware and target-app parity docs and ROADMAP describe the choice, the restarts and the variables.
10 KiB
DesignCraft roadmap
Stage: pre-alpha · next: alpha, ~0% and ~15–25 h away; beta ~28 points and ~420–750 h beyond that (the core-workflow gate, not the percentage: ready for real work is already ~47%)
Last reviewed: 2026-10-10 · Last updated: 2026-10-10 · Change: minor (stage re-normalized to pre-alpha by the core-workflow gate; earlier today: major, full re-measure against InDesign 2026 21.6; restructured to the craftrules progress-docs standard) · Target: Adobe InDesign 2026
DesignCraft aims at full Adobe InDesign parity, and to be better: faster, open (documented JSON format + IDML), scriptable by agents (MCP), and available on the web.
Headline numbers
| Value | Kind | |
|---|---|---|
| Feature breadth | ~80% (69% of InDesign 21.6's in-scope menu items matched; 99% of our 275-row checklist) | measured bounds, estimate between |
| Ready for real work | ~47% (range 42–52%) | estimated: weighted sum over the dimensions below |
| Mainstream practitioner | ~42% (weekly core 60.7% × 0.88 interaction × 0.92 stability × 0.85 file exchange) | estimated |
| Essentials user | ~54% (core features 72% × 0.90 launch × 0.88 discoverability × 0.95 files received) | estimated |
| Remaining to beta | 420–750 Opus 5.5 agent-hours (~120–200 wall-clock hours with 4–5 agents) | estimated |
| Remaining to full parity | 1,150–1,950 Opus 5.5 agent-hours | estimated |
| Audience | Ready | Opus 5.5 agent-hours to ~95% | Work that dominates |
|---|---|---|---|
| Full target (ready for real work) | ~47% | 1,150–1,950 (~70% parallel) | depth across all areas, scripting compatibility, localization, typography fidelity, INDD/IDML |
| Mainstream practitioner | ~42% | 500–850 (~65% parallel) | INDD import and IDML corpus, composition fidelity, weekly-area depth, PDF/X-1a and printing |
| Essentials user | ~54% | 135–250 (~50% parallel) | start-up robustness, context menus and copy/paste, threading, fonts, Windows printing |
Why pre-alpha: it fails one of InDesign's core workflows, the alpha gate in docs/roadmap.md. A user can't thread a story from one frame to another: the in and out ports are drawn, but clicking the out port does nothing and there is no thread or unthread command. Long text flows only through autoflow on Place and Primary Text Frame, which users report adding and keeping pages wrongly (#165, #140, #194, #352). Everything else in the gate works end to end on macOS (document setup, styles, images, object editing, PDF export, save and reopen), and ready for real work (~47%) is above the alpha bar, so manual threading is the whole distance to alpha. Beyond that, beta needs INDD opening, IDML proven on real files, composition compared with InDesign's, and basic editing fixed on Windows and Linux. Details: docs/target-app-parity.md.
By dimension
| Dimension | Ready | Hours | Doc |
|---|---|---|---|
| Features (depth) | 55% | 250–400 | target-app-parity.md |
| Typography and composition fidelity | 45% | 120–200 | typography-parity.md |
| File formats | 40% | 150–250 + 20–60 (INDD) | file-format-parity.md |
| UI/UX fidelity | 45% | 120–180 | ui-parity.md |
| Stability | 45% | 60–100 | gaps.md |
| Performance | 65% | 40–80 | gaps.md |
| Localization | 45% | 120–180 | localization-parity.md |
| Platforms | 55% | 40–80 | hardware-parity.md |
| Hardware | 45% | 20–40 | hardware-parity.md |
| Ecosystem (scripting, plug-ins) | 15% | 150–250 | gaps.md |
| AI features | 15% | 60–120 | gaps.md |
All "Ready" figures are estimates; the weights and evidence are in target-app-parity.md.
Features
Breadth from docs/parity-checklist.md (cargo xtask parity,
measured); Ready estimated.
| Area | Breadth | Ready | Hours |
|---|---|---|---|
| Application shell & workspace | 99% | 55% | 40–60 |
| Documents, pages, spreads | 100% | 65% | 20–35 |
| Layers | 100% | 75% | 5–10 |
| Frames, shapes & paths | 100% | 60% | 25–40 |
| Transform | 100% | 70% | 10–20 |
| Fill, stroke, colour | 100% | 60% | 20–35 |
| Effects & transparency | 100% | 50% | 20–30 |
| Placing & links | 93% | 55% | 20–35 |
| Type & text frames | 99% | 45% | 40–60 |
| Typography | 100% | 45% | see typography-parity.md |
| Styles | 100% | 60% | 15–25 |
| Tables | 100% | 55% | 25–40 |
| Long documents | 100% | 45% | 30–50 |
| Interactivity & digital | 100% | 40% | 30–50 |
| Output & production | 92% | 55% | 40–70 |
| XML & automation | 100% | 35% | see Ecosystem |
| View & navigation | 100% | 65% | 10–20 |
| Undo, history, saving | 100% | 70% | 5–10 |
| Accessibility | 100% | 50% | 15–25 |
Languages
Interface strings translated, of 1,724 measured. Detail: docs/localization-parity.md.
| Language | Code | Status | Translated |
|---|---|---|---|
| English | en | full | 100% |
| Simplified Chinese | zh | partial (no canvas IME) | 83% |
| Spanish | es | partial | 83% |
| Hindi | hi | none | 0% |
| Arabic | ar | partial (RTL UI and composition) | 85% |
| French | fr | partial | 83% |
| Portuguese (Brazil) | pt-br | partial | 61% |
| Indonesian | id | none | 0% |
| Japanese | ja | partial (no canvas IME) | 95% |
| German | de | partial | 83% |
| Korean | ko | none | 0% |
| Vietnamese | vi | none | 0% |
Also shipped: Italian (85%), Ukrainian (83%).
Upcoming
Ranked; detail and estimates in docs/roadmap.md, the full list in docs/gaps.md.
- Alpha gate: manual text threading (out-port click, thread/unthread commands) and Primary Text Frame fixes (#165) (15–25 h)
- Basic editing on every platform: copy/paste, context menus, caret formatting (30–50 h)
- CI running
cargo xtask cion pull requests (4–8 h) - Open INDD through a clean-room converter, owner decision on #114 (20–60 h)
- IDML real-file corpus and fixes (60–100 h)
- Composition oracle against InDesign; optical kerning; overset rule (80–140 h)
- Start on any GPU; find every installed font (25–50 h)
Documents
target-app-parity.md (assessment) · gaps.md (work list) · roadmap.md (milestones, plan) · parity-checklist.md (presence checklist) · typography-parity.md · file-format-parity.md · ui-parity.md · hardware-parity.md · localization-parity.md · architecture.md · agents: agents.md, mcp.md, control-protocol.md · CJK: cjk.md, cjk-typography.md · Arabic: arabic-typography.md
Progress log
Newest first.
- 2026-10-10 · Readiness per audience: full ~47% (method aligned with the standard, was 42%), mainstream practitioner ~42%, essentials user ~54%, each with hours to ~95%.
- 2026-10-10 · Stage re-normalized to pre-alpha under craftrules' core-workflow gate: text can't be threaded by hand (no out-port click, no thread command).
- 2026-10-10 · Full re-measure against InDesign 2026 21.6 (menu dump, bundle formats and
localizations, 124 issues): breadth ~80%, ready ~42% (down from the ~83% "including depth" of
2026-10-04, on user evidence and the INDD gap); progress docs restructured to the craftrules
standard (
docs/parity.md→docs/parity-checklist.md; new target-app-parity, gaps, roadmap, architecture, typography, file-format, UI, hardware and localization docs). Landed today: split columns (#295), PDF export options dialog (#196), column rules (#274), Paragraph Rules dialog (#263), Character Style Options (#144), PDF/X-4 RGB handling (#223), Spanish hyphenation (#222), Japanese (#272), Italian (#218) and Ukrainian (#236) interfaces, relocated packages relink (#244), graphics backend chosen before the window (#297), graphics adapter chosen by the display it drives with a restart on the next adapter or backend when one fails at start-up and the reason logged (#273, #334), IDML table styles (#288), warichu (#280), list numbering from IDML (#298), security hardening (#153, #155), many text-composition fixes (#282–#293). - 2026-10-09 · Version 0.5.0; many user reports triaged into issues (#160–#265).
- 2026-10-08 · Version 0.4.0.
- 2026-10-07 · Community PRs: data merge, Brazilian Portuguese, PDF/X-4 flags, Office fonts on macOS; CJK 1.x (Korean line breaking, fallback fonts in Package); PrintCraft renamed PdfCraft.
- 2026-10-06 · Snapping to parent items, Windows packaging, ARM64 workflow, control-channel hardening.
- 2026-10-05 · FreeBSD build; caret-boundary fixes.
- 2026-10-04 · Never-crash pass across importers and the engine; rotate spread view; kashida in styles; previous estimate: breadth 99%, "~83% including depth", ~170 h remaining.
- 2026-10-03 · Release 0.1.1; separations, XML tagging, track changes, PDF/X-4 output intent, split window, buttons, menu customization.
- 2026-10-02 · Hyperlinks/Bookmarks panels, conditional text, tagged PDF, lists across stories, style groups, Glyphs panel, High Contrast theme.
- 2026-10-01 · First commit; skeleton, IDML, PDF, tables, footnotes, cross-references, index, spelling and parallel composition on the first day.
Revision history
| Date | Change | Summary |
|---|---|---|
| 2026-10-10 | minor | Full number 42% → 47% (method aligned with the standard: plain weighted sum, no new evidence); readiness table with hours per audience |
| 2026-10-10 | minor | Headline adds mainstream practitioner (~42%) and essentials user (~54%); method in docs/target-app-parity.md |
| 2026-10-10 | minor | Stage alpha → pre-alpha under the new core-workflow gate: manual text threading is missing; alpha gate table added to docs/roadmap.md |
| 2026-10-10 | major | Restructured to the progress-docs standard; full re-measure; the "Working today" inventory moved to docs/target-app-parity.md, milestones to docs/roadmap.md, the CLI/MCP section left to docs/agents.md (which already covered it) |
| 2026-10-04 | major | Parity estimate: breadth 99%, depth ~83%, ~170 h |
| 2026-10-01 | major | First roadmap: status, milestones |