Two reviewers, no blockers, three majors — all fixed.
- The remembered-tab focus after a Restore-all batch fired at the
message boundary, but the LAST restore's tab inserts one async round
later (its OPFS recent-cache write precedes the insertion) and the
insertion's own activation overwrote the selection. The focus flag
now also consumes at the insertion point, so the batch focus holds
in the common all-success case.
- The service worker cached ANY successful in-scope navigation under
the shell key — a direct navigation to a font or JSON would be
served as the app shell for every offline navigation until the next
online visit. Only the scope root caches as the shell now.
- Local macOS builds broke: sha256sum does not exist there (shasum
does) and BSD sed requires an explicit -i suffix. The wrapper picks
the available hasher and uses a portable in-place edit.
Also: '/app/' gets its own web app manifest (the previous commit's
'installable' claim was true only of the marketing site, whose
start_url sits OUTSIDE the service worker's scope — an installed app
opened offline showed an error page). The README now states the
honest offline timing (assets captured on the second visit — the
first visit's preloads outrun the installing worker) and the
long-lived-window update behavior.
Session recovery: more than one pending snapshot adds a 'Restore all'
button to the Recover Drawing modal — every snapshot goes through the
exact same guarded path as a single restore (deleted only after its
own open succeeds). The open-tab list and active tab persist to
localStorage on web (open/save/close) and the remembered tab is
re-selected once a restore batch settles; the modal notes how many
drawings the session had. Nothing is auto-opened.
Deep links: ?file=<url|name> opens a drawing at boot. http(s) URLs
must end in .dwg/.dxf and are fetched with the same streaming,
64 MB-capped, 8 s-deadline reader the remote images use, then go
through the guarded open queue; cross-origin opens announce the
origin first. Bare names resolve from the browser's recent storage,
with a hint when absent. Everything else (file:, javascript:, drive
letters, non-drawing tails) is refused with a console note.
Offline PWA: the app was installable (manifest) but downloaded its
55 MB bundle on every visit. A cache-first service worker now serves
the whole bundle from Cache Storage after one online visit —
navigations network-first so updates land, hashed assets
stale-while-revalidate, non-GET/cross-origin passthrough, old caches
cleaned on activate, and the waiting-worker update flow documented.
The build wrapper stamps the cache version from the bundle hash and
hard-fails if sw.js is missing or unstamped.
Also fixes a locale group mismatch on the DATALINK pick notice
(catalog said a display group that does not exist in the ftl
sources).
Two reviewers over the #1345/#1348/#1349 wave; no blockers, two
majors, several minors — all fixed.
- La16 was still billed at 4 B/px: the no-conversion special case
keyed on native_bpp == 4, which La16 also is — but only Rgba8 is a
real move in into_rgba8, so La16's decode peak is 8. The case now
keys on the color type, and a table test pins every layout's peak
(plus GIF's screen+frame 8) so the accounting can't silently drift.
- A control-routed open of a damaged-but-recoverable drawing parked
the operation behind the RecoveryPrompt modal (hang until the JS
timeout, silent drop if another op arrived, or false success when a
human declined). The open op now settles as failed with
recovery_required and the parse error; the prompt stays for humans,
automation can retry through the recovery flow.
- json_values_match: integers beyond f64 precision compared by value
(a u64 could round-trip-match a float); integral pairs now compare
exactly and mixed integer/float matches only when the float is whole
and equals the integer.
- A queued web open settles with the explicit queued marker whenever
the open slot was busy (the old name check made same-named in-flight
opens look completed with an empty result).
- ocs-control.js resolves with the API object reference, not
window.ocsControl (undefined if the bridge already existed when the
script ran); DATALINK on web says the typed path is superseded by
the picker instead of silently discarding it; the wipeout depth test
probes past the page background's rectangle; and the plot-path doc
comment names the function it actually mirrors.
Upstream issues, fixed:
#1345 — hatches plotted at draw_depth 0.0: the plot path cloned models
from the hatch cache, which bakes the depth at seed time; the canvas
re-reads the live depth map every frame, which is why the editor was
right and every export (PDF and native printing — both route through
the exporter) was wrong, with hatch-vs-hatch ties broken by HashMap
order. The plot path now re-reads the depth map per entity, and the
exporter's tie-break ranks follow the editor's own pass order
(hatch < image < wire fill < wire < text < wipeout) instead of pinning
hatches between fills and wires. Pinned by content-stream operator-
order tests (back/front hatches, overlapping hatches, interleaved
wire-between-hatches) and a scene test that fails pre-fix.
#1348 — set_properties' optimistic-concurrency compare now matches
numbers by value (10.0 == 10) through json_values_match, compatible
with the shape of upstream PR #1350; nested objects/arrays recurse.
#1349 — web automation contract: control protocol 2 (v1 clients still
accepted) adds {op:'open', name, data_base64} routed through the
guarded open queue; window.ocsControl.ready/request promise wrapper
with an ocs-ready event (web/ocs-control.js); the startup donation
prompt is suppressed once a control request is seen (opt-in pre-boot
via localStorage); docs/automation documents the whole web contract.
Also: DATALINK CSV-to-table works on web (picker + bytes through the
session registry — was std::fs, silently dead); COLORSCHEME accepts
System (prefers-color-scheme on web, AppsUseLightTheme on Windows).
Decode accounting, made honest per the fix-of-fix review: into_rgba8
allocates the RGBA destination while the native buffer is alive for
every non-Rgba8 layout (L8 5, La8 6, Rgb8 7, L16 6, La16 8 B/px
peaks — the formula now says so), GIF bills the screen+frame
coexistence at 8 B/px, and TIFF drops the 3x multipliers both ways —
tiff 0.11 splits the given budget between frame and scratch (a total
ceiling), so the old double-billing refused honest 16-bit GeoTIFFs
~2.25x early while admitting hostile chunk transients at 3x. The
image-refusal-unknown-size locale key moves to the group the catalog
always claimed (file).
Review of the image wave (with empirical reproduction against the
vendored image 0.25) found the budget contract did not hold across
formats: the decoder's Limits reserve checks NATIVE bytes-per-pixel,
so a grayscale/RGB JPEG/PNG/BMP over the RGBA budget decoded anyway
and into_rgba8() then ballooned to up to 4x it — the tab death the
budgets exist to prevent. The probed dimensions already prove an
OverBudget image cannot fit, so it is now refused without decoding,
removing all dependence on per-decoder limit fidelity.
Conversely, image's TIFF codec stacks ~3x scratch buffers on top of
the reserve, so an in-budget TIFF (the flagship GIS scenario) was
refused with a self-contradictory note ('400 MB, over the 512 MB
limit'); the decoder allowance for TIFF is now 3x while tier math
stays on the real budget.
Also: the refusal message saturates instead of overflowing on crafted
headers, the image pick only matches the refusal note naming ITS file
(peek, not drain — other references' notes survive for their own
drains), and the Georgian Supplement range ends at U+2D2F (U+2D30.. is
Tifinagh, which ships no font).
Build tooling: the font files missing from dist was a trunk
finalize_dist failure (Windows file-lock during the dist wipe; staging
had them, the exit code was missed) — not a copy-dir skip; it did not
reproduce in two clean cycles. scripts/build-web.sh is now the
canonical wrapper (MSYS_NO_PATHCONV, trunk, font copy + byte
verification that hard-fails), wired into README, Trunk.toml, and the
Pages workflow.
Giant rasters (500 MB aerial/GIS TIFFs) decoded whole: multiple GB of
RGBA on a wasm heap is tab death, and even desktop silently failed
past the image crate's 512 MiB default allocation cap. Decoding now
probes the header for dimensions first and runs in tiers — full decode
under a per-target budget (512 MB wasm / 4 GB native, enforced through
the decoder's allocation limits so oversize fails fast instead of
ballooning), Lanczos3 downsample to a display cap (4096² wasm /
8192² native) when the decode fits but the buffer is oversized, and a
clear localized refusal naming the dimensions and the limit when it
cannot. The probed source dimensions stay the entity's sizing truth,
so a downsampled buffer never shrinks the on-paper footprint; PDF
export self-corrects (effective DPI derives from placement × stored
pixels). Refusals surface once per image through the pick, open, and
drop/re-resolve paths.
Decode coverage gains webp and gif (pure Rust in image 0.25 — no C,
no lockfile churn), with the attach picker and drop filters matching.
Remote http(s) raster references now fetch on web: CORS-classified
failures name CORS, HTTP errors name the status, bodies over 64 MB and
8 s deadlines are bounded, and fetched bytes join the session source
registry (so ARCHIVE can package them and re-resolution reads cache).
Georgian and Armenian Noto subsets ship (generation blocks included),
so those scripts render instead of skipping; scripts still without
subsets remain silently skipped, unchanged.
The last four big web-vs-desktop gaps.
XREFs: XATTACH reads the picked drawing's bytes and registers them
name-keyed (the image/PDF-underlay pattern), pre-parses on the web
worker, and the commit merge produces the same xref block flow as
native. Opened drawings resolve xrefs from the registry (trailing-name
match) before cache build; unresolved references produce ONE warning
listing the names and the fix — drop the referenced files on the page
(drops and picker opens now register their bytes) and reopen or
XRELOAD. XRELOAD works off the registry on unsaved drawings too.
Data Link & Extraction: spreadsheet links read picked xlsx/csv bytes
(calamine from slices / CSV from string) through a session registry,
so preview, refresh, and table-insert work on web; extraction takes
external drawings via a multi-file picker parsed on the worker, and
outputs CSV/TXT/.dxex via download_bytes. The disabled 'Add folder'
button explains why.
ARCHIVE/ETRANSMIT: assembles the drawing plus registered images, PDF
underlays, and session plot tables into a deterministic ZIP (minimal
writer over flate2 — no new dependency) delivered as
{stem}_transmit.zip; skipped dependencies are reported by name.
Save/autosave off the UI thread: the parse worker gains a serialize
action (protocol v5) — the document is prepared once on the main
thread (clone, styles, dimension bake), then the DWG/DXF writer pass,
the dominant cost, runs on the worker; any worker failure falls back
to inline serialization with the identical shared implementation.
The save path also clears the dirty flag only when the tab didn't
change mid-write, and the runtime-fields sidecar now carries
graphic_data, fixing a pre-existing loss of raw DWG round-trip bytes
on web-opened documents.
Preload one language-aware font resource across the interface, drawing text, and navigation labels.\n\nRender the navigation cube, ring, outlines, and labels through a 4x MSAA target.
The web build has no system-font access, so CAD text in non-Latin scripts
(Cyrillic, Greek, CJK, …) rendered as nothing — build_fallback was a hard
None on wasm (#141). Instead, ship per-script Noto subsets under web/fonts
(one alphabet per file, SIL OFL 1.1) and fetch them lazily over HTTP the
first time a drawing uses that script, then outline glyphs with ttf-parser.
- web_font: Script enum + script_of (char → script font), a per-script
store with lazy fetch, and a pending queue drained by the app loop.
- CJK is split by language (chinese/japanese/korean). Han ideographs share
code points but differ by language, so the shared block is routed by the
document's $DWGCODEPAGE (932→JP, 949→KR, GB/936→CN); kana is always
Japanese, Hangul always Korean. Re-tessellates when the language changes.
- build_fallback (wasm) outlines from the fetched subset; clear_fallback_cache
lets glyphs that missed while a font was in flight reappear on load.
- PollWebFonts subscription (web only) + WebFontLoaded message drive fetches.
- Trunk copy-dir serves web/fonts; nothing is bundled into the native binary
(desktop keeps using system fonts).
Co-Authored-By: Claude Opus 4.8 <[email protected]>
Afonsop