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).
`echo web/fonts/*.ttf` expands to every match at once, so [ -e ] on
the multi-word string failed and skipped the copy step even with all
13 fonts present. find | head -n 1 survives both multiple matches and
spaces in paths.
A second adversarial round (with empirical reproduction against the
vendored image 0.25) found the budget contract still had holes:
- The probe billed 4 bytes/px, but 16-bit formats allocate their
native layout during decode (8/px for RGBA16 PNG; TIFF stacks ~3x
codec scratch on top). In-budget 16-bit images were falsely refused
with self-contradictory notes, and the decode transient could exceed
the budget the tiers claimed to enforce. The probe now reads the
color type alongside the dimensions and bills the real peak (native
layout + RGBA where they coexist, plus the TIFF codec factor), so
refusals are honest and per-platform-correct, and the TIFF allowance
follows the same accounting.
- GIF: the probe reads the LOGICAL SCREEN, but a crafted frame may
declare itself arbitrarily larger (empirically: a 10x10 probed
screen with an 8192x8192 frame — 320 MB and 29 s from a
probe-claimed-400-byte image). The GIF frame buffer is now capped
to the probed size plus slack; frame bombs refuse like any other
oversize.
- Remote bodies on web streamed incrementally with the 64 MB cap
enforced DURING the read (array_buffer() materialized a hostile
chunked reply in full before any check could fire), aborting and
cancelling the reader the moment the cap is crossed.
- Degenerate raster entities (zero/non-finite size or vectors) no
longer produce NaN placement math that aborted whole PDF exports —
they resolve to the broken-reference placeholder.
- Same-name resolution now has a deterministic winner (newest
registration, via insertion order) instead of HashMap iteration
order; URL query strings no longer break by-name matching; the
image pick matches its refusal note by the quoted file name
(substring collisions like a.tif/2011.tif attributed errors to the
wrong file); a dropped remote outcome at the queue cap triggers one
unconditional catch-up re-resolution instead of waiting for another
event; the unknown-size refusal says so instead of printing 0x0;
and the build script fails cleanly when there are no fonts to
verify.
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.
Afonsop