Follow-up to #7580: the escalation it added could never fire.
## What's broken
The auto-run skips a policy that has already run on a file, keyed on
`(categoryId, fileId)`. `recordRunStart` claims that key — and #7580 has
the **browser-side first pass** record its own run under `categoryId:
"classification"` for the uploaded file. So the local heuristic ticks
the very key the server escalation checks, and the AI is never asked, at
any confidence.
Trigger is the default seeded setup: **Classification as the only
on-upload policy**, and a local verdict below `high`. Any other
on-upload policy masks it, because classification then targets that
policy's output — a new file id whose key was never claimed. That's why
this went unnoticed.
Two smaller faults in the same path:
- A chained output carried no `classificationConfidence`, so
`shouldDispatchToAi` waited for a verdict that could never arrive (a
tool-derived file gets no local pass).
- Browser-local runs were polled against the server: 3 × 404 per file,
after which `MAX_NOT_FOUND` marked a local run that had actually
**succeeded** as `FAILED`.
## The fix
- `PolicyRunRecord.browserLocal`; `recordRunStart` skips the dispatch
claim for such a run. It is the first pass, not the policy's run.
- The local pass meters under `classification:local-meter` instead of
the category id, so metering dedupe survives without suppressing
dispatch.
- The poll effect skips browser-local runs.
- `CONSUME_FILES` inherits `classificationConfidence` alongside the
labels, so the verdict survives a version bump.
## How to test
Download
[`low-confidence-classification.pdf`](https://github.com/Stirling-Tools/Stirling-PDF/raw/fix/chained-classification-confidence/frontend/editor/src/proprietary/services/heuristic/fixtures/low-confidence-classification.pdf)
(checked in as a fixture, verdict pinned by a test).
With **Classification as the only on-upload policy**, upload it and
watch the Network tab:
- **Before:** no `POST /api/v1/policies/{id}/run` for classification,
ever. Console shows `local-classification-*` 404s.
- **After:** exactly one, and the engine receives `POST
/api/v1/documents/classify`.
Judge it on that request, not on the resulting label — the model's
answer varies, so a label comparison can pass or fail for the wrong
reason.
Headless equivalent:
```
npx vitest run --project proprietary src/proprietary/components/policies/usePolicyAutoRun.escalation.test.tsx
```
Passes here, fails on `main` on "asks the AI about an unsure verdict
even though the local pass already ran". Its other two cases pass on
both, so the guards still hold: a confident verdict still costs nothing,
and a file with no verdict yet still waits rather than racing the free
pass.
New tests drive the **real** run store — mocking it is what let this
through.
`task frontend:check`: 255 files / 2202 tests.
# Description of Changes
Originally, I wanted to re-enable typed linting on our repo but using
Oxlint this time to avoid the memory and speed issues that ESLint was
causing. Unfortunately, it's not stable enough yet to actually use on
our repo (although it is close, I suspect it'll be stable enough fairly
soon). I was able to remove many of the unnecessary casts that it found
though, so even though this won't be enforced, it's still worth cleaning
up what I've found.
The bundled signing test certificates expired at **07:41:10 UTC on
2026-08-26**. They were issued exactly one year earlier, so they went
from fine to fatal mid-morning with no warning, and they take down
`main` and every open branch, not just one PR.
First casualty was the `docker-compose-tests` job on #6802, which
started at 07:45:
```
java.security.cert.CertificateExpiredException: NotAfter: Wed Aug 26 07:41:10 UTC 2026
at CreateSignatureBase.checkValidity(CreateSignatureBase.java:159)
at CertSignControllerTest.testSignPdfWithPkcs12(CertSignControllerTest.java:205)
```
```
$ openssl x509 -in app/core/src/test/resources/certs/test-cert.pem -noout -dates
notBefore=Aug 26 07:41:10 2025 GMT
notAfter =Aug 26 07:41:10 2026 GMT
```
## What was broken
`CertSignControllerTest` (7 tests) and `PdfSigningServiceImplTest` (2)
fail outright. `ValidateSignatureControllerMoreTest` and
`CertificateValidationServiceMoreTest` read the same fixtures.
Auditing the rest of the repo turned up three more time bombs that had
not gone off yet:
| Fixture | Was | Problem |
|---|---|---|
| `app/core/.../certs/test-cert.*` + `test-key.*` | expired 2026-08-26 |
**already breaking every branch** |
| `test-certs/valid-test.p12`, `valid-test.jks` (proprietary + frontend
copies) | expire 2027-03-25 | same failure, seven months out |
| `test-certs/not-yet-valid-test.p12` | valid **from** 2027-03-25 |
becomes valid, so its test silently stops proving anything, on the same
day |
## What this does
**Regenerates every fixture** with the identical subject DN, alias,
password, key size and signature algorithm as before, changing only the
validity window. Nothing that any test asserts on has moved.
- valid fixtures: `2025-01-01` to `2125-01-01`
- `not-yet-valid-test.p12`: `2125-01-01` to `2126-01-01`, so it stays in
the future
- `expired-test.p12`: pinned to its permanently-past 2024 window
**Adds `scripts/generate-test-certs.sh`** as the source of truth, so the
next regeneration is one command instead of archaeology. It documents
every DN, alias and password, pins the validity windows, and runs on
Linux, macOS and Git Bash.
**Adds two guard tests** that fail with an actionable message, naming
the script, while there is still a year of runway:
- `BundledTestCertificateExpiryTest` (app/core) checks all seven formats
parse, are in their validity window, and have more than 365 days left
- `BundledWorkflowCertificateExpiryTest` (proprietary) does the same for
the valid pair, and additionally asserts the expired fixture is still
expired and the not-yet-valid one is still in the future
That last pair matters: those two fixtures exist to test a validity
outcome, and each one silently stops testing anything once the clock
passes its window.
## Verification
Run locally against the regenerated bytes, on the exact content
committed here:
```
./gradlew :stirling-pdf:test --tests '*CertSignControllerTest*' --tests '*BundledTestCertificateExpiryTest*' \
--tests '*PdfSigningServiceImplTest*' --tests '*ValidateSignatureControllerMoreTest*' \
--tests '*CertificateValidationServiceMoreTest*'
BUILD SUCCESSFUL
./gradlew :proprietary:test --tests '*BundledWorkflowCertificateExpiryTest*' --tests '*CertificateValidationIntegrationTest*' \
--tests '*SigningFinalizationServiceMoreTest*' --tests '*ServerCertificateServiceTest*' \
--tests '*CertificateSubmissionValidatorTest*' --tests '*WorkflowSessionServiceTest*'
BUILD SUCCESSFUL
```
`spotlessCheck` passes on both modules.
# Description of Changes
e2e Playwright tests are currently failing intermittently on all
platforms for different reasons, most notably WebKit, which seems to
fail much more often than the others. This PR attempts to fix the
issues. I've ran the e2e tests a few times now and they don't seem to be
inconsistent any more, but it's difficult to tell if all the issues are
genuinely fixed due to the inconsistent nature. As far as I can tell,
I've not broken anything though.
Review Flow PR 4. Stacked on #7477. Recorded failures appear in a
notification bell, showing each reader the failures they are allowed to
see and the actions they can actually take.
Scope is deliberately viewing and routing only. Resolving a failure —
retry, decrypt-and-retry — is #7479, which also brings the write path
for it; nothing resolution-shaped ships here, not even dark.
## What's added
**A notification bell** in the editor and the processor shell. Polls
`GET /api/v1/notifications` every 30 seconds, shows an unread badge, and
lists open failures newest first. Each row shows the failure's title,
its message with **Copy error** and **Show full message** chips, an
occurrence count, and its available actions.
**A notification API** (`stirling.software.proprietary.notification`),
derived from failures on read rather than stored in its own table:
| Route | Purpose |
|---|---|
| `GET /api/v1/notifications` | the caller's open failures, newest first
|
Read-only by design: every action the bell offers is one the client runs
on its own device, so there is nothing to post back. Every id is
prefixed (`failure:<uuid>`), so the bell never holds a raw failure id it
could hand to a failure endpoint.
**Per-reader actions.** A `FailureKind` declares each action with an
audience (`OWNER`, `TEAM_REVIEWER`, `ANYONE_WHO_SEES`). The server
resolves that against the reader and derives `Ownership` (`MINE` /
`THEIRS` / `UNOWNED`) from the row's actor, so an admin reviewing
someone else's failure is not offered a document their browser does not
hold. Adding a failure kind requires no frontend change.
**Server-run and client-run actions are distinguished.**
`FailureActionId` carries an `Execution` facet; the registry requires a
bean only for server actions, and dispatching a client action on the
failure surface returns 400. The notification projection goes further:
it carries only client-run offers, so the bell cannot be sent a button
it would refuse to draw.
**Actions in the bell:** at most two. The owner of the document gets
**View file** (opens it in the editor); a team reviewer gets **View in
processor** (dev builds only). Dismiss stays on the failure queue in
`/processor/documents` — deciding a failure's fate belongs to the review
surface, not the panel that announces it. An action id the build has not
wired is skipped rather than rendered dead, so the server can ship new
kinds ahead of the clients that understand them.
**Attended policy runs record their document.** `POST
/api/v1/policies/{id}/run` accepts an optional opaque `fileId`, recorded
when the run carries exactly one primary document. This is what lets a
repeat fold onto one incident instead of opening a new one per upload,
lets deleting the file clear its failure, and lets the owner open the
document from the row.
## Behaviour changes
- **The bell re-reads as soon as a failure you caused is recorded**,
rather than leaving you to wait out a poll interval for news of your own
upload. Applies to a failed tool run and to a policy run reaching
`FAILED`. Other people's failures still arrive on the poll, which is
what it is for.
- **An action the reader cannot use is not rendered.** Where the server
gave a reason for withholding it, that reason appears as the row's
one-line note. An action that was never offered to that reader produces
no note.
- **Deleting a document closes every incident about it that the deleter
caused**, including a failed policy run on their own upload, so a user's
own errors leave the bell with the file rather than lingering with a
dead button.
- **The failures list in `/processor/documents` stays behind
`import.meta.env.DEV`**, and View in processor is gated to match so it
cannot navigate to a section that is not mounted. Both lift when
failures get their own review screen.
- **One poll for all bells.** The bell is mounted in three places; the
list, document lookups and read marker are shared, so mounting more than
one does not multiply requests.
- `ACKNOWLEDGE` is no longer offered by any kind. The id, bean and
status remain so existing rows stay readable.
## Known limits
- The poll does not pause when the tab is hidden.
- No retention or per-team cap on `file_run_events`.
## How to test
Needs a proprietary or SaaS build with login enabled. `task dev:all`,
then sign in.
1. **Create a failure.** Add a password-protected PDF to the editor and
choose **Skip for now** when it asks to unlock. The upload starts a
policy run that fails on it.
2. **Watch the bell.** The badge should appear within a second or two,
not after 30 — this is the refresh-on-failure path. Open it: a row
titled "Password-protected document" with the error message and the two
chips.
3. **The buttons should be View file and View in processor, nothing
else.** No Dismiss and no retries: dispositions live on the review
surface, resolutions in #7479.
4. **View file** closes the panel and selects that document in the
editor.
5. **Dismiss from the queue instead.** Open `/processor/documents` (dev
build), find the row in the failures list and dismiss it there; the bell
drops it on its next read.
6. **Confirm the local-document probe.** Create a second failure, then
delete that file from the editor and reload. Its incident closes with
it; a row whose document is still present keeps **View file**.
7. **Confirm attribution end to end.** Sign in as a plain member, run a
shared policy on your own upload so it fails. The member sees their own
row in the bell. Sign in as the team leader: they see it too, but with
**View in processor** instead of **View file**, because the document is
not in their browser.
8. **Confirm folding.** Add the same locked PDF again and skip again.
The existing row's occurrence count increases rather than a second row
appearing.
9. **Confirm one poll for many bells.** Open the editor and the
processor in two tabs. Each tab issues its own poll, but within a tab
the several mounted bells share one — the Network tab should show one
`GET /api/v1/notifications` per 30s per tab, not three.
## Migration
None. No new column and no new value in any CHECK-constrained enum;
`CheckConstrainedEnumsTest` fails if that changes.
Auto-generated by stirlingbot[bot]
This PR updates the frontend license report based on changes to
package.json dependencies.
Signed-off-by: stirlingbot[bot] <stirlingbot[bot]@users.noreply.github.com>
Co-authored-by: stirlingbot[bot] <195170888+stirlingbot[bot]@users.noreply.github.com>
# Description of Changes
Building ontop of a users draft PR for form creation tools
**Fill Form** becomes a full **Form Editor**: fill, create, modify and
delete AcroForm fields visually. Builds on the community form-creation
draft, plus a UX/UI rework pass.
- **Backend**: `/api/v1/form` endpoints — `fields-with-coordinates`,
`add/modify/delete-fields`, combined `edit-fields` (one round-trip),
`fill`, `extract-csv/xlsx`; supports text (multiline, comb), checkbox,
dropdown, list box, radio, button actions (reset/print/URL/submit) and
signature placeholders
- **Create**: type palette, click-or-drag placement with snap guides,
inline property editor, batch "Add N fields"
- **Modify**: move/resize on the page, arrow-nudge + Delete key, X/Y/W/H
inputs, staged edits/deletes with chips, discard
- **Fill**: live progress + required tracking, flatten toggle, Export
menu (JSON/CSV/XLSX), Ctrl/Cmd+S
- **Safety**: confirm dialog before discarding staged work; empty
required fields warn with "Save anyway" instead of blocking
- **UI**: consistent panel skeleton (fixed header / scrolling list /
pinned actions), empty states that link into Create, full i18n with
plural keys
[walkthrough.html](https://github.com/user-attachments/files/30508976/walkthrough.html)
---
## Checklist
### General
- [ ] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [ ] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md)
(if applicable)
- [ ] I have performed a self-review of my own code
- [ ] My changes generate no new warnings
### Documentation
- [ ] I have updated relevant docs on [Stirling-PDF's doc
repo](https://github.com/Stirling-Tools/Stirling-Tools.github.io/blob/main/docs/)
(if functionality has heavily changed)
- [ ] I have read the section [Add New Translation
Tags](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md#add-new-translation-tags)
(for new translation tags only)
### Translations (if applicable)
- [ ] I ran
[`scripts/counter_translation.py`](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/docs/counter_translation.md)
### UI Changes (if applicable)
- [ ] Screenshots or videos demonstrating the UI changes are attached
(e.g., as comments or direct attachments in the PR)
### Testing (if applicable)
- [ ] I have run `task check` to verify linters, typechecks, and tests
pass
- [ ] I have tested my changes locally. Refer to the [Testing
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md#7-testing)
for more details.
---------
Co-authored-by: Denys Vitali <denys@denv.it>
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
# Description of Changes
The Documents tab in the Processor is supposed to be available to all
Processor users, but because the API is built on top of the Audit data,
which is only for enterprise users, the API call always fails with 403.
This means that it never fills the query cache, so every time you go
back to the tab it has to reload all the data for a couple of seconds
(and will fail again). This fixes the API so that it's available to any
Processor user instead of just enterprise users. Also, the documents
data was only being written to the log on an enterprise license, so I've
changed it so that data is always tracked in the audit log because
otherwise the Documents tab would still be useless to non-enterprise
users.
The Audit Log tab was also available to all Processor users, but would
have the same issue where the table would never load because the API
would 403 as well. I've just made the Audit Log tab disabled for
non-enterprise users now. We might want to do something to signpost it a
bit more that it's an enterprise-specific feature, but it's better than
nothing for now.
Split out of #7574 — this is the classification half, which is
independent of the editor-source work and can land on its own.
## What this does
- **Runs the local heuristic first and only escalates an unsure verdict
to the AI.** A high-confidence local answer stands; anything less (or a
file the heuristic hasn't reached yet) goes to the engine. A wrong label
costs more than an engine call, so the bar is deliberately strict.
- **Makes `classify` an authorable pipeline task**, so it can be used as
a step like any other tool, and skips files that are already classified.
- **Leaves the seeded Classification policy unowned** rather than naming
a `system` placeholder that was never a real user; existing seeds are
repaired on boot.
## Review feedback applied
From @jbrunton96 on #7574:
- **The generic runner no longer names classification.** Everything
classification-specific moved into
`proprietary/data/classificationPolicy.ts`, and `usePolicyAutoRun` now
asks capability questions instead: `policyRewritesDocument`,
`policyDeliversOutputFiles`, `policyRequiresAiEngine`,
`shouldDispatchToAi`. There is no `id === "classification"` left in the
runner.
- **Ordering is no longer a name in the runner.**
`pinClassificationLast` is gone; the runner sorts annotating policies
after rewriting ones. The constraint is real: an annotating policy is
non-blocking, so a rewriting one running after it forks from the
pre-annotation version and drops the labels. To be straight about what
this is and isn't - see "Still open" below - `policyRewritesDocument` is
still keyed on the category id, not on a property each policy declares.
The check moved out of the runner; it did not stop being a check on one
id.
- **Confidence is typed.** New `ClassificationConfidence` union in
`core/types/fileContext.ts`, reused by `fileStorage`,
`HeuristicConfidence`, and the trusted-verdict constant instead of being
respelled at each site.
- **Comments trimmed** to the repo's 2-line guideline, and a stale
seeder javadoc that still claimed an internal-user owner was corrected.
## Still open, deliberately
`classificationPolicy.ts` answers its capability questions with
`categoryId === "classification"`. That is the same check relocated, not
removed, and the module doc now says so outright.
Deliberate, for two reasons:
- **The concept it would be declared against is going away.** Policies
are becoming pipelines with labels behind a separate enforcement layer,
which removes the category the flag would live on. A capability system
built on `categoryId` today gets migrated twice.
- **Classification is genuinely privileged, not accidentally special.**
It is the only policy with a browser-side implementation, so it can
answer without the server. That is a product decision, and a local-only
mode for set scenarios is planned - the flag for it should be designed
with that feature, not guessed at now.
The end state for the rest: an in-place output mode retires the ordering
rule and `policyDeliversOutputFiles`, and a run result that can carry
findings as well as files retires the remainder. Both touch the import
path, which is the most delicate code in `usePolicyAutoRun` - not
something to bolt on to a PR that has already been split once.
Nothing is broken by leaving it. A user-built classify pipeline still
gets its labels: the generic import path reads them off the returned
PDF. It versions the file instead of labelling in place, and it misses
the local-heuristic shortcut, so it always bills the engine.
## Testing
- `classificationPolicy.test.ts` — 12 cases covering each capability and
the escalation rule
- Full frontend `proprietary` project: 39 files / 442 tests
- `:proprietary:test` for `DefaultClassificationPolicySeederTest` +
`ClassifyLabelControllerTest`
- `tsc --noEmit` on core, proprietary, portal, saas, desktop, cloud
---------
Co-authored-by: James Brunton <jbrunton96@gmail.com>
# Description of Changes
Prettier takes about 10 seconds to run over our frontend folder, but
[Oxfmt](https://oxc.rs/docs/guide/usage/formatter.html) does an (almost)
identical job in 0.2 seconds. This PR converts our Prettier integration
to an equivalent Oxfmt integration. There's exactly 2 files in the
frontend folder that Oxfmt formats differently to Prettier so it'll
barely cause any disruption to the source.
I've removed the `--check` option from the frontend tool models
generator as part of this because we can do the same thing with Task
easily enough and Oxfmt isn't directly importable like Prettier since
it's a Rust binary instead of a JS library. Originally I was shelling
out to Oxfmt on single-file mode to keep it all in memory but it just
seemed more likely that there'd be config mismatches between that script
and the task so I think it's better this way.
# Description of Changes
Follow-on from #7334. Fix more `any` type usages and ban them in the
linter. We're starting to get down to only difficult folders left now,
so some of these fixes replace an excluded folder with a couple of
individual files to reduce scope to manageable levels.
There are two real behaviour changes in this PR because of bugs that
were never caught due to the lack of proper typing:
- In the Google Drive service, `lastModified` was always `undefined`
because it should have been read via `lastModifiedUtc`, which it now is.
This means that files being read from Google Drive should now accurately
retain their last modified date from Drive.
- In the error toasts, there was translation logic to try and make
friendlier error messages, but it'd never actually fire since it relied
on `i18n` being written to `globalThis`, which it never was. It now
imports the singleton instead so that translation should start working.
I also had to tweak the way that FitText works because it was relying on
`any` typing to mix refs between different places where they weren't
technically compatible but I've changed it to go via a function and the
behaviour doesn't change.
# Description of Changes
Follow-up to #7518, picking up two mobile rough edges found while going
over that branch. Two changes, one commit each.
## 1. Tool search back in the tool list (mobile)
Tool search lives in the workbench bar's super search, which on mobile
sits on the Workspace slide. So searching for a tool meant swiping off
the tool list, typing, then swiping back. This puts a filter at the head
of the tool panel on mobile. Reuses the existing `ToolSearch` component
in `mode="filter"`, the same one the desktop fullscreen picker uses.
Drives `setSearchQuery` on `ToolWorkflowContext`, so the query,
filtering and grouped results are all existing paths. `ToolPanel` takes
a new `showSearch` prop; `RightSidebar` passes `showSearch={isMobile}`.
Desktop renders exactly as before.
**To test:**
- Open the editor at a phone-width viewport (under 1024px).
- A "Search tools..." field should sit above Favourites / Recommended in
the Tools pane.
- Typing filters into grouped results. Clearing goes back to the compact
list.
- It hides once a tool is open, and comes back on the way out.
- On desktop the field should not appear at all.
## 2. The mobile overflow menu opened with nothing in it
`WorkbenchBarMobileActions` rendered its kebab trigger unconditionally.
But every item inside is gated on `currentView === "viewer"` or
`!isCustomView`. In a `custom:*` workbench both are false, so the
dropdown was empty. `WorkbenchBarDesktopActions` renders nothing in that
case, so this only showed on phones. Now returns `null` when neither
group applies, with the two conditions named so the trigger and the
items can't drift apart again.
**To test:**
- Phone-width viewport, load a PDF.
- Open a tool with its own workbench view: Compare, Get Info report,
Show JS, Validate Signature, Edit Table of Contents, or PDF Text Editor.
- The kebab at the right of the workbench bar should be gone entirely,
rather than opening an empty menu.
- Back in the viewer or page editor it should still be there, with Print
/ Download / Save As / Close.
# Description of Changes
The macOS Dock icon renders noticeably larger than every other app. The
cause is that
`icon.icns` was **100% full-bleed** - the red rounded square filled all
1024x1024 with zero
margin. macOS does not mask or inset legacy `.icns` icons, so the
artwork has to carry Apple's
grid itself: an **824x824 body centred on a 1024x1024 canvas**.
Full-bleed therefore rendered
**24% wider and 54% larger in area** than its neighbours.
Linux had the same defect for the same reason - the hicolor PNGs were
94.9-100% full-bleed,
and GNOME's HIG says an app icon is drawn within the canvas but must not
fill it (~10% margin,
so a body around 80%). Those small existing margins were resampling
artifacts, not padding.
Windows is deliberately **left full-bleed**. Microsoft imposes no inset:
target-size assets are
drawn without tile padding and the taskbar simply scales the bitmap into
the slot. `app.ico` is
a pure rename here, byte-identical to before.
## What changed
Icons are now split per platform, since the three platforms disagree
about how much of the
canvas the artwork may fill:
| Path | Owner | Treatment |
| --- | --- | --- |
| `icons/macos/app.icns` | `dmg`, `app` | 824/1024 Apple grid |
| `icons/macos/app-512.png` | build-time only | see note below |
| `icons/linux/app-{16..512}.png` | `deb`, `rpm`, `appimage` | ~10%
margin, KDE's small-size exception at 16/32 |
| `icons/windows/app.ico` | `msi`, NSIS | unchanged, full-bleed |
Linux is selected by a new `tauri.linux.conf.json`. Tauri merges
platform configs with
JSON Merge Patch (RFC 7396), so `bundle.icon` is **replaced wholesale**
rather than appended.
## Notes for reviewers
- **`icons/macos/app-512.png` is build ballast, not a real asset.**
`tauri-codegen` requires a
PNG in the icon list for every non-Windows target, with a hardcoded
fallback to
`icons/icon.png` - a file this PR deletes. Without it the build fails.
It is embedded as
`default_window_icon`, which tao's macOS backend discards
(`set_window_icon` there is a no-op:
"macOS doesn't have window icons"). Nothing renders it.
- **Linux icon order matters.** The bundler derives the hicolor
directory from each PNG's real
pixel dimensions, so `app-128.png` lands in `128x128/`. `app-512.png` is
listed first because
the first PNG in the list also becomes the window icon, which GTK does
honour.
- **`.imgbotconfig` had to be repointed.** Its previous entry named
`icons/icon.png`, a path this
PR deletes. That exclusion is load-bearing: ImgBot once optimised the
icon to an indexed
palette and `tauri::generate_context!()` rejects non-RGBA icons,
breaking the desktop build
(#6990). All 15 generated PNGs, including the eight inside the `.icns`,
are verified colour
type 6.
- **Not fixed here:** our corner radius is 14.3% of the body where macOS
and GNOME neighbours sit
near 22%, so the icon still reads squarer than its neighbours. That is a
brand-silhouette
decision rather than the sizing bug, so it was left alone.
- The 15 pre-existing unused assets (`Square*Logo.png`, `mstile-*`,
`android-chrome-*`,
`android/`, `ios/`) are untouched. No configured bundle target consumes
them.
## Verification
`task check` was **not** run - this PR touches no Java, TypeScript or
engine Python, so it
cannot exercise the change. What was verified directly instead:
- Simulated the RFC 7396 merge and Tauri's `find_icon` resolution per
platform: Windows resolves
to `app.ico`, macOS to `app.icns` plus the stub PNG, Linux to its own
six PNGs. Every path exists.
- Both configs validate against the bundled
`@tauri-apps/cli/config.schema.json`, base and merged.
- Every PNG's real dimensions match its filename, and every body
measures exactly its nominal
inset (410/512, 154/192, 102/128, 52/64, 28/32, 14/16).
- An overlay diff of the new macOS body against the old artwork shows
only 1px antialiasing
hairlines - the mark itself is unchanged, only inset.
- Pre-commit hooks pass.
---
## Checklist
### General
- [x] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [x] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md)
(if applicable)
- [x] I have performed a self-review of my own code
- [x] My changes generate no new warnings
### Documentation
- [ ] I have updated relevant docs on [Stirling-PDF's doc
repo](https://github.com/Stirling-Tools/Stirling-Tools.github.io/blob/main/docs/)
(if functionality has heavily changed)
- [ ] I have read the section [Add New Translation
Tags](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md#add-new-translation-tags)
(for new translation tags only)
### Translations (if applicable)
- [ ] I ran
[`scripts/counter_translation.py`](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/docs/counter_translation.md)
### UI Changes (if applicable)
- [ ] Screenshots or videos demonstrating the UI changes are attached
(e.g., as comments or direct attachments in the PR)
### Testing (if applicable)
- [ ] I have run `task check` to verify linters, typechecks, and tests
pass
- [x] I have tested my changes locally. Refer to the [Testing
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md#7-testing)
for more details.
Auto-generated by stirlingbot[bot]
This PR updates the frontend license report based on changes to
package.json dependencies.
Signed-off-by: stirlingbot[bot] <stirlingbot[bot]@users.noreply.github.com>
Co-authored-by: stirlingbot[bot] <195170888+stirlingbot[bot]@users.noreply.github.com>
### Description of Changes
This Pull Request was automatically generated to synchronize updates to
translation files and documentation. Below are the details of the
changes made:
#### **1. Synchronization of Translation Files**
- Updated translation files
(`frontend/editor/public/locales/*/translation.toml`) to reflect changes
in the reference file `en-US/translation.toml`.
- Ensured consistency and synchronization across all supported language
files.
- Highlighted any missing or incomplete translations.
- **Format**: TOML
#### **2. Update README.md**
- Generated the translation progress table in `README.md` using
`counter_translation_v3.py`.
- Added a summary of the current translation status for all supported
languages.
- Included up-to-date statistics on translation coverage.
#### **Why these changes are necessary**
- Keeps translation files aligned with the latest reference updates.
- Ensures the documentation reflects the current translation progress.
---
Auto-generated by [create-pull-request][1].
[1]: https://github.com/peter-evans/create-pull-request
Co-authored-by: stirlingbot[bot] <195170888+stirlingbot[bot]@users.noreply.github.com>
Auto-generated by stirlingbot[bot]
This PR updates the frontend license report based on changes to
package.json dependencies.
Signed-off-by: stirlingbot[bot] <stirlingbot[bot]@users.noreply.github.com>
Co-authored-by: stirlingbot[bot] <195170888+stirlingbot[bot]@users.noreply.github.com>
# Description of Changes
Adds a PDF/UA converter, an accessibility report, and PDF/A conformance
level A.
**New: `POST /api/v1/convert/pdf/ua`** (Convert tool, "PDF/UA" target).
Tags an untagged PDF, marks
decorative content as artifacts, embeds missing fonts and applies the
document-level PDF/UA
requirements (title, language, tab order, form-field descriptions), then
validates with veraPDF. The
`pdfuaid` declaration is written only if validation passes, so a
returned file never claims more
than it delivers; response headers report whether it was declared, how
many checks still fail and
how many images still need a description.
**New: `POST /api/v1/security/accessibility-report`.** Reports what
fails, what the converter can fix
on its own, what needs a person, and lists the figures needing a
description with the keys the
conversion accepts back. Read-only; does not modify the file. Capped at
100 MB / 2000 pages and
weighted `LARGE_WEIGHT`, since it runs a full veraPDF pass plus the
converter's layout analysis over
every page.
**PDF/A level A.** `pdfa-1a`, `pdfa-2a` and `pdfa-3a` output formats on
the existing
`/api/v1/convert/pdf/pdfa` endpoint. Level A is level B plus tagging, so
the document is tagged
after Ghostscript (which discards any structure tree it is given) and
the level A claim is written
only if veraPDF agrees. Optional `pdfUa=true` additionally declares
PDF/UA alongside PDF/A, again
only if it validates.
Honesty rules the implementation holds to:
- **Never claim a level that was not reached.** If tagging fails, the
file is returned at level B and
is named `_PDFA-2b.pdf`, not `_PDFA-2a.pdf`. With `strict=true` the
request fails outright rather
than returning a level B file against a level A request, and a level B
pass no longer satisfies a
strict level A request.
- **Never relabel a document's language.** The requested language
(default `en-GB`) is applied only
when the document declares none; a French PDF stays French unless the
caller sets
`overrideLanguage`, and ignoring a requested language is reported as a
warning.
- **Never invent alternative text.** Descriptions come from the caller.
The Convert panel can list
the images needing one (via the report endpoint) and send them back per
figure; any image left
undescribed blocks the conformance claim rather than being papered over.
- **Never certify hidden content.** Marking images decorative, or
suppressing text that could not be
tagged reliably, withdraws the claim instead of passing the checker by
hiding content.
PDF/UA-1 and PDF/UA-2 are both offered; UA-2 raises the file to PDF 2.0
and namespaces the structure
tree, and its test asserts conformance rather than merely reporting it.
Convert steps saved in Automations/Pipelines round-trip their PDF/UA
settings (profile, language,
override, title, font embedding, descriptions).
---
## Checklist
### General
- [ ] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [ ] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md)
(if applicable)
- [ ] I have performed a self-review of my own code
- [ ] My changes generate no new warnings
### Documentation
- [ ] I have updated relevant docs on [Stirling-PDF's doc
repo](https://github.com/Stirling-Tools/Stirling-Tools.github.io/blob/main/docs/)
(if functionality has heavily changed)
- [ ] I have read the section [Add New Translation
Tags](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md#add-new-translation-tags)
(for new translation tags only)
### Translations (if applicable)
- [ ] I ran
[`scripts/counter_translation.py`](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/docs/counter_translation.md)
### UI Changes (if applicable)
- [ ] Screenshots or videos demonstrating the UI changes are attached
(e.g., as comments or direct attachments in the PR)
### Testing (if applicable)
- [ ] I have run `task check` to verify linters, typechecks, and tests
pass
- [ ] I have tested my changes locally. Refer to the [Testing
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md#7-testing)
for more details.
# Description of Changes
This PR bumps the Stirling PDF application version from `2.14.2` to
`2.14.3` across the project.
Changes include:
- Updated the Gradle project version in `build.gradle` to `2.14.3`.
- Updated the Tauri desktop application version in
`frontend/editor/src-tauri/tauri.conf.json`.
- Updated the AUR package version for `stirling-pdf-desktop`.
- Updated the AUR package version for `stirling-pdf-server-bin`.
- Updated the mocked `appVersion` used by the core frontend server
experience simulations.
- Updated the mocked `appVersion` used by the proprietary frontend
server experience simulations.
- Kept all application, desktop, packaging, and test/simulation version
references synchronized for the `2.14.3` release.
The change prepares the project metadata and packaging configuration for
the `2.14.3` release and prevents different components from reporting or
packaging the previous `2.14.2` version.
---
## Checklist
### General
- [ ] I have read the [Contribution
Guidelines](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/CONTRIBUTING.md)
- [ ] I have read the [Stirling-PDF Developer
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md)
(if applicable)
- [ ] I have read the [How to add new languages to
Stirling-PDF](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md)
(if applicable)
- [ ] I have performed a self-review of my own code
- [ ] My changes generate no new warnings
### Documentation
- [ ] I have updated relevant docs on [Stirling-PDF's doc
repo](https://github.com/Stirling-Tools/Stirling-Tools.github.io/blob/main/docs/)
(if functionality has heavily changed)
- [ ] I have read the section [Add New Translation
Tags](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/devGuide/HowToAddNewLanguage.md#add-new-translation-tags)
(for new translation tags only)
### Translations (if applicable)
- [ ] I ran
[`scripts/counter_translation.py`](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/docs/counter_translation.md)
### UI Changes (if applicable)
- [ ] Screenshots or videos demonstrating the UI changes are attached
(e.g., as comments or direct attachments in the PR)
### Testing (if applicable)
- [ ] I have run `task check` to verify linters, typechecks, and tests
pass
- [ ] I have tested my changes locally. Refer to the [Testing
Guide](https://github.com/Stirling-Tools/Stirling-PDF/blob/main/DeveloperGuide.md#7-testing)
for more details.
## What
The super-search work pinned the editor's `WorkbenchBar` visible on
every view except My Files, even with no file open. That left an empty
workbench showing a fully painted bar whose only live control was the
search — download / close / print / save were all disabled, because
those actions only make sense with a file open.
This stops forcing the bar. When nothing is open, only the global search
floats (unpainted, centered), mirroring how the Processor already works.
When a file **is** open, the `WorkbenchBar` renders exactly as before.
Also fixes a smaller Processor issue: its floating search strip was
shorter than the sidebar logo row, so the search sat higher than the
brand. Its height now matches the logo row (51px) so they line up.
## Changes
- **`Workbench.tsx`** — render the `WorkbenchBar` only when a file is
open (or a custom view supplies content); otherwise render the new
floating search. My Files and `hideTopControls` custom views are
unchanged.
- **`WorkbenchFloatingSearch.tsx` / `.css`** (new) — the editor's
`SuperSearch` floated in an unpainted strip, mirroring
`PortalSearchBar`. Its vertical band matches the bar's so opening a file
swaps in the bar without a shift.
- **`PortalSearchBar.css`** — strip height matched to
`.portal-sidebar__logo` (51px) so the Processor search aligns with the
logo.
The notification bell is intentionally out of scope — it ships in a
separate PR.
## Before / after
(ignore the bell icon in the after that’s not live yet)
<img width="2056" height="1077" alt="Screenshot 2026-08-20 at 2 28
17 AM"
src="https://github.com/user-attachments/assets/144f5216-4784-42b2-8c09-afda43577ad0"
/>
<img width="2056" height="1071" alt="Screenshot 2026-08-20 at 2 28
31 AM"
src="https://github.com/user-attachments/assets/63af9303-af8a-48dd-b113-485169fb4924"
/>
- **Editor, no file:** painted bar with disabled buttons → just a
floating search.
- **Editor, file open:** unchanged.
- **Processor:** search now vertically aligned with the logo.
## Testing
- `task frontend:check` — lint (incl. colour linters) + typecheck +
tests (247 files / 2137 tests) all pass.
- Processor alignment verified in Storybook (`Portal/Shell/AppShell`):
logo row, search strip, and search pill share the same vertical center.
- Editor float not verified in-browser (local backend is behind a login
gate); covered by types/tests and reuses the verified Processor pattern.
# Description of Changes
Currently in the Processor's Pipelines page, none of the tools which
require supporting files are usable because it's never been hooked up to
the new API to upload supporting files. This PR hooks it up to that so
all tools using supporting files work in the processor. I had to tweak
the type generation a little for this so we have a static map of which
params are for supporting files so we know to handle them differently.
The `Test with a file` button has to work a little differently than the
main run since it's running an ad-hoc pipeline so the files haven't
necessarily been saved to the server yet. In this case, it'll use
whatever local changes the user has made for those pipeline steps, and
for all other steps, it'll just use what's saved in the server.
Auto-generated by stirlingbot[bot]
This PR updates the frontend license report based on changes to
package.json dependencies.
Signed-off-by: stirlingbot[bot] <stirlingbot[bot]@users.noreply.github.com>
Co-authored-by: stirlingbot[bot] <195170888+stirlingbot[bot]@users.noreply.github.com>
## What
Both sidebars ended in a different bottom section. The editor showed an
account row (avatar, name, settings); the processor showed a "Link
Stirling account" CTA plus a `Settings` nav item and no identity at all.
They are now **one shared `<NavFooter>`** rendering the same rows in
both apps, in this order:
1. the link-account CTA (self-hosted, when unlinked)
2. free credits remaining
3. **Open \<the other app\>**
4. the account row — avatar, name, settings
It's a **single surface** with hairline dividers between rows, not
stacked cards. Rows are assembled as a list, so a row this build doesn't
show (no wallet, no processor access, nothing to link) takes its divider
with it rather than leaving a stray line.
This also fixes the profile-picture/initials desync between the sidebar
and the account settings page.
## Screenshots
Captured with the stubbed Playwright harness at 1600x900, scoped to the
sidebar and auto-cropped to the region that actually changed. Base is
`origin/main`; every state is driven by dummy backend stubs so all the
nav-bar permutations are covered.
<img width="2104" height="3044" alt="montage_cloud-dark"
src="https://github.com/user-attachments/assets/667c9a71-3ab7-4582-9259-08cda521e238"
/>
<img width="2104" height="3044" alt="montage_cloud-light"
src="https://github.com/user-attachments/assets/126bc994-efa3-436b-bf50-31d76f574eaa"
/>
<img width="2104" height="1492" alt="montage_editor-dark"
src="https://github.com/user-attachments/assets/c725726d-ea90-4fa9-bf0a-6014f3729869"
/>
<img width="2104" height="1490" alt="montage_editor-light"
src="https://github.com/user-attachments/assets/d0d34e0a-ac60-4f8c-af15-afb66fbd679e"
/>
<img width="2104" height="1066" alt="montage_processor-dark"
src="https://github.com/user-attachments/assets/51ccfe63-c327-41f4-ab91-74555b6f7148"
/>
<img width="2104" height="1068" alt="montage_processor-light"
src="https://github.com/user-attachments/assets/c90d7aba-90fa-4998-ac8f-26ce666dde71"
/>
The free-credits meter is a cloud-build surface, so the self-hosted
capture can't reach it. Those states come from the new Storybook stories
with dummy wallet data (`Shared/NavFooter`), which is also where the
credit tone bands and the collapsed rail are easiest to review.
## How it's wired
`NavFooter` is purely presentational. Each app resolves its own data
through three `@app/*` seams, so core carries no build-specific gating
and any box whose data is absent is dropped rather than rendered empty.
| Seam | core | cloud / proprietary / saas |
|---|---|---|
| `useFreeCreditsSummary` | `null` — self-hosted editor installs aren't
metered | cloud reads `freeRemaining` / `freeAllowance` off the same
`useWallet()` the Plan page's free meter uses, so the sidebar and Plan
can't disagree |
| `useOtherAppSwitch` | `null` — core ships no processor | gated on
`portalAccess` (`/api/v1/auth/me` in SaaS, the Spring session flag
self-hosted) |
| Link-account CTA | n/a | unchanged conditions — passed in as
`accountExtras`, still only when `linkState === "unlinked"`, still a
no-op in SaaS |
- The processor reads the meter through its own
`@portal/hooks/useFreeCreditsSummary` rather than the editor's `@app`
one. Self-hosted resolves `@app/*` as proprietary → core, where the
cloud wallet hook isn't in the cascade, and the implementation can't
live in `proprietary/` because core/desktop builds ship no portal and
must never resolve `@portal`. Keeping it in `portal/` gets the figure to
the linked self-hosted processor without weakening that rule; it reads
the same `GET /api/v1/payg/wallet` the Usage page's trial meter already
renders, gated on link state and behind the portal's query cache.
`portal-saas/` just re-exports the cloud hook, so both footers share one
fetch.
The processor-access gate previously lived in two near-identical
`AppSwitcher` copies. It moves into `useOtherAppSwitch`, `AppSwitcher`
now reads it too, and the duplicate
`saas/components/shared/AppSwitcher.tsx` is deleted — the logo switcher
and the footer row can no longer disagree about access.
## Profile picture sync
One `useAccountIdentity` hook now backs the editor footer, the processor
footer and the account settings page. Previously settings derived its
initial from `email[0]` while the sidebar used `displayName[0]`, and the
two drew different blue discs. Alongside that, the shared `Avatar`:
- falls back to initials when a picture URL fails to load, instead of
leaving an empty disc
- renders one letter for single-word names (`admin` → "A", not "AD")
- gains an `xl` size so the settings hero disc is the same component
## Notes
- Labelled **"Free credits"** rather than "free monthly credits":
`freeAllowance` is documented as a one-time lifetime grant, not a
monthly reset, so "monthly" would misdescribe the data. Happy to change
if the backend semantics differ from the type comments.
## Testing
- `task frontend:check` and `task frontend:typecheck:all` pass (all 9
build variants).
- 9 new `Shared/NavFooter` stories pass the Chromium + axe story scan;
`frontend:storybook:a11y:changed` reports no regressions.
- Stubbed E2E suite passes, including the `config-button` tour/settings
specs that target the account row. Two failures (`console-clean ›
landing`, `viewer-text-selection › Ctrl+C`) also fail on `origin/main`
locally — they need a backend on :8080 and clipboard permissions.
Every top bar styled itself, so none of them matched the new UI. Also,
colors on the premium banner (and possibly others) clashed since the
theme changes.
## Before Example Issue
<img width="1934" height="348" alt="Screenshot 2026-08-17 at 11 47
20 PM"
src="https://github.com/user-attachments/assets/b6f13207-2f47-4084-bd3b-2392f572c1a1"
/>
## After (all)
<img width="2880" height="800" alt="danger__dark"
src="https://github.com/user-attachments/assets/6b311dec-23e7-4059-a6bb-75527cbd2e34"
/>
<img width="2880" height="800" alt="danger__light"
src="https://github.com/user-attachments/assets/3b04b109-5146-4874-88b3-d99770ea51f8"
/>
<img width="2880" height="800" alt="default-app__dark"
src="https://github.com/user-attachments/assets/b0101b98-fce8-467a-99a6-dd40b8864da1"
/>
<img width="2880" height="800" alt="default-app__light"
src="https://github.com/user-attachments/assets/8aa21f01-6549-4d78-b217-c5547d60ab5b"
/>
<img width="2880" height="800" alt="free-tier-limit__dark"
src="https://github.com/user-attachments/assets/07fd7498-f44f-408f-8c79-9b5ea55e13df"
/>
<img width="2880" height="800" alt="free-tier-limit__light"
src="https://github.com/user-attachments/assets/f38c527b-9f64-4db0-b84c-56ca48e464cc"
/>
<img width="2880" height="800" alt="server-attention__dark"
src="https://github.com/user-attachments/assets/447a36fa-056d-4ca9-8b30-04aa0ccd6ed1"
/>
<img width="2880" height="800" alt="server-attention__light"
src="https://github.com/user-attachments/assets/f0311d45-a218-4846-b0ac-47996e2637c5"
/>
<img width="2880" height="800" alt="team-invitation__dark"
src="https://github.com/user-attachments/assets/cc368473-3b9f-4ab0-878a-67da993874c2"
/>
<img width="2880" height="800" alt="team-invitation__light"
src="https://github.com/user-attachments/assets/19d52a3e-4028-46f0-8562-9bd9cef9397a"
/>
<img width="2880" height="800" alt="upgrade-prompt__dark"
src="https://github.com/user-attachments/assets/e9d20daa-f41f-46d9-b827-a84f276e8af1"
/>
<img width="2880" height="800" alt="upgrade-prompt__light"
src="https://github.com/user-attachments/assets/9b6250c1-6a61-40d6-a7ab-e37398d67322"
/>
## What changed
- `InfoBanner` exposed 8 colour-override props (`background`,
`borderColor`, `textColor`, `iconColor`, `buttonColor`,
`buttonTextColor`, `closeIconColor`, `buttonVariant`), so every caller
invented its own look. Replaced with a closed tone set: `info` · `promo`
· `warning` · `danger`.
- Tone drives the whole bar — fill, border, icon and the button — so a
CTA can't drift from the bar it sits on. Text is neutral in every tone;
only the icon carries the tone colour.
- All colour comes from `--c-*` tokens mixed over `--c-surface`, so the
bars follow light and dark instead of ignoring them. The old bars were
hardcoded: in dark mode the two licence warnings stayed cream-on-white.
- `promo` keeps the gradient it was always meant to have, built from the
existing `--c-hue-indigo`/`--c-hue-purple` stops (documented in
`colors.css` as gradient hues, deliberately not accent-following), with
the existing `premium` button accent on it.
- Deleted the hardcoded colours from all four callers: the purple
gradient (`#667eea`→`#764ba2`), the orange soup (`#FFF4E6` / `#9A3412` /
`#EA580C`) duplicated across the urgent banner and the admin plan
section, and the fixed dark bar (`--mantine-color-dark-7`) on the team
invitation.
- `UpgradeBanner|AdminPlanSection` sat on the theme linter's exemption
list, which is how those colours survived the theme migration. Exemption
removed, so `code-colors` now guards them.
- The banner's class was colliding with `core/ui/Banner.css`'s
`.sui-banner` (16 live rules), which restyled it in the app but not in
Storybook — that's why the two disagreed on radius, border and tone.
Renamed to `.app-banner`; the two surfaces now render identically.
- Bar is square and full-bleed with a single hairline rule underneath;
button labels are optically centred.
- Added `--c-warning-subtle`, matching the existing `--c-danger-subtle`
/ `--c-success-subtle`.
- New `Shared → Top bars` story renders all six bars at once, so a
change to the shared component is visible against the whole set.
- Unrelated one-liner: `frontend/.prettierignore` now ignores the
gitignored `editor/screenshots/` capture artifacts, which were failing
`format:check` locally. Happy to drop it if you'd rather keep this PR to
the bars.
## Testing
- `task frontend:check` — typecheck, lint (oxlint + 4 theme-lint passes
+ stylelint), format, 244 files / 2119 tests.
- `frontend:storybook:a11y:changed` — clean in light and dark.
- The a11y gate caught a real defect mid-change: giving each banner
`role="region"` with the same label produced duplicate landmarks, which
the app hits for real whenever two banners show at once. Landmark
removed.
- All six bars captured in the running editor, light and dark, and
diffed against `origin/main`'s component rendered with each caller's
original props.
# Description of Changes
Overlay PDFs and Change Metadata both crashed in the Processor because
they required `FilesModalContext` and `ViewerContext` respectively.
Neither of those contexts make sense to provide in the Processor because
there are no files in context and there is no Viewer, so redesign both
tool settings to only optionally require these contexts. Their behaviour
is unchanged in the Editor but they now work in the Processor (just
without the extra info about the active files, since there are none).
Also hooks up the Reorganise Pages settings so that it can be used from
Automate. The component already existed but just wasn't being used,
which just looks like an oversight.
James Brunton