Add fonts, md-editor extensions, the ffmpeg core, the tesseract core and
English/Simplified Chinese OCR data so tools stop loading them from
third-party CDNs. Remove currency-exchanger-js and @ffmpeg/util, which
nothing imports anymore. Patch curlconverter to resolve its wasm files
against the page's <base href> so curl-converter works in a subfolder.
Refs sharevb/it-tools#571
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
* ci: move to pnpm 12, pnpm/setup v3 and faster, pinned CI
pnpm 12.6.0 is declared in devEngines.packageManager instead of
packageManager, so pnpm/setup, the Dockerfile and local installs all pick
the same version; the lockfile records the resolved binary. Every workflow
moves to pnpm/setup v3, and the Vercel workflows gain a setup step so
`vercel build` runs with that pnpm rather than whatever is on the runner.
CI time:
- ci runs format check, lint, unit tests, typecheck and build as parallel
steps instead of one after another.
- E2E builds the app once and shares dist with the three browser jobs
instead of rebuilding it in each.
- ci and E2E cancel superseded runs on the same ref.
- The Docker latest/baseurl workflows no longer repeat ci's checks. They
run on workflow_run after ci succeeds for a push, and build the commit
ci tested.
Runners are pinned to ubuntu-26.04-arm. CodeQL stays on x64
(ubuntu-26.04): only the x64 image ships the CodeQL bundle in its tool
cache. The Playwright cache key now includes the runner architecture.
Every action is pinned to its latest release by commit SHA with a version
comment, the Dockerfile base images by digest, and the pnpm and Vercel CLI
installs to exact versions, matching the Renovate presets already enabled
(helpers:pinGitHubActionDigests, docker:pinDigests).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
* ci: deploy to Vercel from one workflow with previews and a concurrency group
Replaces sharevb-vercel-preview.yml and sharevb-vercel-prod.yml with
deploy.yaml, which builds here and deploys the prebuilt output:
- One concurrency group per ref with cancel-in-progress, so a burst of
pushes costs one deployment instead of one per push.
- A GitHub environment per run (Production or Preview), which restores the
Environments sidebar and the PR's "View deployment" button.
- Pull requests build their own head commit, named after the PR branch so
Vercel shows the branch rather than "HEAD"; runs from forks are skipped,
since they get no secrets.
- Fails early on missing secrets, refuses to deploy a build without
index.html, and pins the Vercel CLI.
The production hostname for sitemap.xml and robots.txt comes from the
VERCEL_PRODUCTION_URL repository variable instead of being hardcoded, so
the same file serves any deployment. Secrets can live on the repository
or on the Production/Preview environments.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
* build: stop compiling tree-sitter's native addon
curlconverter depends on tree-sitter and tree-sitter-bash for its Node
shell parser, but the app only uses curlconverter in the browser, where its
`browser` field swaps in web-tree-sitter (WASM). The native addon is never
loaded.
Building it anyway broke installs: tree-sitter 0.21 ships prebuilds only for
darwin-x64 and win32-x64, so Linux compiles it from source, and its C++17
flags fail against Node 24.21's headers, which require C++20. That is what
failed pnpm install on the arm64 runners.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
* ci(e2e): run the browser jobs on x64
On ubuntu-26.04-arm the Playwright suite ran over 7x slower: chromium was
still in the test step after 27 minutes, against under 4 on x64. The build
job stays on arm; its output is architecture-independent.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
* test(e2e): stop the preview server with SIGTERM so runs exit
pnpm 12 starts the script it runs in its own process group. Playwright
stops its webServer by SIGKILLing the `pnpm preview` process group, which
no longer contains vite: vite survives, keeps the output pipes open, and
Playwright waits on it forever. The E2E jobs finished every test and then
hung until the job timeout.
pnpm 12.6 forwards SIGTERM to the script it runs, so stop the server with
SIGTERM instead. Locally with pnpm 12.6.0: 112 tests in 1.8 min, clean
exit, nothing left listening on the port.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
* Revert "ci(e2e): run the browser jobs on x64"
The slowdown blamed on arm was the Playwright webServer hang fixed in the
previous commit: every test had finished and the job was waiting on an
orphaned vite process. The suite is not slower on arm, so the browser jobs
go back to ubuntu-26.04-arm like the rest.
This reverts commit 9f5af383.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
* ci(docker): publish one image at a time, newest wins
Two workflow_run publishes could overlap, and whichever finished last set
the tags, even if it built the older commit. A single concurrency group per
workflow with cancel-in-progress fixes that: ci already cancels superseded
runs on the branch, so publishes arrive in commit order and cancelling the
older one leaves the newest commit on the tags. Raised in review.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017s2w39tKNdBdD8kyfThuLy
---------
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: sharevb <sharevb@gmail.com>
* ci(e2e): one job per browser with three workers
Replace the 3-way shard matrix with a per-browser matrix. Each job now
runs one Playwright project with three workers instead of three jobs each
running a mixed slice of all browsers at one worker.
The old layout gave each of three runners a single worker, so each job
used one of its four cores; a browser failure also showed up as
"shard 2/3" with no indication of which engine broke.
- `workers: 1` -> `3` on CI. Measured on a 4-core box: 9.4 min at one
worker vs 5.1 min at three, with the box ~95% busy. A fourth worker
added nothing and left no core for the preview server.
- `fail-fast: false` so one engine failing still reports the other two.
- Install and cache only the browser a job actually needs, instead of all
three in every job.
- Fix the cache key, which read `.dependencies.playwright` and always
resolved to `null`; the package is `devDependencies["@playwright/test"]`,
so the key never changed when Playwright was upgraded.
- `set_node_mem.sh` grows a `NODE_MAX_OLD_SPACE_MB` override, used to cap
each worker at 4GB. Playwright workers inherit `NODE_OPTIONS`, so
otherwise every worker claims a machine-sized heap ceiling.
The script is now POSIX. It is sourced from package.json scripts, which
run under dash, where `[[` does not exist -- every run was logging
"[[: not found" and silently falling through to the Linux branch, so the
macOS detection never worked when invoked that way.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* refactor(set_node_mem): capture memory sizes before the arithmetic
Addresses review feedback on the sourced heap-sizing script. Reads
closer to the original, which also assigned the raw value first.
Behaviour is unchanged: dash still produces --max-old-space-size=15051
by default and honours NODE_MAX_OLD_SPACE_MB.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* fix(e2e): give CI more timeout headroom now that workers share a runner
The webkit job failed on `markdown-diff`, which mounts a Monaco diff
editor -- the heaviest tool in the suite -- on the slowest engine:
1 failed [webkit] markdown-diff Has correct title
1 flaky [webkit] markdown-diff Diff editor is visible with two panes
107 passed (9.3m)
Both are first-paint timing, not behaviour. The 15s expect timeout was
sized for one worker per runner; with three sharing four cores, first
paint of a lazy-loaded tool takes longer, and the failure survived two
retries.
Raise the CI budget to 30s for assertions. The test timeout has to move
with it -- it defaults to 30s, so a 30s expect timeout would trip the
test timeout first and report a timeout instead of the failing
assertion. Local runs keep the previous 15s/30s.
Timeouts are upper bounds, so passing tests are unaffected.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* fix(e2e): drop the inline timeout that caps markdown-diff below the CI budget
Raising the CI expect timeout to 30s fixed one of the two webkit
failures in markdown-diff but not the other:
before: 1 failed + 1 flaky, 107 passed
after: 1 failed, 108 passed
The surviving failure asserts the title with an inline
`{ timeout: 15000 }`, which overrides the global expect timeout, so it
kept the old budget while its neighbour picked up the new one.
The inline value only restated what the global used to be, so removing
it is a no-op locally (still 15s) and lets CI use its 30s budget.
The four inline overrides in mermaid-exporter are left alone: those
raise the timeout to 30s rather than capping it.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* perf(test): scope jsdom to the suites that need it
`test:unit` ran `vitest --environment jsdom`, which builds a jsdom for
every test file. Only 7 of 128 need one -- the other 121 are pure
functions and pass unchanged under the node environment.
Measured on a 4-core box, `pnpm test`, warm cache:
before 89s wall, environment 128.79s, tests 4.70s
after ~56s wall, environment 7.37s, tests 5.30s
Environment setup was the single largest cost in the run and is now
~7s. The tests themselves were always ~5s.
Default to the node environment and let the 7 DOM-dependent suites opt
in with a `// @vitest-environment jsdom` docblock:
src/ui/c-input-text (@vue/test-utils mount)
src/ui/n-input-number-i18n (@vue/test-utils mount)
src/tools/certificate-key-parser (webcrypto-liner-shim)
src/tools/extract-text-from-html
src/tools/markdown-preview
src/tools/file-splitter
src/tools/django-secret-generator
Same 128 files and 1035 tests pass before and after.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* test: drop jsdom from three more suites and fix two assertions
Down from 7 jsdom suites to 4. The three removed did not need a DOM;
they needed it by accident.
certificate-key-parser
The service and its infos module each did a side-effect
`import 'webcrypto-liner-shim'`. That shim polyfills WebCrypto for
browsers without a secure context (see README: WebCrypto needs HTTPS),
and it touches browser globals at import time, so pulling it into a
node process fails before any test runs.
Node has WebCrypto natively, so the polyfill only belongs on the
browser path. Move the import to certificate-key-parser.vue, which is
what pgp-keygen.vue and pgp-encryption.vue already do. It now also
evaluates before sshpk rather than after it, and no longer loads
eagerly through the tool's infos module.
That let the test drop its jsdom workaround too: it was replacing
global.TextEncoder/TextDecoder with the text-encoding-utf-8 polyfill,
which node does not need. The `no-restricted-globals` disable existed
only for those writes and is gone with them.
Side effect: the run no longer emits the shim's asm.js linking
warning, which was the only warning left in the unit suite.
extract-text-from-html
`validateHtml` parsed the input with DOMParser, discarded the
document, and validated with a regex instead. `parseFromString` with
'text/html' recovers from bad markup rather than throwing, so the
catch was unreachable and the whole block was dead -- and the only
reason the file needed a DOM.
django-secret-generator
Used `window.crypto`; `globalThis.crypto` is the same API in browsers
and in node.
Two assertions did not test what they claimed:
expect(getTextFromHtml('<p>..</p>')).toString();
`.toString()` on the assertion object asserts nothing -- this passed
even if the function returned null. Now checks the type.
expect(getSecretKey()).toMatch(/[a-z0-9!@#$%^&*(-_=+)]{50,60}/);
`toMatch` searches rather than anchors, so any key of 50 chars or more
passed regardless of length. Split into an explicit length range and an
anchored charset check. The old class was also wrong: `(-_` reads as the
range `(` to `_`, which accepts digits and uppercase.
Also moved a valid-html case that sat in the invalid block, and fixed
the `invlid` / `django-secret-generato` typos.
128 files pass, 1035 tests -> 1036 (the django assertion is now two).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* chore(deps): drop the now-unused text-encoding-utf-8
The certificate-key-parser test was the only consumer: it replaced
global.TextEncoder/TextDecoder with this polyfill so the suite could run
under jsdom. That workaround went away when the suite moved to the node
environment, which provides both natively.
pnpm also prunes an orphaned transitive crypto-js@4.2.0, whose own
metadata marks it as no longer maintained. The lockfile change is 22
deletions and no additions.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* ci: enable oxlint type-aware linting, drop typecheck from build and CI
Adds oxlint-tsgolint and turns on `options.typeAware`, which brings 59
type-aware rules from typescript-eslint into `pnpm lint`. It found 59
issues; `--fix` resolved 3 (useless `= undefined` defaults).
The remaining 56 are semantic -- floating promises, base-to-string,
misused string spread -- spread over ~20 files, so they are staged as
`warn` rather than fixed blind in a CI change. Lint exits 0 and the
findings are visible in the output, ready to burn down by rule.
TypeScript stays on 6.0.3. The docs ask for TypeScript 7+, but that is
about tsgolint's own toolchain, not the project's: oxlint-tsgolint ships
a self-contained binary with no npm dependencies, and produces the same
56 findings against TypeScript 6.
Upgrading the project to TypeScript 7.0.2 was tried and reverted -- it
breaks the build:
[@vue/compiler-sfc] No fs option provided to `compileScript` in
non-Node environment. File system access is required for resolving
imported types.
Isolated by rebuilding with the same scripts on 6.0.3, which passes. TS
7's native port also drops the `./lib/tsc` subpath export, so vue-tsc
cannot start at all -- neither 3.3.6 nor the current 3.3.10.
Script and workflow cleanup:
- `build` no longer runs `vue-tsc --noEmit` first; it just builds.
- The `ci` workflow drops its Type check step. It already runs `pnpm
lint`, which now carries the type-aware rules. No other workflow had
a typecheck step.
- `precommit` and the PR template drop `pnpm typecheck` for the same
reason.
The `typecheck` script itself is kept, and so is vue-tsc. tsgolint
analyses no .vue files at all -- 0 of the 59 findings came from one --
so vue-tsc remains the only thing that type-checks single-file
components, and deleting it would drop that coverage with nothing
replacing it. It is simply no longer wired into the build or CI.
Renovate's TypeScript toolchain group gains oxlint-tsgolint, whose
version tracks TypeScript releases.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* refactor(oxlint): move config to oxlint.config.ts
Replaces .oxlintrc.json with a typed config built from oxlint's own
`defineConfig`, so rule names and options are checked by the editor
instead of being free-form JSON strings, and the staged type-aware rules
can carry an explanation next to them.
Verified as a faithful translation by diffing `oxlint --print-config`
between the two: identical resolved rule set (167 rules), no rule added,
dropped or re-levelled, and `options.typeAware` equal. Same 56 warnings
and 0 errors either way.
TypeScript configs are auto-discovered but marked experimental upstream
and require running under Node, which `pnpm lint` already does.
Also records why vue-tsc stays, since oxlint's vue plugin invites the
opposite conclusion: those 46 vue rules are style and correctness lint
rules, not a type checker. Planting
const deliberateTypeError: number = getSecretKey();
in a .vue script block is reported by neither `--type-aware` nor
`--type-check` -- oxlint exits 0 -- while vue-tsc catches it as TS2322.
tsgolint only analyses .ts files.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* ci: restore typecheck and run lint + typecheck on every build path
oxc-project/oxc#15761 ("oxlint: better vue support") is the tracking
issue for Vue in oxlint. It is open, unimplemented, and scoped to
parsing template bindings so structural lint rules can work -- not to
type checking. Even once it ships it would not cover what vue-tsc does,
and the maintainer notes eslint-plugin-vue cannot be fully replicated
because Vue templates carry no JavaScript bindings for the linter.
So vue-tsc stays and goes back into CI. Confirmed the gap is real by
planting `const x: number = getSecretKey()` in a .vue script block:
oxlint exits 0 under both --type-aware and --type-check, vue-tsc reports
TS2322.
Every job that runs `pnpm build` directly now runs lint and typecheck
first:
sharevb-ci + typecheck
sharevb-docker-nightly-release + typecheck
sharevb-docker-realease-latest + typecheck
sharevb-docker-realease-baseurl + typecheck
sharevb-github-pages-publish + lint, typecheck (had none)
sharevb-releases + lint, typecheck (had none)
The two publish workflows were building and shipping without any check
at all. Left alone: the nightly github-release job, already gated by
`needs: ci` on the same commit, and the e2e workflow, whose three
browser jobs build on the same commit that the ci workflow is already
checking in parallel.
`precommit` and the PR template get typecheck back for the same reason.
oxfmt config moves to oxfmt.config.ts alongside oxlint's, using oxfmt's
own defineConfig. Same 832 files are flagged before and after, so the
translation is faithful.
Format checking is deliberately NOT wired into CI yet: `pnpm fmt:check`
currently fails on 832 of 1613 files (410 .vue, 359 .ts, 50 .md, 12
other). The scripts exist but have never been enforced, so the tree has
drifted. Enabling the gate needs a repo-wide reformat, which belongs in
its own commit rather than buried here.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* ci: report formatting drift without enforcing it
Adds `pnpm fmt:check` to the ci workflow with continue-on-error, so the
832 files that predate oxfmt being run in CI are visible on every PR
without blocking anything. Drop continue-on-error once a repo-wide
reformat lands.
Only the ci workflow gets it. It is the one that runs on pull_request,
so it is where a non-blocking signal is actually read; adding it to the
docker and release workflows would just add noise to logs nobody opens.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* refactor(test): pin the timezone in config instead of every script
`TZ=UTC` was repeated across four scripts, including once redundantly
(`test` set it and then called `test:unit`, which set it again).
It is genuinely load-bearing for the unit suite. Running vitest on
America/New_York fails 7 tests across 5 files -- date-time-converter,
days-calculator, crontab-generator and friends assert on formatted dates
and offsets. Moved to `test.env` in vite.config.ts, which is enough on
its own: the full suite passes with the host on America/New_York and on
Asia/Tokyo with no inline prefix.
That also fixes `pnpm coverage`, which never had the prefix and so would
fail those same 7 tests on any machine not set to UTC.
For e2e it was doing nothing. The browser timezone is already pinned by
`timezoneId: 'Europe/Paris'`, and the specs do no date maths on the node
side -- the one `Date.now` override runs inside `addInitScript`, so it
is browser-side. The timezone-sensitive specs pass on Asia/Tokyo without
it. Pinned in playwright.config.ts anyway so the runner stays
deterministic if a spec ever does compute a date in node.
Also switches `test` from `npm run` to `pnpm run`, since everything else
in the repo uses pnpm.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* chore: drop the `test` alias in favour of `test:unit`
`test` was a pass-through to `test:unit` and nothing else, so it only
added a level of indirection and a second name for the same thing --
confusing next to `test:e2e`, which is a different suite.
Repointed all five callers: the ci, nightly-release and two docker
release workflows, plus the README's "Run Unit Tests" snippet.
Note `pnpm test` no longer resolves, since npm's implicit test
lifecycle has nothing to fall back on now.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* refactor(e2e): fold test:e2e:dev into test:e2e
The two scripts were not two environments, they were two answers to
"who owns the web server":
test:e2e Playwright starts `pnpm preview` on :5050 and tests the
production build
test:e2e:dev assumes `pnpm dev` is already up on :5173 and starts
nothing
That is one variable, not a second workflow, so it does not need a
second script. The config now derives it: if E2E_BASE_URL points
somewhere other than the preview URL, you are supplying the server and
Playwright leaves the lifecycle alone. NO_WEB_SERVER is gone -- it only
ever restated what a custom base URL already implies.
Renamed BASE_URL to E2E_BASE_URL along the way. vite.config.ts reads
BASE_URL as the app's base path (`/it-tools/`, set by the Pages and
Docker builds), so one exported value meant two incompatible things and
would have pointed Playwright at a path instead of a URL.
Both paths verified:
pnpm test:e2e 5/5, preview started
E2E_BASE_URL=http://localhost:5173 pnpm ... 5/5, dev server reused
Documented both in the README, which had no e2e section at all -- the
script was the only place the dev flow was written down. Also notes that
a cold dev server compiles on demand and can blow the assertion timeouts
on a first run; that cost 2 failures until the routes were warmed.
Also fixes webServer.command, which shelled out to `npm run preview`.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* chore: delete set_node_mem.sh
The script raised the V8 heap to total RAM minus 1GB and was sourced by
build, test:unit and test:e2e. Nothing needs it any more.
The build does not come close to the heap. It completes on Node's
default (2096MB here) and still completes when capped at 512MB -- a
thirtieth of what the script was handing out:
cap 15051MB (script) 92s, 93s
default 2096MB 94s, 103s
cap 1024MB 91s
cap 512MB 95s
Unit tests are the same story: 35s and 37s with, 36s and 37s without.
No measurable difference anywhere, so the script was not buying speed
either.
The reason is that the build is not a JavaScript-heap workload. It peaks
around 6.5GB of system memory while running fine under a 512MB heap cap,
because rolldown does the work in Rust, outside V8. The README's "you
need around 16GB" dates from the old JS bundler; corrected to ~8GB with
the measured figure.
Removing it also retires two things it dragged along: the
NODE_MAX_OLD_SPACE_MB override added for Playwright, which existed only
because every worker inherited a machine-sized ceiling from this script,
and a POSIX bug where `[[` failed under dash on every single run.
Verified with NODE_OPTIONS unset, so at the 2096MB default CI will see:
build 87s, 129 files / 1072 unit tests, e2e 5/5.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* fix(e2e): key the browser cache on the installed Playwright version
The key read the range from package.json, so it was `^1.62.0` rather
than `1.62.0`. Those agree today, but the range also matches 1.62.3 and
1.63.0: a lockfile bump inside the same range leaves the key unchanged,
so the run restores browsers for the old version and the cache is never
refreshed, because that key already exists.
Read node_modules/@playwright/test/package.json instead. pnpm/setup has
already installed dependencies by this step, so the resolved version is
available.
Raised by Copilot on the upstream PR.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
---------
Co-authored-by: Claude <noreply@anthropic.com>
* ci: use pnpm/setup for pnpm setup and install
Replace the corepack bootstrap + actions/setup-node + separate
`pnpm install` steps with a single `pnpm/setup` step, which installs
pnpm (version read from package.json `packageManager`), installs Node,
caches the pnpm store and runs `pnpm install`.
Node 24 stays pinned via `runtime: node@24`, and the GitHub Pages
workflow drops `pnpm/action-setup` in favour of the same step (it now
caches the store too).
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* chore: pin Node in package.json, bump pnpm to 11.22.0
Move the Node version out of the workflows and into package.json, so the
`pnpm/setup` steps are just `cache: true`:
- `engines.node` pins Node 24 as package metadata.
- `devEngines.runtime` is the field `pnpm/setup` actually reads to pick
the runtime, and pnpm itself uses it to run scripts, so it is needed
for the pin to take effect.
Bump `packageManager` from pnpm 11.11.0 to 11.22.0 (latest 11.x).
pnpm records the runtime in the lockfile (`node@runtime:24.19.0`), and
the 11.22.0 resolver writes peer suffixes differently from 11.11.0, so
pnpm-lock.yaml is regenerated. Lint, typecheck, unit tests and build all
pass on the updated lockfile.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
* chore: bump .nvmrc to Node 24.19.0
The repo was already on Node 24 everywhere else (Dockerfile, workflows,
and now engines/devEngines in package.json) while .nvmrc still selected
22.19.0. Align it with the rest, matching the node@runtime:24.19.0
version pnpm resolved into the lockfile.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01YAwAJ6L3Mai8eSxXf5e6SA
---------
Co-authored-by: Claude <noreply@anthropic.com>
* fix(deps): move @xmldom/xmldom off the versions upstream flagged as critical
Upstream deprecated 0.8.13 and 0.9.10 with 'this version has critical issues,
please update to the latest version'. 0.8.14 and 0.9.11 are the patched
releases of those same lines.
The direct dependency moves to ^0.8.14, which also lifts xml-crypto and
xml-encryption since both accept the whole 0.8 line. @boxyhq/saml20 pins
0.9.10 exactly, so a version scoped override takes it to 0.9.11.
epubjs asks for ^0.7.5 and the 0.7 line has no release that is not deprecated,
so 0.7.13 stays until epubjs moves; overriding it across two minors of an xml
parser is not something to do blind.
Also refreshes the stale minimumReleaseAgeExclude entry, which still named
0.8.12 and 0.8.13. The policy is not currently switched on, but 0.8.14 was
published three days ago and the list is where this repo records that.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* test(crontab-generator): pin the cron dialect detection and the execution times
getLastExecutionTimes was only asserted to return five items, and getCronType
only covered four expressions. Neither is enough to move a cron library under.
Adds a classification table spanning the standard and aws dialects, including
the six field expressions where they overlap, and freezes the clock so the
returned execution times can be pinned exactly, timezone offsets included.
No behaviour change: this is the reference the cron-parser v5 upgrade has to
reproduce.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* fix(deps): upgrade cron-parser to v5, keeping the aws dialect detected
Upstream deprecated v4 with 'v4 is no longer maintained, upgrade to v5'. The
rename is mechanical (parseExpression -> CronExpressionParser.parse) but the
upgrade is not, because v5 accepts the six field aws dialect that v4 rejected.
getCronType relied on that rejection: it tried the standard parser first and
treated a throw as the signal to fall through to event-cron-parser. Under v5
'0 11-22 ? * MON-FRI *' parses as standard, so aws expressions would have been
classified as standard, given the wrong next execution times and reported as
invalid whenever the aws dialect was requested.
Detection now keys off the '?' marker, which AWS EventBridge requires in one of
the two day fields and which standard cron gives no meaning to. Checked against
v4 over a 32 expression corpus spanning both dialects: identical classification
everywhere, including '* * * * * *', which is valid in both and stays standard.
getLastExecutionTimes keeps returning string[]: v5 types toJSON() as nullable
and the caller joins the array, so a hole would have silently rendered as a
blank line.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* fix(build): stub bare fs/promises, which cron-parser v5 reaches for
cron-parser v5 adds CronFileParser, exported from the package entry, and it
does require('fs/promises') to read crontab files. The browser build has no
business loading that, and the alias list only stubbed 'node:fs/promises'.
Vite matches aliases by prefix, so the bare 'fs' entry rewrote 'fs/promises'
into '_empty.ts/promises' and the build failed with UNLOADABLE_DEPENDENCY:
'Not a directory'. The new entry has to sit before 'fs' for that reason.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* refactor(crontab-generator): run each cron parser at most once per classification
From review feedback. getCronType could evaluate the standard branch twice for
an expression that is neither dialect, and getLastExecutionTimes asked for the
type twice, which repeated the whole classification. The '?' marker now selects
which dialect to try first and each parser runs once; the caller keeps the type
in a local.
Behaviour is unchanged, the 30 assertions of the untouched test file still pass.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
---------
Co-authored-by: Claude <noreply@anthropic.com>
* refactor(websocket-tester): use the native WebSocket instead of w-websocket-client
w-websocket-client ships a single 1465 byte UMD that wraps the platform
WebSocket: it builds `${url}/?token=${token}`, wires four handlers and returns
the socket. It nevertheless declares wsemi and lodash-es as runtime
dependencies, which its published build never imports, and wsemi alone pulls in
20 packages. Installing all of that to avoid writing twenty lines is a bad
trade, so the wrapper now lives here, behind tests.
Removes 14 packages from the lockfile (w-websocket-client, wsemi, diff2html,
tippy.js, @popperjs/core, seamless-scroll-polyfill, tree-kill, eventemitter3,
xxhash-wasm and their own transitives), and drops the Node-only `ws` from the
browser app's dependency graph; `ws` now only comes in through ethers.
Two follow-on effects:
- wsemi was the only subdependency wanting xlsx, which is why the xlsx CDN
override forced pnpm's blockExoticSubdeps protection off repo-wide. Only this
package depends on xlsx now, so the protection goes back to its default. A
forced re-resolution of all 2484 packages passes with it enabled and leaves
the lockfile byte identical.
- Connecting with an invalid url used to return { error } from the wrapper
instead of a socket, so the next send() threw 'wsc.send is not a function'
and nothing was ever logged. The failure is now reported in the log.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* fix(websocket-tester): make connection errors readable and normalize the url
Both from review feedback on the pull request:
- The log showed 'Error: [object Event]' for a failed connection, because the
native onerror handler receives a bare Event whose reason the websocket spec
deliberately withholds. formatWebSocketError() turns it into 'connection
failed', keeps the message of a real Error, and stringifies anything else.
The old client had the same flaw, but this pr claims to report connection
failures in the log, so it should actually report something.
- buildWebSocketUrl() now trims a trailing slash, so 'ws://host/' produces
'ws://host/?token=...' rather than 'ws://host//?token=...'. This is a
deliberate deviation from the previous client, which concatenated blindly.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
---------
Co-authored-by: Claude <noreply@anthropic.com>
* test: extract testable services and pin current crypto behaviour
Move the hashing, hmac, crc and ULA logic out of the single file components
into service modules next to them, and cover them with characterization tests
that pin the exact digests, encodings and ciphertexts the tools produce today.
The encryption tool had no tests at all; it now has round trip coverage for
every algorithm plus pinned ciphertexts for the deterministic ones, including
OpenSSL salted payloads produced by previous versions of it-tools.
No behaviour change: this is the reference point the dependency replacements
that follow have to reproduce exactly.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* chore(deps): replace the five deprecated packages flagged by renovate
Renovate's dependency dashboard lists five deprecated or replaced packages;
this removes all of them, keeping the observable behaviour identical.
- crypto-js -> crypto-es. Upstream discontinued crypto-js; crypto-es is its
maintained ESM/TypeScript successor and is byte for byte compatible, which
the characterization tests added in the previous commit verify: the same
1012 assertions pass before and after the swap, including the OpenSSL
salted ciphertexts of the encryption tool. It ships its own types, so
@types/crypto-js goes away too.
- plausible-tracker -> @plausible-analytics/tracker, the official replacement.
The app keeps providing the same 'plausible' injection shape, so nothing
downstream changes. apiHost maps onto the new endpoint option and keeps
posting to the very same url, trackLocalhost onto captureOnLocalhost, and
the pageviews that enableAutoPageviews() used to start are now captured by
init() itself. init() throws without a domain, unlike the old tracker, so
an enabled but domain-less config now falls back to the no-op instance
instead of breaking the app bootstrap.
- svg2png-wasm -> @resvg/resvg-wasm, the replacement named in its deprecation
notice. The renderer moves into a service covered by tests that rasterize
real svg content, and the 2 MB binary no longer has to be committed under
public/: vite emits it from the package as a hashed asset, so it can never
drift from the js glue on upgrade.
- @types/bcryptjs and @types/uuid are deprecated stubs, bcryptjs and uuid
having shipped their own types for a while.
Also drops the now dead renovate rules that only existed to keep uuid and
@types/uuid in lockstep, and fixes the tool metadata a new guard test caught:
the encryption tool advertised 'crypto-s' and the wysiwyg editor still
advertised monaco and prettier.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* test: pin that svg text is actually drawn, document the load bearing empty iv
The resvg swap is pixel identical to svg2png-wasm for shapes and gradients,
and it also fixes text that names the embedded Roboto family: svg2png-wasm
returned an empty canvas there, so the assertion guards a real improvement
rather than only the image dimensions.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
* fix: address review feedback on the plugin and renderer initialization
Two of the review comments are real regressions against the behaviour the
replaced packages had, so both are fixed with a test that fails without the fix:
- The plausible plugin called init() on every install. init() throws on the
second call, so installing the plugin on a second app instance (tests, hot
reload) would have taken the app down, where the old tracker was happy to be
constructed repeatedly. The tracker is now initialized once and reused. The
api host also gets its trailing slash trimmed so the endpoint cannot end up
with a double slash.
- initializeSvgRenderer() memoized the initialization promise including its
rejections, so one failed wasm download left the tool broken until a page
reload. The old code only latched its flag after a successful load. A failed
attempt is no longer cached.
Also extends the tool metadata guard to the removed stub type packages, and
pins the negative counters that verifyHOTP walks through when the window is
wider than the counter: their hexadecimal form keeps the minus sign, which the
hex parser reads as ffffffff rather than rejecting, and crypto-es reproduces
that exactly.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Mf5V2jMpFFXxGVXZU5rHBq
---------
Co-authored-by: Claude <noreply@anthropic.com>
* feat: mobile UX overhaul, theme flash prevention and build cleanup
UI/UX (mostly mobile):
- Prevent white flash on load: pre-paint inline style + script in
index.html applies the persisted theme before first paint.
- Match browser chrome to the app background: theme-color meta kept in
sync for Android/older iOS, and body background pinned to the Layout
color for Safari 26+, which ignores theme-color and samples the page.
- Make the document the app's scroller instead of n-layout's nested
container: single scrollbar, native iOS tap-status-bar-to-scroll-top,
natural toolbar collapsing; desktop sider becomes sticky and router
scrollBehavior restores/resets scroll on navigation.
- Full-width mobile menu that fades in under an always-visible sticky
top bar, seamless app background without the hero gradient, closes on
navigation (including About while already on /about), scroll-locked
behind, symmetric spacing around the title block and icon row.
- Language selector and theme toggle become icon dropdowns; theme
supports System/Light/Dark via useColorMode (same storage as before).
- Route-change loading spinner scoped to the page area so the nav bar
stays visible; spinner remains cosmetic-only (350ms delay).
- Halve content padding on small screens; reduce home top padding.
- Full-width expand/collapse tools button so its width doesn't jump.
- No sticky hover styles on touch devices (hover gated behind
@media (hover: hover)).
Build:
- Rename deprecated rolldown output.advancedChunks to codeSplitting.
- Fix unocss 'unmatched utility' warnings in the pretty-scrollbar
shortcut; override js-sha256 to 0.11.x (drops eval); bump
javascript-obfuscator to 5.4.6.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TNayL5ianBBucVapfkM8CC
* Remove unnecessary comment
* Refresh lockfile, remove releaseage ignore
Signed-off-by: Sebastian Danielsson <sebastian.danielsson@proton.me>
* chore: remove unused LocaleSelector component
Language selection moved into the NavbarButtons dropdown; nothing
references this component anymore. Its i18n keys are left in place.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TNayL5ianBBucVapfkM8CC
---------
Signed-off-by: Sebastian Danielsson <sebastian.danielsson@proton.me>
Co-authored-by: Claude <noreply@anthropic.com>
* perf(i18n): bundle only English eagerly, lazy-load other locales on demand
The entry bundle previously inlined all 19 locales (11 MB minified, 2.4 MB
gzip) via the virtual @intlify messages module. Now only en (plus tool-level
en files) is compiled into the entry; every other locale becomes its own lazy
chunk fetched the first time it is selected, with vue-i18n falling back to
English until the messages arrive.
VITE_AVAILABLE_LOCALES now filters the locales offered at runtime instead of
trimming the build; unlisted locale chunks are emitted but never fetched.
Also adds shims for *.yml default imports and @tabler/icons-vue per-icon
deep imports (used by the follow-up icon-loading commit).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(tools): load tool icons lazily and parallelize boot config fetches
The eager tool-metadata glob pulled ~460 static icon component modules into
the entry graph. Tool definitions now reference their icon via
defineAsyncComponent + per-icon dynamic import, and an advancedChunks group
merges all of them into a single lazy 'tool-icons' chunk fetched off the
critical path when the menu/home first renders. Exact same glyphs as before.
Entry-graph components (navbar, layouts, ToolCard, copy buttons, command
palette, home) switch from runtime icon packages to build-time-inlined
~icons/tabler imports so no static reference drags the icon chunk into the
entry bundle.
src/tools/index.ts also fetched tools-filter.json and external-tools.json
sequentially under top-level await, blocking first render on two serial
round-trips; they are now fetched in parallel.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(pwa): precache only the app shell, cache chunks and wasm on demand
The service worker precached every JS chunk and WASM binary — 1,679 files,
~162 MB — on first visit, saturating bandwidth while the page was trying to
become interactive. Precache is now limited to the shell (css/html) with
CacheFirst runtime caching for same-origin hashed scripts, workers and wasm,
so offline support covers everything the user has actually opened.
VITE_PWA_FULL_PRECACHE=true restores the previous cache-everything behavior
for self-hosted full-offline deployments.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(menu): eliminate layout thrash in sider and category animations
The sider show/hide animates CSS width, which re-laid-out all ~460 menu
items plus the page content on every frame; category expand/collapse
animates grid-template-rows with the same cost. Two changes cut the
per-frame work:
- content-visibility: auto on each category container so offscreen
categories skip layout and paint entirely during animations
- the sider scroll content is pinned to the expanded width, so the width
transition clips the menu instead of re-wrapping every item per frame
Measured with the Long Animation Frames API (dev mode, full hide+show
cycle): 4 long frames totalling ~395 ms before, 2 totalling ~236 ms after;
animation frames after the initial commit now stay under 50 ms.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(deps): replace moment with date-fns in countdown and alarm tools
moment was only used by three tools, each pulling its ~70 KB gzip bundle
into their chunks while date-fns (already a dependency) covers the same
needs tree-shaken. Also fixes a latent bug: moment's mutating add()/set()
modified the shared 'now' ref on every computed evaluation, so the deadline
could drift by one duration per re-render; date-fns operations are
immutable.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(deps): replace lodash with es-toolkit
Full CJS lodash (~70 KB gzip, no tree-shaking) was in the entry graph via
the tools store and shared composables, plus four lodash.* micro-packages.
es-toolkit provides the same functions as tree-shakeable ESM:
- import _ from 'lodash' -> import * as _ from 'es-toolkit/compat'
(namespace member access tree-shakes under Rolldown)
- the 20 _.chain pipelines are rewritten as plain array/object operations
or es-toolkit calls (es-toolkit has no chain, deliberately)
- lodash.defaultsdeep/flattendeep/last/mergewith micro-packages replaced
with es-toolkit core/compat equivalents
- lodash's RecursiveArray type is declared locally in generate-tree.ts
- files whose only lodash use was inside a rewritten chain lose the
dependency entirely
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* fix: catch lodash usages the codemod missed (.js/.tsx files, bare _() wrapper)
outputFormatter.js and diff-viewer.models.tsx were outside the codemod's
*.ts/*.vue glob, and otp.service.ts used the implicit-chain _(...) wrapper
form that the _. usage regex didn't match (its import was wrongly removed).
Both builds still passed because pnpm's hoisted store made 'lodash'
resolvable even after removal from package.json.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(boot): drop the splash screen and parallelize the settings fetch
The splash screen existed to cover a multi-second blank period while the old
13 MB entry bundle parsed; with the startup payload at ~3 MB the app shell
paints fast enough to render directly (measurements in the PR description).
Removes vite-plugin-splash-screen entirely.
tools-settings.json was fetched under top-level await in main.ts, serially
after the config fetches in src/tools/index.ts. It now lives in a sibling
async module so all three optional config fetches share one round-trip.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* fix: defer preload-error reload so it cannot abort an in-flight navigation
Firefox fires vite:preloadError for module preloads cancelled by a user
navigation (NS_BINDING_ABORTED), and the immediate location.reload() raced
and aborted that navigation. With the tool icons now loading via dynamic
imports there are ~460 preloads in flight at startup, which made the race
deterministic in the date-time-converter Playwright specs on firefox.
Deferring the reload by a tick resolves it: if the page is genuinely
navigating away its timers die with the document; on a real chunk-load
failure the reload still happens.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* fix(types): satisfy the vitest tsconfig typecheck
pnpm typecheck runs vue-tsc against tsconfig.vitest.json, which lacks the
unplugin-vue-i18n type shim and infers stricter unions:
- tools/index.ts: type the fetch fallbacks so Promise.all doesn't widen
filterConfig to 'ToolsFilter | {}'
- i18n.plugin.ts: collapse es-toolkit merge's recursive mapped type before
it reaches vue-i18n's generics (TS2589), watch the locale through the
getter, and cast the messages record
- generate-tree.ts: drop flattenDeep/RecursiveArray for a plain spread —
child output is already joined strings, so there was only one level to
flatten
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* perf(bundle): cut startup payload from ~1 MB to ~560 KB gzip
Four eager-loading leaks accounted for almost half the startup JS:
- Font Awesome (298 KB gzip): main.ts registered the entire solid icon set
via library.add(fas) although only the pomodoro timer uses it. The six
icons it needs are now registered inside the tool's lazy chunk.
- tool-icons chunk (67 KB gzip): rolldown's advancedChunks defaults to
includeDependenciesRecursively, which captured shared helpers into the
icon chunk and made entry chunks statically depend on it, dragging the
whole lazy icon set back into the startup payload. Disabled per-group.
- markdown-it + entities + dompurify (~50 KB gzip): eagerly reachable via
three paths — the external-tools mapping in tools/index.ts (now a
conditional dynamic import), vue-markdown-render in tool.layout.vue and
Home.custom.vue (now defineAsyncComponent).
- demo gallery routes: registered behind a runtime env check the bundler
cannot eliminate; now a conditional dynamic import gated on
import.meta.env.DEV so production builds drop it entirely.
Startup JS: 3.2 MB -> 1.92 MB raw, ~1.05 MB -> 562 KB gzip.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
* fix(lint): drop redundant spread flagged by oxlint
Object.values already returns a fresh array, so it can be pushed into
directly.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01RKAJuYi4dP1mwsjPi4v5Ea
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* chore: upgrade toolchain to Vite 8, Vue 3.5, Vitest 4, TypeScript 6 and pnpm 11; adopt oxlint/oxfmt
- vite 4 -> 8.1.3 (rolldown-based: build.rollupOptions -> rolldownOptions,
drop optimizeDeps.esbuildOptions), @vitejs/plugin-vue 4 -> 6, plugin-vue-jsx 3 -> 5
- vue 3.3 -> 3.5.39, vue-tsc 1.8 -> 3.3.6, vitest 0.34 -> 4.1.9, jsdom 22 -> 29
- typescript 5.2 -> 6.0.3 with moduleResolution=bundler (baseUrl/node10 are
hard errors in TS 6, removed in TS 7); TS 7 RC not adoptable: its native
tsc lacks the JS compiler API vue-tsc needs (vuejs/language-tools#5381)
- unocss 0.57 -> 66 (presetUno -> presetWind3), unplugin-* to current majors
(unplugin-vue-components dtsTsx disabled: emits invalid declarations for
dotted component names)
- @vueuse/head 1.0.0 -> 1.3.1 (1.0.0 silently stopped applying page titles
under Vue 3.5)
- replace ESLint (@antfu config) + Prettier with oxlint + oxfmt; configs
mirror the previous style rules. Existing files are intentionally NOT
mass-reformatted (oxfmt has no brace-style/operator-position options to
match the old @antfu style) — only files with real changes are formatted
- pnpm 9 -> 11.10.0: settings move to pnpm-workspace.yaml, fresh lockfile
passing supply-chain policies, explicit allowBuilds decisions,
blockExoticSubdeps off (xlsx CDN override patches vulnerable npm 0.18.5)
- isolated-vm (17 MB) and onnxruntime-node (208 MB) stubbed via overrides:
native-only transitives already aliased to an empty module for the bundle
- elliptic overridden to npm ^6.6.1 (github fork tarball of
webcrypto-liner-shim unreachable from the build environment)
- CI: node 22 -> 24, all actions pinned to latest patch releases, automatic
Vercel workflows disabled (fork deploys via Vercel Git integration),
vercel.json: git.deploymentEnabled removed, buildCommand skips no-op vue-tsc
- Dockerfile: node:24-alpine, pnpm@11, ship pnpm-workspace.yaml and stubs
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
* fix: resolve type, lint and runtime issues from the toolchain upgrade
vue-tsc 3 checks templates far more strictly than 1.8 and TS 5.7+ made
Uint8Array generic over its backing buffer. Widen shared component props
(InputCopyable, c-table, c-input-text, UnitsConverter), convert values at
call sites, fix invalid local .d.ts declarations, and repair issues the
new tooling exposed:
- WireGuard generator: pre-shared-keys checkbox used v-model:value which
never binds on a standalone n-checkbox (now v-model:checked)
- mask-closable="false" string attrs became real boolean bindings in
countdown/cron-alarm/time-alarm
- v-model casts (x as any) fail to compile under plugin-vue 6 + rolldown;
replaced with :value/@update:value split bindings
- invalid '::ffff::' IPv6 literal in ipv6registry.json, exposed by stricter
parsing in cidr-tools 11.3 (IPv4-mapped range never matched)
- fanger pinned to 0.3.1 (0.3.2 published without dist/)
- countries-and-timezones namespace imports, Vitest 4 snapshot format,
invalid ul-inside-p HTML in dead-pixel, test assertion outside it() block
vue-tsc clean; oxlint clean; vitest 665/665; vite build + PWA OK;
Playwright e2e 100/100; all 457 tool routes render without errors
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
* chore: drop whole-tree oxfmt from precommit instructions
The tree deliberately keeps its existing @antfu-era formatting; running
'pnpm fmt' (oxfmt over all of src) would reformat ~800 untouched files.
Format only the files you change, or run 'pnpm fmt' in a dedicated PR.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
* fix: escape backslashes before dots in parquet column keys
CodeQL flagged incomplete string escaping (js/incomplete-sanitization):
escaping '.' to '\.' without first escaping existing backslashes makes
column names containing backslashes ambiguous in naive-ui's nested-key
syntax. Escape backslashes first.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
* test: allow 30s for initial mermaid render in e2e
mermaid is a large lazy-loaded chunk; webkit on cold CI runners exceeds
Playwright's default 5s expect timeout even across retries.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
* refactor: replace 'as any' input type casts with proper typings
Copilot review on the upstream PR flagged the casts. c-input-text
renders any non-password type as a text input, so widen its type union
to include 'url' instead of casting; for naive-ui's NInput use the
documented input-props to set the native input's type=number.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
(cherry picked from commit 906820336b271aa6d96ecb11efa03219bf869475)
* fix: reconcile lockfile with upstream sync; repair date-picker typing in Docker Label Generator
- fold upstream's new CodeMirror/picomatch dependencies into the pnpm 11
lockfile; add @lezer/highlight to optimizeDeps (kept from upstream)
- dockerfile-label-generator: n-date-picker emits a number timestamp,
not a Date — calling toISOString() on it would throw at runtime the
moment a date was picked. Type as number and convert via new Date()
- drop stray tsconfig.vite-config.tsbuildinfo build artifact; ignore
*.tsbuildinfo
vue-tsc clean; oxlint clean; vitest 665/665; vite build OK
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
* chore: keep upstream Vercel CLI deployment setup
The fork-specific Vercel changes (Git-integration deploys, disabled CLI
workflows, build:vercel command) are excluded from the upstream
contribution; upstream deploys via its own Vercel CLI workflows.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
(cherry picked from commit ed182ec55648358556343e60e1667129acdbb966)
* fix: escape backslashes in Docker label values; raise global e2e expect timeout
- dockerfile-label-generator escaped quotes without escaping existing
backslashes first (CodeQL js/incomplete-sanitization, same class as
the parquets-reader fix): a value ending in a backslash would break
the generated LABEL quoting
- playwright: 15s global expect timeout — first paint of heavy lazy
chunks exceeds the 5s default on slow CI browsers (webkit), which
intermittently failed title assertions across retries
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01M6deEDJXFzgDzPiYPAcsv1
(cherry picked from commit 5b3822e10113a49bc519661a5b01ad62660d56a9)
* Update .gitignore to include specific file types
* Add stats.html to .gitignore
* Delete tsconfig.vite-config.tsbuildinfo
* Delete stats.html
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: sharevb <sharevb@gmail.com>