Files
DriverVault/TRANSLATIONS.md
T
tajniak81andClaude Opus 5 a7cab50e06 Apprise: a gateway to hand a message to, not a hundred protocols to carry
Apprise is a Python library that speaks 100+ notification services behind one URL
grammar — mailto://, tgram://, ntfy://, discord://. None of that is portable to a
server that takes no dependencies, and none of it needs to be: caronc/apprise-api
wraps the library in HTTP and is meant to run as a container beside us. So the
connector carries no notification protocols of its own. It posts a body to an
endpoint the operator runs and lets Apprise fan it out, which is also why adding
a service later costs nothing here.

Targets are addressed one of two ways and configKey is the switch. Stateful means
the URLs live on the Apprise server under a key, narrowed by a tag expression, and
recipients are then edited there — no credential for any downstream service is
ever held in DriverVault. Stateless means the URLs travel with the request, from a
secret config field, which is simpler for one destination and worse for ten. A
call that names its own key or urls takes that destination alone rather than
merging with the configured one: honouring a caller's URLs while still falling
back to the configured key would deliver the message somewhere nobody asked for.

baseUrl is Required, which no other connector's address is. Toyota, Anker and
Greencell leave everything blank at the global layer because the superadmin → org
→ user cascade exists to fill it in, and a blank there means "let the user
choose". There is no cascade behind this one — a notification gateway is
infrastructure the operator runs, not an account a driver owns — so nothing
further down can supply the address, and a blank is simply a plugin that cannot
work. Better to fail at enable than at the first notification nobody sees.

Three limits are choices rather than gaps. /add and /del are not implemented: the
Apprise config belongs to the operator, we post to it, and a connector that can
delete a notification config has a wider blast radius than one that can only send
through it. privacy=1 is forced on /json/urls rather than offered as a parameter,
so a target listing reads mailto://user:****@host and downstream tokens stay on
the Apprise side of the wire. Attachments are remote URLs the Apprise server
fetches; multipart upload is the API's own path for files and not ours.

Health follows the rule Greencell set. A reachable server whose config holds
nothing to notify is degraded, not down: the half we address works and the missing
half is the operator's config. Two cases earn their own line — a config key set
against a server running with stateful mode disabled can never resolve, and /status
answers 417 rather than 500 when Apprise finds a problem with itself, so that is a
parsed answer and not a transport failure. A proxy that strips our Accept header
gets the same codes back as plain text, which is read rather than called
unreadable; an HTML error page from something that is not Apprise is not, and a
test pins the difference.

Notifications needed a category of their own, and that is the one change outside
the plugin: the constant, the tab order in PluginsCard.vue, and the label in all
three panel languages. The cost is now written down in the plugins README beside
the Descriptor example, since the previous five categories predate anyone having
to add a sixth.

The plugin's tests run against an apprise-api stand-in built from that project's
views.py — both notify paths, the override rules, 204-as-empty against
424-as-failure, and every health branch. builtin_test.go is the other half: the
blank-import list in builtin.go is a silent failure mode, since a connector left
out of it compiles, passes its own tests, and never appears in the panel. What is
not covered is a live instance; there is no Docker on this machine, so the wire
contract comes from reading upstream's source rather than from running it, and a
smoke test against a real deployment is still worth doing.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-30 22:43:07 +02:00

7.0 KiB

Translations (i18n)

DriverVault's user interface is translatable. Each surface reads its text from per-language files — nothing hardcodes English in the parts that have been converted — so adding a language is a matter of dropping in a new file, not editing screens.

Three languages ship today: English (en), Polish (pl) and Danish (da). English is the base and the fallback: any key missing from another language renders the English string, so a partial translation is always safe to ship.

The language is the language half of the user's BCP-47 locale (locale = language-REGION, e.g. pl-PL). The Settings Language picker sets it; the Region half keeps steering date, number and currency formatting independently, so the two can be mixed freely (English text with Polish number formatting, say). The picker offers every European language because the choice also drives date/number formatting — the ones without a translation file fall back to English UI text and say so beneath the picker.

Where the files live

Surface Language files Loader Language source
Web App (Vue) Web App/web/src/i18n/{en,pl,da}.json Web App/web/src/i18n/index.js signed-in profile locale (reactive prefs)
API Server panel (Vue) API Server/panel/src/i18n/{en,pl,da}.json API Server/panel/src/i18n/index.js localStorage (dh-panel-lang) — the panel has no user profile
Phone App (Flutter) Phone App/assets/i18n/{en,pl,da}.json Phone App/lib/i18n.dart signed-in profile locale (via AppSettings)
Phone App — Android Auto (Kotlin) the same files, read out of the APK's flutter_assets/ Phone App/android/…/phoneapp/car/CarStrings.kt the same profile locale, read from shared_preferences

All three use the same JSON shape and the same t() contract, so a translator learns one format.

The t() contract

t("settings.appearance.title")                 // simple lookup (dot path = JSON nesting)
t("forms.share.title", { name: car.name })     // {name} placeholder interpolation
t("dashboard.serviceRecords", { n: count })    // plural — see below
  • Keys are dot paths matching the nesting in the JSON.

  • Placeholders are named ({name}), never positional, so a translator can reorder them to suit the target grammar.

  • Plurals are an object keyed by CLDR category, selected for the active language by Intl.PluralRules (web/panel) or Intl.plural (Flutter):

    "serviceRecords": {
      "one":  "{n} service record",
      "other": "{n} service records"
    }
    

    This is why Polish works: it needs one / few / many where English has only one / other, and a naive n === 1 check would get "5 samochodów" wrong. Polish files therefore carry all four forms.

  • The web/panel loaders also expose tSplit(key, name) for the few strings that wrap one value in its own markup (a monospace URL, a bolded car name). It returns { before, after } around the placeholder so the value keeps its styling without splitting the sentence into word-order-assuming fragments or putting a translated string on a v-html path.

Adding a language

  1. Copy en.json to <code>.json in each surface you want to cover (fr.json, say) and translate the string values. Keep the keys and the {placeholders} unchanged. For a language with more plural categories than English, expand the plural objects (one/few/many/other as CLDR requires for that language).
  2. Register it in the loader's MESSAGES map / translatedLanguages list:
    • Web: Web App/web/src/i18n/index.js — add to the imports and MESSAGES.
    • Panel: API Server/panel/src/i18n/index.js — same.
    • Phone: Phone App/lib/i18n.dart — add the code to translatedLanguages (the file is loaded from assets/i18n/ automatically; it's covered by the assets/i18n/ directory entry in pubspec.yaml).
  3. The Settings picker already lists every European language, so the new one becomes selectable immediately and the "not translated yet" hint disappears for it. Untranslated keys still fall back to English.

No screen code changes are needed to add a language.

Coverage

  • Web App — fully translated: every view, component, form, and the status labels in lib/format.js. Both apps are complete in all three languages. The one place the wording is deliberately not translated is proper nouns: protocol and product names (OCPP, CSMS, MQTT, MQTTS, TLS, Toyota Connected, MyToyota, Anker Solix, Greencell, HabuDen, Lexus, Apprise) read the same in every file, as do the units — and so does the Greencell device command QUERY, which is a literal the charger listens for.

  • API Server panel — UI chrome, cards, login, status, and the API section titles are translated. The individual REST endpoint descriptions in the API reference table are intentionally left in English as developer reference documentation.

  • Phone App — navigation, login, lock screen, dashboard, the server picker and its add/sign-in sheet, the full Settings panel (including the language picker), the status/badge wording in lib/format.dart, the admin users screen, and the whole car screen: its tabs, every record tile, the share and delete-car dialogs, and all of the form sheets (car_form_sheet, record_form_sheets, attachment_field).

    The strings the two apps share are copied out of Web App/web/src/i18n/ rather than retyped, so a phrase has one translation across both and cannot drift. Only what the phone alone needs is written here: tooltips (the web labels its buttons), the record tiles' running prose (the web lays the same data out as table columns), client-side validation (the web leans on the browser's required), and the snackbars.

    The Android Auto screens read those very files again, by the same keys, out of the APK — the car has no Flutter engine to run i18n.dart in, so car/CarStrings.kt does the same lookup over the same JSON. Almost everything it asks for is a key a phone screen already uses, the status badges included; only carApp.* (four strings — nobody signed in, token refused, server unreachable, Refresh) is the car's own. Plurals are the one part it leaves out: no car screen needs one, and a plural key comes back as the key, which is what i18n.dart does with one it cannot render either.

    test/models_format_test.dart guards two things the analyzer cannot see. The lookups built from a key at render time (car.tabs.$key, enums.fuelType.$v, admin.roles.$r, the delete dialog's plural counts, the connected service's readings, the Service history columns and the parts a service can change) must resolve to a real label in every language — a catalogue entry with no translation fails the test rather than reaching a screen as a raw key path. And every key en.json carries must exist in pl.json and da.json, so a phrase added in English alone is caught at the point it is added rather than by whoever next reads a half-translated screen.