chore(deps): Update i18next (major) #337
No reviewers
Labels
No labels
dependencies
major
security-pin
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set
Reference
trails-cool/trails!337
Loading…
Reference in a new issue
No description provided.
Delete branch "renovate/major-i18next"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
This PR contains the following updates:
^26.3.6→^27.0.0^17.0.9→^18.0.0Release Notes
i18next/i18next (i18next)
v27.1.1Compare Source
t()skips the interpolation and nesting regexp scans when the text has no interpolation or nesting prefix, about 6-14% faster for short texts and up to 75% for long ones. The output does not change. Thanks @g4lb (#2469)v27.1.0Compare Source
countis missing,t()uses the_otherform of the same language before the next fallback language, as CLDR does (lateral inheritance), instead of returning the key. Current CLDR data gives French, Spanish, Italian and Portuguese amanyform for exact millions, so files with only_oneand_othershowed the key (or the fallback language) forcount: 1000000. The plain key still comes first, sokey+key_otherkeeps resolvingkey; with a context,key_<context>_othercomes beforekey_other; ordinals use_ordinal_other; anddefaultValue_otheris used afterdefaultValue. A form filled in from_otheris no longer reported as missing (saveMissing,missingKeyHandler,exists()). Thanks @TenPotatoes (#2468)v27.0.1Compare Source
cacheInBuiltFormats: true, and formats added withaddCached. The cache key holds allt()options, including the interpolated values, so every new amount, date or name added a formatter that was never released: about 50 MB per 100k values, growing without end in a long-running server. A dropped formatter is made again from the same language and options, so the output does not change. Thanks @liustack (#2467).v27.0.0Compare Source
This is a major breaking release. It changes defaults that most setups had to override, gives every instance its own logger and post-processors, refuses language codes that look like markup or paths, and removes options that were broken or had no effect. 26.7 warned in debug mode about most of the runtime changes. The migration guide has the details; to keep the v26 behavior where an option exists:
Changed defaults
fallbackLngdefaults tofalseinstead of['dev']. Without afallbackLng, i18next no longer requestsdevfrom the backend (/locales/dev/translation.jsonwith i18next-http-backend), andsaveMissingsends missing keys for the current language instead ofdev. WithsupportedLngsand nofallbackLng, an unsupported language leaveslanguageandlanguagesunset, asfallbackLng: falsealready did. SetfallbackLngto your main language, or to'dev'for the previous behavior.returnEmptyStringdefaults tofalse: an empty translation falls back like a missing value (to the fallback language, the default value or the key), asnullalready does. i18next-cli writes''for untranslated keys in the other languages, which now fall back instead of rendering blank. An explicit''default from code still counts:t('key', '')andt('key', { defaultValue: '' })return''. SetreturnEmptyString: true(ininit()and inCustomTypeOptions) if an empty translation is meant to render nothing, for example for a Locize project with "preserve empty strings" on.'EN'becomes'en'. Only the casing changes: aliases liketl,iworengstay as they are ('TL'becomes'tl'), and so do codes that are not BCP 47 (en_US,my_custom). i18next 26.7.0 and 26.7.1 warned that v27 turnsiwintoheortlintofil; that was wrong and 26.7.2 corrects it.iw,tland the like stay as they are. SetcleanCode: falseto keep codes without a region exactly as they are.cleanCodeis deprecated.strictInterpolationChecksdefaults totrue:t()needs the options for a string with variables, so a{{name}}cannot be left out of a call. Type a wrapper aroundt()withParseKeysWithoutVariables, declare the variablesinterpolation.defaultVariablesfills in withdefaultVariablesinCustomTypeOptions, or setstrictInterpolationChecks: falseto keep the v26 types (also for ICU and MF2 messages, whose arguments are required either way). The fixes below close the gap plural keys left.returnEmptyStringdefaults tofalseinTypeOptionstoo, so a key whose value is''is typed as the key.Changed behavior
use()of a logger or a post-processor on one instance no longer changes the others, and theinit()of one instance no longer rewires the logging of all of them. An instance fromcreateInstance()does not see what the defaulti18nextinstance (or any other) registered: calluse()on each instance. A clone (cloneInstance()) keeps its parent's logger and post-processors, also those its parent registers later, and what it registers itself stays its own. A plugin that callsuse()from its owninit()(the Locize InContext editor) registers right away, as before.services.loggerstays the instance's logger, withforward(). In debug mode, apostProcessname that is not registered on the instance logs a warning.returnObjectsis handled like a missing key:t()returns the default value if one is passed, else the key (withappendNamespaceToMissingKey,parseMissingKeyHandlerand the post-processors applied), instead of the sentencekey '... (en)' returned an object instead of string., which ended up in the UI. It is not saved as missing, since the key exists. The debug warning stays, andreturnedObjectHandlerstill decides what is returned when it is set (returnedObjectHandler: (key) => `key '${key}' returned an object instead of string.`brings a sentence back).addResources()keeps every value: numbers, booleans and objects were dropped silently, only strings and arrays were added./:'$10/month'stays'$10/month'instead of'$10/month', and dates in react-i18next'sTransno longer render as17/12/2025(discussion 2199).&,<,>,"and'are still escaped, which covers HTML text and quoted attributes; a/cannot end either. Update snapshots that contain/, or pass your owninterpolation.escapeto keep escaping it.lng,changeLanguage(), a language detector orloadLanguages()that does not have the shape ofisSafeLanguageCode()(letters, digits,-,_,@and., no.., at most 128 characters) is refused unless it is configured: listed insupportedLngs, in afallbackLngthat is not a function or inpreload, or a language inresources, as given toinit().changeLanguage()continues as with an unsupported code: with the detector's next code, else the fallback language, else none (withfallbackLng: false, a code refused atinit()leaveslanguageunset);loadLanguages()skips it. WithsupportedLngs, the input is matched as before ('de-DE;q=0.9'still gives'de'), and only an unsafe match is refused, which onlynonExplicitSupportedLngscan return ('de-<x>'for a supported'de'). WithoutsupportedLngs, a detector that returned"><img src=x onerror=alert(1)>made that stringi18next.language, which apps put into<html lang>, and backends were asked to load it. BCP 47, POSIX (en_US.UTF-8@euro,sr@latin) and private-use codes (en-x-pirate) pass; to use a code with other characters (+, spaces,/), list it insupportedLngs. In debug mode a refused code logs a warning.createInstance(options)merges the defaults right away, without initializing: beforeinit(),instance.optionshas the defaults,getFixedT()no longer throws (t()returns''untilinit()), anddebug: truealready shows the warnings beforeinit().init()still tells the options you set from the defaults, so akeySeparatorornsSeparatorpassed tocreateInstance()counts as set by you (#2426).loadNamespaces()beforeinit()does nothing and calls back, as it did for an instance without options.keySeparator: falsemeans flat keys only: a lookup no longer splits a key at the textfalse(t('nestedfalsec')returned the value ofnested.c), and a selector path is joined with.like a key prefix, so selectors work on the flat keys ($ => $['a.b']and$ => $.a.blook up the flat key'a.b';$ => $.nested.cno longer resolvesnested, thenc). types: withkeySeparator: falseinCustomTypeOptions, the keys are the flat keys (nonestedfalseckeys), a key containingfalseis no longer split, andkeyPrefixtakes the parts of the flat keys before a., which is howgetFixedT()joins it ('nested'for'nested.c'). A key prefix combined with selectors is not typed for flat keys.Removed
getResource('en.translation.key'),addResource('en.translation.key', value)andaddResourceBundle('en.translation', resources). Pass language, namespace and key as separate arguments;addResource(),addResources()andaddResourceBundle()throw aTypeErrorfor the dotted form. A language code with a.(de_DE.UTF-8) is now a language like any other.interpolation.prefixEscaped,suffixEscaped,nestingPrefixEscapedandnestingSuffixEscaped, which were inserted into the RegExp unescaped. Pass the literal delimiters asprefix,suffix,nestingPrefixandnestingSuffix.compatibilityJSON(it had no effect since v24).locizeLastUsedis declared by locize-lastused (5.1.0 and later).ReactOptionsstays importable from'i18next'but is empty: react-i18next fills it in (17.1.13 and later, and 18), so thereactoptions are checked when react-i18next is in the program._and$helpers intypescript/t.d.tsandtypescript/options.d.ts($FirstNamespace,_ValueOfand 84 more) are no longer importable; every declaration of those files counted as exported. The other types of those files (TOptionsBase,TFunctionSignature,ParseTReturn, ...) are now exported on purpose. If you imported one of the hidden helpers, please open an issue: exporting it again does not break anything.index.js, a CommonJS shim that theexportsmap made unreachable.Support
engines); CI tests 22, 24 and 26.defaults, which moved with every update of the browser data. The output is the same as 26.7's.^5 || ^6 || ^7, unchanged.Kept
callbackApistill work; i18next guesses their calling convention from the number of parameters and logs a deprecation warning in debug mode.isInitialized,initializedStoreOnceandinitializedLanguageOncestay writable, so react-i18next 17'suseSSRand next-i18next's Pages Router work with 27. Writing them is deprecated: callhydrate().services.backendConnector.backendstays readable.Type fixes for
strictInterpolationChecksstrictInterpolationChecks, a plural key no longer switches the check off. Withoutcounta plural key likeitem(item_one,item_other) has no value of its own, and thestringit resolved to hid the variables of every other key in a union:(key: 'greeting' | 'item') => t(key)and(key: ParseKeys<'ns'>) => t(key)compiled as soon as the namespace had a plural key, while the same wrapper without one failed. Now such a call is checked against the plural forms, ordinal ones and'ns:item'keys included:t('item')needscountwhen the forms use it, and a wrapper over a union with a plural key fails as it does without one. To migrate, passcount, or type the wrapper withParseKeysWithoutVariables.strictInterpolationChecks, a call with options needs the variables of a plural key's forms too:t('item', { defaultValue: 'x' })and(key: 'plain' | 'item') => t(key, { defaultValue: 'x' })needcountwhen the forms use it, ast('item')does. Before, any options object turned the check off for a plural key withoutcount. Without the flag the options types are unchanged.ParseKeysWithoutVariablesleaves out a key whose ordinal forms takecount, and keeps a key that has its own value next to its plural forms, sincet()returns that value withoutcount.t('ns:item')for a plural key withoutcountreturnsstring, ast('item')does, instead ofnever.i18next/react-i18next (react-i18next)
v18.0.1Compare Source
@types/reactas an optional peer dependency. The type declarations import fromreact, and with pnpm's global virtual store (virtualStoreType: global) TypeScript resolved that to the untypedreact/index.js, so every React type in them becameany: wrong props compiled, callback parameters failed with TS7006. Under the default pnpm layout it only worked through pnpm's hidden hoisting. The range is*, so projects without TypeScript install nothing extra and an older@types/reactdoes not make npm fail (#1948)>= 27.0.1), which keeps at most 1000 formatters in each cached formatv18.0.0Compare Source
Released together with i18next 27. The migration guide lists every change with what to do: https://react.i18next.com/latest/migrating-v17-to-v18
Breaking changes
>= 27.0.0) and React 18 or later (peer>= 18.0.0). React 16 and 17 apps stay on react-i18next 17: on i18next 27, use react-i18next 17.1.13 or later (it brings the types of thereactinit options), or 18.useSyncExternalStorecomes from React itself, souse-sync-external-storeis no longer a dependency.Transoutput no longer depends oninterpolation.escapeValue: it always decodes HTML entities after parsing, asIcuTransdoes, in text and in attribute values, both withreact.unescape(attribute values used the built-in decoder before). With i18next's defaultescapeValue: true, a valueTom & Jerryrendered asTom & Jerryand a date17/12/2025as17/12/2025; both now render as written, and so do entities in the translation itself (&,<, , ...). Values still render as text, never as markup (since 17.1.10). TheshouldUnescapeprop andreact.transDefaultProps.shouldUnescapeare ignored and warn (DEPRECATED_OPTION, in development and withdebug: true); remove them. Snapshot tests that contain entities insideTransoutput need an update (#1903, #1893).TransandIcuTransre-render on language change by themselves: they subscribe to thereact.bindI18n(defaultlanguageChanged) andreact.bindI18nStoreevents, likeuseTranslation. Before, they only updated when a parent re-rendered (#1710). They still neither load translations nor suspend; with atprop, thattstill decides the language.TransWithoutContextandIcuTransWithoutContextstay hook-free.reactoptions of an instance stay on that instance.initReactI18nextcopied them into module-wide defaults, so a second instance withoutreactoptions inherited, for example, the first one'suseSuspense: false. Set them on every instance, or callsetDefaults()once.i18nreturned byuseTranslation(and passed bywithTranslationandTranslation) forwards reads and writes to the instance. It was a copy taken when the language changed, so afterawait i18n.changeLanguage('de')the copy still saiden, and properties set on it never reached the instance. It is still a new object whenever the language changes, for dependency arrays and the React Compiler. Because the old object now reads the current language too, a check likeprevProps.i18n.language !== this.props.i18n.language(or the language of ausePrevious(i18n)) no longer sees a change: compare the objects (prevProps.i18n !== this.props.i18n) or keep the previous language yourself.useSSRandwithSSRcall i18next'shydrate()instead of writingresourceStore.data,options.ns,initializedStoreOnce,isInitializedandinitializedLanguageOncethemselves. Same behavior with real instances, including the opt-out for cloned instances. Test mocks ofi18nused withuseSSR/withSSRnow need ahydratemethod, mocks that load beforeinit()needonce, and mocks used withTransorIcuTransneedonandoff, since both subscribe to events now (withbindI18nStoreset, alsostore.onandstore.off): a minimal mock like{ t, options, language }throwsi18n.on is not a function.react-i18next/icu.macroBabel macro (deprecated since 17.0.21) is removed, with the placeholder exportsdate,time,number,select,pluralandselectOrdinal. WriteIcuTransdirectly.HTMLAttributes['children']augmentation forallowObjectInHTMLChildrenmoved to the opt-inreact-i18next/augment-react. Default installs no longer change React's DOM types. WithallowObjectInHTMLChildren: true, addimport 'react-i18next/augment-react';next to yourCustomTypeOptions; with current@types/reactit needsskipLibCheck: true(#1928).I18nContextis typed{ i18n: i18n; defaultNS?: string | string[] } | undefined, as at runtime:useContext(I18nContext)isundefinedoutside anI18nextProvider.getI18n()keeps its type.reactinit options. i18next 27 keeps an emptyReactOptionsinterface that react-i18next fills in, soimport type { ReactOptions } from 'i18next'keeps working, andReactOptionsis also exported fromreact-i18next. ThedefaultTransParentdocs no longer claim a'div'default.i18next(load i18next's UMD build first) and shrinks from 71.3 KB to 26.2 KB minified. The AMD build (dist/amd) is removed; AMD loaders can use the UMD file.engines.nodeis>=22, and the build targets ES2020 syntax with a fixed browser list instead of browserslistdefaults.Other changes
react-serverexport condition forreact-i18next. In a React Server Component,import { Trans } from 'react-i18next'failed withcreateContext is not a function; it now resolves to an entry withTransandIcuTrans(the variants without React context),TransWithoutContext,IcuTransWithoutContext,nodesToString,initReactI18next,setI18n/getI18nandsetDefaults/getDefaults. Hooks, HOCs,Translation,I18nextProviderand the SSR helpers stay client-only, so importing them there is a build error. Without ani18nprop, the serverTransandIcuTransuse the global instance (the oneinitReactI18nextregistered), which is the wrong one with an instance per request: pass that instance asi18n(tis optional). Atalone is not enough, because atfromgetFixedTcarries no instance. In development and withdebug: truethey warn about it (GLOBAL_INSTANCE_ON_SERVER) (#1759, #1840).<0/>) with children of its own is no longer remounted on every render.Transcreated a new component type for it each time, so its state and effects were reset.cloneInstance(), for example one per request in i18next-http-middleware) share their parent's, as they share its logger, so they do not repeat them.init()usesi18n.once('initialized')instead of removing its listener with a delay.keySeparator: falseand a key prefix (useTranslation(ns, { keyPrefix }),getFixedT), thevaluesofTransandIcuTransare typed from the right key. The prefix and the key were joined with the textfalse(nestedfalsec), so no variables were checked; they are now joined with., asgetFixedTdoes.useSSRtakes its third argument{ i18n }, and theuseTranslationoptions takebindI18nStore, as at runtime.ErrorCodelistsI18N_NOT_INITIALIZED,ICU_TRANS_RENDER_ERRORandGLOBAL_INSTANCE_ON_SERVER.fileswhitelist replaces.npmignore;@babel/runtimeis no longer a dependency.Configuration
📅 Schedule: (in timezone Europe/Copenhagen)
🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.
♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.
👻 Immortal: This PR will be recreated if closed unmerged. Get config help if that's undesired.
This PR has been generated by Mend Renovate CLI.