import{type LyraFlagUrlResolver}from'./flag.js';
/**
* Register the bulk-resolution `@aceshooting/lyra-flags` resolver through that package's
* tier-committed `./standard` entry — the pairing of `flag-peer-bulk.js`'s one-shared-fetch
* batching with `./standard`'s one-tier module graph. Import this entry **instead of**
* `flag-peer.js` or `flag-peer-bulk.js`, never alongside either: each `setFlagUrlResolver()` call
* simply replaces the previous resolver, so importing two entries just makes whichever loaded last
* win.
*
* Use it when both conditions hold: the page renders most or all flags at once (a country table, a
* full locale picker), *and* every `<lr-flag>` stays on the default `fidelity="standard"`.
* `flag-peer-bulk.js` reaches the same batching through the peer package's root entry, which
* statically imports all three fidelity tiers' generated loader maps so its `flagUrl()` can honour
* a per-call variant — so a bundler emits the detailed and compact tiers' whole lazy-chunk graph
* even though bulk resolution only ever reads the standard-tier eager map. On a real production
* build with a 156-country flag column, that difference was +15.8MB of emitted assets (+65 detailed
* SVGs, +31 compact WebPs) that no route rendered.
*
* The tradeoff this entry makes in exchange: it is committed to one tier, so `fidelity="compact"`
* or `fidelity="detailed"` on an individual element resolves to that code's *standard* asset
* rather than reaching for a tier the registered resolver deliberately does not ship. That is a
* silent visual no-op, not an error — for per-instance fidelity, import `flag-peer-bulk.js` (bulk,
* all three tiers) or `flag-peer.js` (per-code lazy, all three tiers) instead.
*/
export declare function registerLyraFlagStandardBulkPeer():Promise<LyraFlagUrlResolver|null>;