//#region src/scan/parseHtml.d.ts
/**
* Tiny dependency-free HTML extraction helpers.
*
* The hosted backend audit relies on Cheerio + a real browser, but the CLI scan
* must stay dependency-light. These regex-based helpers cover the handful of
* head/anchor signals the score needs. They are intentionally forgiving: when a
* tag can't be parsed it is simply skipped rather than throwing.
*/
/** Compute the UTF-8 byte length of a string in both Node and browser builds. */
export declare const byteLength: (text: string) => number;
/** Extract the `lang` attribute of the `` element, if present. */
export declare const extractHtmlLang: (html: string) => string | undefined;
/** Extract the `dir` attribute of the `` element, if present. */
export declare const extractHtmlDir: (html: string) => string | undefined;
/** Extract the document `` text. */
export declare const extractTitle: (html: string) => string;
/** Extract the `<meta name="description">` content. */
export declare const extractMetaDescription: (html: string) => string;
/** Extract the `<meta property="og:image">` content. */
export declare const extractOgImage: (html: string) => string | undefined;
/** Whether a `<link rel="canonical">` element is present. */
export declare const hasCanonical: (html: string) => boolean;
/** A parsed `<link rel="alternate" hreflang>` element. */
export type HreflangLink = {
hreflang: string;
href: string;
};
/** Extract every `<link rel="alternate" hreflang="…" href="…">` element. */
export declare const extractHreflangs: (html: string) => HreflangLink[];
/**
* Extract every eagerly-loaded script URL: `<script src>`,
* `<link rel="modulepreload">` and `<link rel="preload" as="script">`.
*
* @param html - The raw HTML document.
* @param baseUrl - Base URL used to resolve relative script URLs.
* @returns Absolute, de-duplicated script URLs.
*/
export declare const extractScriptUrls: (html: string, baseUrl: string) => string[];
/** A parsed `<a href>` anchor. */
export type Anchor = {
href: string;
text: string;
};
/** Extract every `<a href="…">text</a>` anchor from the document. */
export declare const extractAnchors: (html: string) => Anchor[];
/**
* Extract visible text snippets from an HTML document (scripts, styles and
* tags stripped). Used to approximate the rendered content size without a DOM.
*/
export declare const extractVisibleTextStrings: (html: string) => string[];
//#endregion
//# sourceMappingURL=parseHtml.d.ts.map