@adonis-agora/agent/dist/src/rag-media/text-extractor.d.ts

Governed, durable-backed AI agent (chat + tool-calling + governance) for AdonisJS — part of the Agora ecosystem.

/**
 * Turns a media file's raw bytes into the plain text that gets chunked, embedded, and indexed. The
 * extraction seam — so the RAG stack itself stays format-agnostic. Ships the text-family extractors;
 * bring your own for PDF/DOCX/etc. by registering a parser on a {@link MimeTextExtractor} (so no PDF
 * library is ever a hard dependency of `@adonis-agora/agent`). Mirrors the reference `rag-media`
 * `TextExtractor` contract exactly.
 */
export interface TextExtractor {
    extract(bytes: Buffer, contentType: string): Promise<string>;
}
/** Thrown when no extractor is registered for a content type — ingestion treats it as "skip, don't index". */
export declare class UnsupportedContentTypeError extends Error {
    readonly contentType: string;
    constructor(contentType: string);
}
/** How a single content type (or `type/*` family) turns bytes into text. */
export type ExtractFn = (bytes: Buffer, contentType: string) => string | Promise<string>;
/**
 * A {@link TextExtractor} that dispatches by content type. Register exact types (`application/json`) or
 * a whole family (`text/*`); an exact match always wins over a family. Unregistered types throw
 * {@link UnsupportedContentTypeError}. This is the extension point for binary formats:
 * `.register('application/pdf', pdfFn)` — the PDF/DOCX library stays entirely on the host side.
 */
export declare class MimeTextExtractor implements TextExtractor {
    private readonly exact;
    private readonly families;
    register(contentType: string, fn: ExtractFn): this;
    extract(bytes: Buffer, contentType: string): Promise<string>;
    /** Whether a content type would resolve to an extractor (without reading bytes). */
    supports(contentType: string): boolean;
    private resolve;
}
/** Strip any `; charset=…` parameter off a content type and lower-case it. */
export declare function normalizeContentType(contentType: string): string;
/** Decode bytes as UTF-8 — the extractor for `text/*` and JSON. */
export declare function decodeUtf8(bytes: Buffer): string;
/** Decode UTF-8 then strip HTML: drop `<script>`/`<style>` bodies and tags, decode common entities. */
export declare function extractHtmlText(bytes: Buffer): string;
/**
 * The default extractor: `text/plain`, `text/csv`, `text/markdown` (and the whole `text/*` family) plus
 * `application/json` decode as UTF-8, `text/html` is tag-stripped, and everything else — PDF, DOCX,
 * images — throws {@link UnsupportedContentTypeError} (so binary formats are skipped rather than indexed
 * as garbage). Extend it for those: `.register('application/pdf', myPdfExtractor)`.
 */
export declare function defaultTextExtractor(): MimeTextExtractor;
//# sourceMappingURL=text-extractor.d.ts.map