localrag 0.1.0

package/dist/embedding/embedding-service.d.ts

@@ -0,0 +1,22 @@
+/**
+ * EmbeddingService
+ *
+ * Wraps @xenova/transformers (all-MiniLM-L6-v2) for local, normalized
+ * sentence embeddings. Uses a dynamic import because transformers is ESM-only
+ * while the rest of the package targets CommonJS.
+ *
+ * The model (~25 MB ONNX) is cached automatically in ~/.cache/huggingface
+ * after the first download. A progress spinner is shown on first load.
+ */
+export declare class EmbeddingService {
+    /**
+     * Generate normalized L2 embeddings for one or more texts.
+     * Returns a 2D array: one 384-dim vector per input text.
+     */
+    embed(texts: string[]): Promise<number[][]>;
+    /**
+     * Embed a single text, returning a 384-dim number[].
+     */
+    embedOne(text: string): Promise<number[]>;
+}
+//# sourceMappingURL=embedding-service.d.ts.map

package/dist/embedding/embedding-service.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"embedding-service.d.ts","sourceRoot":"","sources":["../../src/embedding/embedding-service.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AAwCH,qBAAa,gBAAgB;IAC3B;;;OAGG;IACG,KAAK,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC;IAajD;;OAEG;IACG,QAAQ,CAAC,IAAI,EAAE,MAAM,GAAG,OAAO,CAAC,MAAM,EAAE,CAAC;CAIhD"}

package/dist/embedding/embedding-service.js

@@ -0,0 +1,99 @@
+"use strict";
+/**
+ * EmbeddingService
+ *
+ * Wraps @xenova/transformers (all-MiniLM-L6-v2) for local, normalized
+ * sentence embeddings. Uses a dynamic import because transformers is ESM-only
+ * while the rest of the package targets CommonJS.
+ *
+ * The model (~25 MB ONNX) is cached automatically in ~/.cache/huggingface
+ * after the first download. A progress spinner is shown on first load.
+ */
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.EmbeddingService = void 0;
+// eslint-disable-next-line @typescript-eslint/no-require-imports
+const ora = require('ora');
+const VECTOR_DIM = 384;
+// eslint-disable-next-line @typescript-eslint/no-explicit-any
+let pipelineInstance = null;
+async function getPipeline() {
+    if (pipelineInstance)
+        return pipelineInstance;
+    const spinner = ora({
+        text: 'Loading embedding model (first run downloads ~25 MB — one-time only)…',
+        spinner: 'dots',
+    }).start();
+    try {
+        // Dynamic import: @xenova/transformers is ESM; works fine from CJS via import()
+        const { pipeline, env } = await Promise.resolve().then(() => __importStar(require('@xenova/transformers')));
+        // Allow downloading from HuggingFace Hub; cache locally forever
+        env.allowLocalModels = true;
+        env.useBrowserCache = false;
+        pipelineInstance = await pipeline('feature-extraction', 'Xenova/all-MiniLM-L6-v2', { quantized: true });
+        spinner.succeed('Embedding model ready');
+        return pipelineInstance;
+    }
+    catch (err) {
+        spinner.fail('Failed to load embedding model');
+        throw err;
+    }
+}
+class EmbeddingService {
+    /**
+     * Generate normalized L2 embeddings for one or more texts.
+     * Returns a 2D array: one 384-dim vector per input text.
+     */
+    async embed(texts) {
+        if (texts.length === 0)
+            return [];
+        const pipe = await getPipeline();
+        const output = await pipe(texts, { pooling: 'mean', normalize: true });
+        // Output is a Tensor with shape [texts.length, 384].
+        // .tolist() converts to nested JS arrays.
+        // eslint-disable-next-line @typescript-eslint/no-explicit-any
+        const nested = output.tolist();
+        return nested;
+    }
+    /**
+     * Embed a single text, returning a 384-dim number[].
+     */
+    async embedOne(text) {
+        const results = await this.embed([text]);
+        return results[0] ?? new Array(VECTOR_DIM).fill(0);
+    }
+}
+exports.EmbeddingService = EmbeddingService;
+//# sourceMappingURL=embedding-service.js.map

package/dist/embedding/embedding-service.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"embedding-service.js","sourceRoot":"","sources":["../../src/embedding/embedding-service.ts"],"names":[],"mappings":";AAAA;;;;;;;;;GASG;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAEH,iEAAiE;AACjE,MAAM,GAAG,GAAG,OAAO,CAAC,KAAK,CAAC,CAAC;AAE3B,MAAM,UAAU,GAAG,GAAG,CAAC;AAEvB,8DAA8D;AAC9D,IAAI,gBAAgB,GAAQ,IAAI,CAAC;AAEjC,KAAK,UAAU,WAAW;IACxB,IAAI,gBAAgB;QAAE,OAAO,gBAAgB,CAAC;IAE9C,MAAM,OAAO,GAAG,GAAG,CAAC;QAClB,IAAI,EAAE,uEAAuE;QAC7E,OAAO,EAAE,MAAM;KAChB,CAAC,CAAC,KAAK,EAAE,CAAC;IAEX,IAAI,CAAC;QACH,gFAAgF;QAChF,MAAM,EAAE,QAAQ,EAAE,GAAG,EAAE,GAAG,wDAAa,sBAAsB,GAAC,CAAC;QAE/D,gEAAgE;QAChE,GAAG,CAAC,gBAAgB,GAAG,IAAI,CAAC;QAC5B,GAAG,CAAC,eAAe,GAAG,KAAK,CAAC;QAE5B,gBAAgB,GAAG,MAAM,QAAQ,CAC/B,oBAAoB,EACpB,yBAAyB,EACzB,EAAE,SAAS,EAAE,IAAI,EAAE,CACpB,CAAC;QAEF,OAAO,CAAC,OAAO,CAAC,uBAAuB,CAAC,CAAC;QACzC,OAAO,gBAAgB,CAAC;IAC1B,CAAC;IAAC,OAAO,GAAG,EAAE,CAAC;QACb,OAAO,CAAC,IAAI,CAAC,gCAAgC,CAAC,CAAC;QAC/C,MAAM,GAAG,CAAC;IACZ,CAAC;AACH,CAAC;AAED,MAAa,gBAAgB;IAC3B;;;OAGG;IACH,KAAK,CAAC,KAAK,CAAC,KAAe;QACzB,IAAI,KAAK,CAAC,MAAM,KAAK,CAAC;YAAE,OAAO,EAAE,CAAC;QAElC,MAAM,IAAI,GAAG,MAAM,WAAW,EAAE,CAAC;QACjC,MAAM,MAAM,GAAG,MAAM,IAAI,CAAC,KAAK,EAAE,EAAE,OAAO,EAAE,MAAM,EAAE,SAAS,EAAE,IAAI,EAAE,CAAC,CAAC;QAEvE,qDAAqD;QACrD,0CAA0C;QAC1C,8DAA8D;QAC9D,MAAM,MAAM,GAAgB,MAAc,CAAC,MAAM,EAAE,CAAC;QACpD,OAAO,MAAM,CAAC;IAChB,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,QAAQ,CAAC,IAAY;QACzB,MAAM,OAAO,GAAG,MAAM,IAAI,CAAC,KAAK,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC;QACzC,OAAO,OAAO,CAAC,CAAC,CAAC,IAAI,IAAI,KAAK,CAAC,UAAU,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;IACrD,CAAC;CACF;AAzBD,4CAyBC"}

package/dist/extractors/docx-extractor.d.ts

@@ -0,0 +1,12 @@
+import { Extractor, ExtractedChunk } from './extractor.interface';
+/**
+ * DOCX extractor using mammoth.
+ *
+ * mammoth.extractRawText() strips all formatting and returns clean plain text,
+ * which is ideal for embedding. Word documents don't have a reliable page-
+ * number concept in the file format, so page is left undefined.
+ */
+export declare class DocxExtractor implements Extractor {
+    extract(filePath: string): Promise<ExtractedChunk[]>;
+}
+//# sourceMappingURL=docx-extractor.d.ts.map

package/dist/extractors/docx-extractor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"docx-extractor.d.ts","sourceRoot":"","sources":["../../src/extractors/docx-extractor.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAElE;;;;;;GAMG;AACH,qBAAa,aAAc,YAAW,SAAS;IACvC,OAAO,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,cAAc,EAAE,CAAC;CAoB3D"}

package/dist/extractors/docx-extractor.js

@@ -0,0 +1,29 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.DocxExtractor = void 0;
+/**
+ * DOCX extractor using mammoth.
+ *
+ * mammoth.extractRawText() strips all formatting and returns clean plain text,
+ * which is ideal for embedding. Word documents don't have a reliable page-
+ * number concept in the file format, so page is left undefined.
+ */
+class DocxExtractor {
+    async extract(filePath) {
+        // eslint-disable-next-line @typescript-eslint/no-require-imports
+        const mammoth = require('mammoth');
+        const result = await mammoth.extractRawText({ path: filePath });
+        if (result.messages && result.messages.length > 0) {
+            const errors = result.messages.filter((m) => m.type === 'error');
+            if (errors.length > 0) {
+                throw new Error(`DOCX extraction errors in ${filePath}: ${errors.map((e) => e.message).join(', ')}`);
+            }
+        }
+        const text = result.value.trim();
+        if (!text)
+            return [];
+        return [{ text }];
+    }
+}
+exports.DocxExtractor = DocxExtractor;
+//# sourceMappingURL=docx-extractor.js.map

package/dist/extractors/docx-extractor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"docx-extractor.js","sourceRoot":"","sources":["../../src/extractors/docx-extractor.ts"],"names":[],"mappings":";;;AAEA;;;;;;GAMG;AACH,MAAa,aAAa;IACxB,KAAK,CAAC,OAAO,CAAC,QAAgB;QAC5B,iEAAiE;QACjE,MAAM,OAAO,GAAG,OAAO,CAAC,SAAS,CAAC,CAAC;QAEnC,MAAM,MAAM,GAAG,MAAM,OAAO,CAAC,cAAc,CAAC,EAAE,IAAI,EAAE,QAAQ,EAAE,CAAC,CAAC;QAEhE,IAAI,MAAM,CAAC,QAAQ,IAAI,MAAM,CAAC,QAAQ,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YAClD,MAAM,MAAM,GAAG,MAAM,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,CAAmB,EAAE,EAAE,CAAC,CAAC,CAAC,IAAI,KAAK,OAAO,CAAC,CAAC;YACnF,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACtB,MAAM,IAAI,KAAK,CACb,6BAA6B,QAAQ,KAAK,MAAM,CAAC,GAAG,CAAC,CAAC,CAAsB,EAAE,EAAE,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CACzG,CAAC;YACJ,CAAC;QACH,CAAC;QAED,MAAM,IAAI,GAAI,MAAM,CAAC,KAAgB,CAAC,IAAI,EAAE,CAAC;QAC7C,IAAI,CAAC,IAAI;YAAE,OAAO,EAAE,CAAC;QAErB,OAAO,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC;IACpB,CAAC;CACF;AArBD,sCAqBC"}

package/dist/extractors/extractor.interface.d.ts

@@ -0,0 +1,14 @@
+/** A unit of text extracted from a source document, optionally carrying a page number. */
+export interface ExtractedChunk {
+    text: string;
+    page?: number;
+}
+/** Common interface all file-type extractors must implement. */
+export interface Extractor {
+    extract(filePath: string): Promise<ExtractedChunk[]>;
+}
+export declare function isSupportedFile(filePath: string): boolean;
+export declare function getSupportedExtensions(): string[];
+/** Returns the appropriate extractor for a given file path, or null if unsupported. */
+export declare function getExtractor(filePath: string): Promise<Extractor | null>;
+//# sourceMappingURL=extractor.interface.d.ts.map

package/dist/extractors/extractor.interface.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"extractor.interface.d.ts","sourceRoot":"","sources":["../../src/extractors/extractor.interface.ts"],"names":[],"mappings":"AAIA,0FAA0F;AAC1F,MAAM,WAAW,cAAc;IAC7B,IAAI,EAAE,MAAM,CAAC;IACb,IAAI,CAAC,EAAE,MAAM,CAAC;CACf;AAED,gEAAgE;AAChE,MAAM,WAAW,SAAS;IACxB,OAAO,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,cAAc,EAAE,CAAC,CAAC;CACtD;AAaD,wBAAgB,eAAe,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAEzD;AAED,wBAAgB,sBAAsB,IAAI,MAAM,EAAE,CAEjD;AAED,uFAAuF;AACvF,wBAAsB,YAAY,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,SAAS,GAAG,IAAI,CAAC,CAK9E"}

package/dist/extractors/extractor.interface.js

@@ -0,0 +1,63 @@
+"use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.isSupportedFile = isSupportedFile;
+exports.getSupportedExtensions = getSupportedExtensions;
+exports.getExtractor = getExtractor;
+const path = __importStar(require("path"));
+// ── Supported extensions ─────────────────────────────────────────────────
+const SUPPORTED = {
+    '.txt': async () => { const { TextExtractor } = await Promise.resolve().then(() => __importStar(require('./text-extractor'))); return new TextExtractor(); },
+    '.md': async () => { const { TextExtractor } = await Promise.resolve().then(() => __importStar(require('./text-extractor'))); return new TextExtractor(); },
+    '.pdf': async () => { const { PdfExtractor } = await Promise.resolve().then(() => __importStar(require('./pdf-extractor'))); return new PdfExtractor(); },
+    '.docx': async () => { const { DocxExtractor } = await Promise.resolve().then(() => __importStar(require('./docx-extractor'))); return new DocxExtractor(); },
+    '.xlsx': async () => { const { XlsxExtractor } = await Promise.resolve().then(() => __importStar(require('./xlsx-extractor'))); return new XlsxExtractor(); },
+    '.pptx': async () => { const { PptxExtractor } = await Promise.resolve().then(() => __importStar(require('./pptx-extractor'))); return new PptxExtractor(); },
+};
+function isSupportedFile(filePath) {
+    return path.extname(filePath).toLowerCase() in SUPPORTED;
+}
+function getSupportedExtensions() {
+    return Object.keys(SUPPORTED);
+}
+/** Returns the appropriate extractor for a given file path, or null if unsupported. */
+async function getExtractor(filePath) {
+    const ext = path.extname(filePath).toLowerCase();
+    const factory = SUPPORTED[ext];
+    if (!factory)
+        return null;
+    return factory();
+}
+//# sourceMappingURL=extractor.interface.js.map

package/dist/extractors/extractor.interface.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"extractor.interface.js","sourceRoot":"","sources":["../../src/extractors/extractor.interface.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AA0BA,0CAEC;AAED,wDAEC;AAGD,oCAKC;AAxCD,2CAA6B;AAe7B,4EAA4E;AAE5E,MAAM,SAAS,GAA6C;IAC1D,MAAM,EAAE,KAAK,IAAI,EAAE,GAAG,MAAM,EAAE,aAAa,EAAE,GAAG,wDAAa,kBAAkB,GAAC,CAAC,CAAC,OAAO,IAAI,aAAa,EAAE,CAAC,CAAC,CAAC;IAC/G,KAAK,EAAG,KAAK,IAAI,EAAE,GAAG,MAAM,EAAE,aAAa,EAAE,GAAG,wDAAa,kBAAkB,GAAC,CAAC,CAAC,OAAO,IAAI,aAAa,EAAE,CAAC,CAAC,CAAC;IAC/G,MAAM,EAAE,KAAK,IAAI,EAAE,GAAG,MAAM,EAAE,YAAY,EAAE,GAAI,wDAAa,iBAAiB,GAAC,CAAC,CAAE,OAAO,IAAI,YAAY,EAAE,CAAC,CAAE,CAAC;IAC/G,OAAO,EAAC,KAAK,IAAI,EAAE,GAAG,MAAM,EAAE,aAAa,EAAE,GAAG,wDAAa,kBAAkB,GAAC,CAAC,CAAC,OAAO,IAAI,aAAa,EAAE,CAAC,CAAC,CAAC;IAC/G,OAAO,EAAC,KAAK,IAAI,EAAE,GAAG,MAAM,EAAE,aAAa,EAAE,GAAG,wDAAa,kBAAkB,GAAC,CAAC,CAAC,OAAO,IAAI,aAAa,EAAE,CAAC,CAAC,CAAC;IAC/G,OAAO,EAAC,KAAK,IAAI,EAAE,GAAG,MAAM,EAAE,aAAa,EAAE,GAAG,wDAAa,kBAAkB,GAAC,CAAC,CAAC,OAAO,IAAI,aAAa,EAAE,CAAC,CAAC,CAAC;CAChH,CAAC;AAEF,SAAgB,eAAe,CAAC,QAAgB;IAC9C,OAAO,IAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,WAAW,EAAE,IAAI,SAAS,CAAC;AAC3D,CAAC;AAED,SAAgB,sBAAsB;IACpC,OAAO,MAAM,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;AAChC,CAAC;AAED,uFAAuF;AAChF,KAAK,UAAU,YAAY,CAAC,QAAgB;IACjD,MAAM,GAAG,GAAG,IAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,WAAW,EAAE,CAAC;IACjD,MAAM,OAAO,GAAG,SAAS,CAAC,GAAG,CAAC,CAAC;IAC/B,IAAI,CAAC,OAAO;QAAE,OAAO,IAAI,CAAC;IAC1B,OAAO,OAAO,EAAE,CAAC;AACnB,CAAC"}

package/dist/extractors/pdf-extractor.d.ts

@@ -0,0 +1,11 @@
+import { Extractor, ExtractedChunk } from './extractor.interface';
+/**
+ * PDF extractor using pdf-parse.
+ *
+ * Uses pdf-parse's pagerender hook to collect per-page text.
+ * Falls back to full-document text if per-page extraction fails.
+ */
+export declare class PdfExtractor implements Extractor {
+    extract(filePath: string): Promise<ExtractedChunk[]>;
+}
+//# sourceMappingURL=pdf-extractor.d.ts.map

package/dist/extractors/pdf-extractor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"pdf-extractor.d.ts","sourceRoot":"","sources":["../../src/extractors/pdf-extractor.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,SAAS,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAElE;;;;;GAKG;AACH,qBAAa,YAAa,YAAW,SAAS;IACtC,OAAO,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,cAAc,EAAE,CAAC;CA6C3D"}

package/dist/extractors/pdf-extractor.js

@@ -0,0 +1,89 @@
+"use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.PdfExtractor = void 0;
+const fs = __importStar(require("fs"));
+/**
+ * PDF extractor using pdf-parse.
+ *
+ * Uses pdf-parse's pagerender hook to collect per-page text.
+ * Falls back to full-document text if per-page extraction fails.
+ */
+class PdfExtractor {
+    async extract(filePath) {
+        // eslint-disable-next-line @typescript-eslint/no-require-imports
+        const pdfParse = require('pdf-parse');
+        const buffer = fs.readFileSync(filePath);
+        const pageTexts = [];
+        let pageIndex = 0;
+        try {
+            await pdfParse(buffer, {
+                // eslint-disable-next-line @typescript-eslint/no-explicit-any
+                pagerender: (pageData) => {
+                    const pageNum = ++pageIndex;
+                    return pageData
+                        .getTextContent()
+                        // eslint-disable-next-line @typescript-eslint/no-explicit-any
+                        .then((content) => {
+                        const text = content.items
+                            // eslint-disable-next-line @typescript-eslint/no-explicit-any
+                            .map((item) => item.str)
+                            .join(' ');
+                        pageTexts[pageNum - 1] = text;
+                        return text;
+                    })
+                        .catch(() => {
+                        pageTexts[pageNum - 1] = '';
+                        return '';
+                    });
+                },
+            });
+            const result = pageTexts
+                .map((text, idx) => ({ text: text.trim(), page: idx + 1 }))
+                .filter(c => c.text.length > 0);
+            return result.length > 0 ? result : [];
+        }
+        catch (err) {
+            const msg = err instanceof Error ? err.message : String(err);
+            // Surface common failure modes
+            if (msg.includes('encrypted') || msg.includes('password')) {
+                throw new Error(`PDF is encrypted/password-protected: ${filePath}`);
+            }
+            throw new Error(`Failed to parse PDF ${filePath}: ${msg}`);
+        }
+    }
+}
+exports.PdfExtractor = PdfExtractor;
+//# sourceMappingURL=pdf-extractor.js.map

package/dist/extractors/pdf-extractor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"pdf-extractor.js","sourceRoot":"","sources":["../../src/extractors/pdf-extractor.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA,uCAAyB;AAGzB;;;;;GAKG;AACH,MAAa,YAAY;IACvB,KAAK,CAAC,OAAO,CAAC,QAAgB;QAC5B,iEAAiE;QACjE,MAAM,QAAQ,GAAG,OAAO,CAAC,WAAW,CAAC,CAAC;QAEtC,MAAM,MAAM,GAAG,EAAE,CAAC,YAAY,CAAC,QAAQ,CAAC,CAAC;QACzC,MAAM,SAAS,GAAa,EAAE,CAAC;QAC/B,IAAI,SAAS,GAAG,CAAC,CAAC;QAElB,IAAI,CAAC;YACH,MAAM,QAAQ,CAAC,MAAM,EAAE;gBACrB,8DAA8D;gBAC9D,UAAU,EAAE,CAAC,QAAa,EAAmB,EAAE;oBAC7C,MAAM,OAAO,GAAG,EAAE,SAAS,CAAC;oBAC5B,OAAO,QAAQ;yBACZ,cAAc,EAAE;wBACjB,8DAA8D;yBAC7D,IAAI,CAAC,CAAC,OAAY,EAAE,EAAE;wBACrB,MAAM,IAAI,GAAW,OAAO,CAAC,KAAK;4BAChC,8DAA8D;6BAC7D,GAAG,CAAC,CAAC,IAAS,EAAE,EAAE,CAAC,IAAI,CAAC,GAAa,CAAC;6BACtC,IAAI,CAAC,GAAG,CAAC,CAAC;wBACb,SAAS,CAAC,OAAO,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC;wBAC9B,OAAO,IAAI,CAAC;oBACd,CAAC,CAAC;yBACD,KAAK,CAAC,GAAG,EAAE;wBACV,SAAS,CAAC,OAAO,GAAG,CAAC,CAAC,GAAG,EAAE,CAAC;wBAC5B,OAAO,EAAE,CAAC;oBACZ,CAAC,CAAC,CAAC;gBACP,CAAC;aACF,CAAC,CAAC;YAEH,MAAM,MAAM,GAAqB,SAAS;iBACvC,GAAG,CAAC,CAAC,IAAI,EAAE,GAAG,EAAE,EAAE,CAAC,CAAC,EAAE,IAAI,EAAE,IAAI,CAAC,IAAI,EAAE,EAAE,IAAI,EAAE,GAAG,GAAG,CAAC,EAAE,CAAC,CAAC;iBAC1D,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC,CAAC,CAAC,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;YAElC,OAAO,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC;QACzC,CAAC;QAAC,OAAO,GAAY,EAAE,CAAC;YACtB,MAAM,GAAG,GAAG,GAAG,YAAY,KAAK,CAAC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC;YAC7D,+BAA+B;YAC/B,IAAI,GAAG,CAAC,QAAQ,CAAC,WAAW,CAAC,IAAI,GAAG,CAAC,QAAQ,CAAC,UAAU,CAAC,EAAE,CAAC;gBAC1D,MAAM,IAAI,KAAK,CAAC,wCAAwC,QAAQ,EAAE,CAAC,CAAC;YACtE,CAAC;YACD,MAAM,IAAI,KAAK,CAAC,uBAAuB,QAAQ,KAAK,GAAG,EAAE,CAAC,CAAC;QAC7D,CAAC;IACH,CAAC;CACF;AA9CD,oCA8CC"}

package/dist/extractors/pptx-extractor.d.ts

@@ -0,0 +1,12 @@
+import { Extractor, ExtractedChunk } from './extractor.interface';
+/**
+ * PPTX extractor.
+ *
+ * PPTX files are ZIP archives whose slides live at ppt/slides/slideN.xml.
+ * We unzip the file with adm-zip and extract all <a:t> text nodes from
+ * each slide's XML using fast-xml-parser, treating each slide as a page.
+ */
+export declare class PptxExtractor implements Extractor {
+    extract(filePath: string): Promise<ExtractedChunk[]>;
+}
+//# sourceMappingURL=pptx-extractor.d.ts.map

package/dist/extractors/pptx-extractor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"pptx-extractor.d.ts","sourceRoot":"","sources":["../../src/extractors/pptx-extractor.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAElE;;;;;;GAMG;AACH,qBAAa,aAAc,YAAW,SAAS;IACvC,OAAO,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,cAAc,EAAE,CAAC;CAgD3D"}

package/dist/extractors/pptx-extractor.js

@@ -0,0 +1,98 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.PptxExtractor = void 0;
+/**
+ * PPTX extractor.
+ *
+ * PPTX files are ZIP archives whose slides live at ppt/slides/slideN.xml.
+ * We unzip the file with adm-zip and extract all <a:t> text nodes from
+ * each slide's XML using fast-xml-parser, treating each slide as a page.
+ */
+class PptxExtractor {
+    async extract(filePath) {
+        // eslint-disable-next-line @typescript-eslint/no-require-imports
+        const AdmZip = require('adm-zip');
+        // eslint-disable-next-line @typescript-eslint/no-require-imports
+        const { XMLParser } = require('fast-xml-parser');
+        const zip = new AdmZip(filePath);
+        const parser = new XMLParser({
+            ignoreAttributes: true,
+            parseTagValue: true,
+            parseAttributeValue: false,
+            trimValues: true,
+        });
+        // Collect slide entries sorted by slide number
+        const entries = zip
+            .getEntries()
+            .filter((e) => /^ppt\/slides\/slide(\d+)\.xml$/.test(e.entryName))
+            .map((e) => {
+            const m = e.entryName.match(/slide(\d+)\.xml$/);
+            return { name: e.entryName, index: m ? parseInt(m[1], 10) : 0 };
+        })
+            .sort((a, b) => a.index - b.index);
+        const chunks = [];
+        for (const { name, index } of entries) {
+            const entry = zip.getEntry(name);
+            if (!entry)
+                continue;
+            const xml = entry.getData().toString('utf-8');
+            let parsed;
+            try {
+                parsed = parser.parse(xml);
+            }
+            catch {
+                continue; // skip malformed slides
+            }
+            const textParts = collectTag(parsed, 'a:t');
+            const text = textParts.join(' ').replace(/\s+/g, ' ').trim();
+            if (text.length > 0) {
+                chunks.push({ text, page: index });
+            }
+        }
+        return chunks;
+    }
+}
+exports.PptxExtractor = PptxExtractor;
+// ── Helpers ───────────────────────────────────────────────────────────────
+/**
+ * Recursively collect all values of a given XML tag name from a parsed object.
+ */
+function collectTag(node, tag) {
+    if (node === null || node === undefined)
+        return [];
+    if (typeof node === 'string')
+        return [];
+    if (typeof node === 'number' || typeof node === 'boolean')
+        return [];
+    if (Array.isArray(node)) {
+        return node.flatMap(item => collectTag(item, tag));
+    }
+    if (typeof node === 'object') {
+        const results = [];
+        for (const [key, value] of Object.entries(node)) {
+            if (key === tag) {
+                if (typeof value === 'string')
+                    results.push(value);
+                else if (typeof value === 'number')
+                    results.push(String(value));
+                else if (Array.isArray(value)) {
+                    for (const v of value) {
+                        if (typeof v === 'string')
+                            results.push(v);
+                        else
+                            results.push(...collectTag(v, tag));
+                    }
+                }
+                else {
+                    results.push(...collectTag(value, tag));
+                }
+            }
+            else {
+                results.push(...collectTag(value, tag));
+            }
+        }
+        return results;
+    }
+    return [];
+}
+//# sourceMappingURL=pptx-extractor.js.map

package/dist/extractors/pptx-extractor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"pptx-extractor.js","sourceRoot":"","sources":["../../src/extractors/pptx-extractor.ts"],"names":[],"mappings":";;;AAEA;;;;;;GAMG;AACH,MAAa,aAAa;IACxB,KAAK,CAAC,OAAO,CAAC,QAAgB;QAC5B,iEAAiE;QACjE,MAAM,MAAM,GAAG,OAAO,CAAC,SAAS,CAAC,CAAC;QAClC,iEAAiE;QACjE,MAAM,EAAE,SAAS,EAAE,GAAG,OAAO,CAAC,iBAAiB,CAAC,CAAC;QAEjD,MAAM,GAAG,GAAG,IAAI,MAAM,CAAC,QAAQ,CAAC,CAAC;QACjC,MAAM,MAAM,GAAG,IAAI,SAAS,CAAC;YAC3B,gBAAgB,EAAE,IAAI;YACtB,aAAa,EAAE,IAAI;YACnB,mBAAmB,EAAE,KAAK;YAC1B,UAAU,EAAE,IAAI;SACjB,CAAC,CAAC;QAEH,+CAA+C;QAC/C,MAAM,OAAO,GAA2C,GAAG;aACxD,UAAU,EAAE;aACZ,MAAM,CAAC,CAAC,CAAwB,EAAE,EAAE,CAAC,gCAAgC,CAAC,IAAI,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC;aACxF,GAAG,CAAC,CAAC,CAAwB,EAAE,EAAE;YAChC,MAAM,CAAC,GAAG,CAAC,CAAC,SAAS,CAAC,KAAK,CAAC,kBAAkB,CAAC,CAAC;YAChD,OAAO,EAAE,IAAI,EAAE,CAAC,CAAC,SAAS,EAAE,KAAK,EAAE,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,EAAE,CAAC;QAClE,CAAC,CAAC;aACD,IAAI,CAAC,CAAC,CAAoB,EAAE,CAAoB,EAAE,EAAE,CAAC,CAAC,CAAC,KAAK,GAAG,CAAC,CAAC,KAAK,CAAC,CAAC;QAE3E,MAAM,MAAM,GAAqB,EAAE,CAAC;QAEpC,KAAK,MAAM,EAAE,IAAI,EAAE,KAAK,EAAE,IAAI,OAAO,EAAE,CAAC;YACtC,MAAM,KAAK,GAAG,GAAG,CAAC,QAAQ,CAAC,IAAI,CAAC,CAAC;YACjC,IAAI,CAAC,KAAK;gBAAE,SAAS;YAErB,MAAM,GAAG,GAAW,KAAK,CAAC,OAAO,EAAE,CAAC,QAAQ,CAAC,OAAO,CAAC,CAAC;YACtD,IAAI,MAAe,CAAC;YACpB,IAAI,CAAC;gBACH,MAAM,GAAG,MAAM,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;YAC7B,CAAC;YAAC,MAAM,CAAC;gBACP,SAAS,CAAC,wBAAwB;YACpC,CAAC;YAED,MAAM,SAAS,GAAG,UAAU,CAAC,MAAM,EAAE,KAAK,CAAC,CAAC;YAC5C,MAAM,IAAI,GAAG,SAAS,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC;YAE7D,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACpB,MAAM,CAAC,IAAI,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,CAAC,CAAC;YACrC,CAAC;QACH,CAAC;QAED,OAAO,MAAM,CAAC;IAChB,CAAC;CACF;AAjDD,sCAiDC;AAED,6EAA6E;AAE7E;;GAEG;AACH,SAAS,UAAU,CAAC,IAAa,EAAE,GAAW;IAC5C,IAAI,IAAI,KAAK,IAAI,IAAI,IAAI,KAAK,SAAS;QAAE,OAAO,EAAE,CAAC;IACnD,IAAI,OAAO,IAAI,KAAK,QAAQ;QAAE,OAAO,EAAE,CAAC;IACxC,IAAI,OAAO,IAAI,KAAK,QAAQ,IAAI,OAAO,IAAI,KAAK,SAAS;QAAE,OAAO,EAAE,CAAC;IAErE,IAAI,KAAK,CAAC,OAAO,CAAC,IAAI,CAAC,EAAE,CAAC;QACxB,OAAO,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC,EAAE,CAAC,UAAU,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;IACrD,CAAC;IAED,IAAI,OAAO,IAAI,KAAK,QAAQ,EAAE,CAAC;QAC7B,MAAM,OAAO,GAAa,EAAE,CAAC;QAC7B,KAAK,MAAM,CAAC,GAAG,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,IAA+B,CAAC,EAAE,CAAC;YAC3E,IAAI,GAAG,KAAK,GAAG,EAAE,CAAC;gBAChB,IAAI,OAAO,KAAK,KAAK,QAAQ;oBAAE,OAAO,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;qBAC9C,IAAI,OAAO,KAAK,KAAK,QAAQ;oBAAE,OAAO,CAAC,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,CAAC;qBAC3D,IAAI,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,EAAE,CAAC;oBAC9B,KAAK,MAAM,CAAC,IAAI,KAAK,EAAE,CAAC;wBACtB,IAAI,OAAO,CAAC,KAAK,QAAQ;4BAAE,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;;4BACtC,OAAO,CAAC,IAAI,CAAC,GAAG,UAAU,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC,CAAC;oBAC3C,CAAC;gBACH,CAAC;qBAAM,CAAC;oBACN,OAAO,CAAC,IAAI,CAAC,GAAG,UAAU,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,CAAC;gBAC1C,CAAC;YACH,CAAC;iBAAM,CAAC;gBACN,OAAO,CAAC,IAAI,CAAC,GAAG,UAAU,CAAC,KAAK,EAAE,GAAG,CAAC,CAAC,CAAC;YAC1C,CAAC;QACH,CAAC;QACD,OAAO,OAAO,CAAC;IACjB,CAAC;IAED,OAAO,EAAE,CAAC;AACZ,CAAC"}

package/dist/extractors/text-extractor.d.ts

@@ -0,0 +1,10 @@
+import { Extractor, ExtractedChunk } from './extractor.interface';
+/**
+ * Plain-text extractor for .txt and .md files.
+ * Returns the entire file content as a single chunk (the chunking service
+ * will split it into appropriately sized pieces).
+ */
+export declare class TextExtractor implements Extractor {
+    extract(filePath: string): Promise<ExtractedChunk[]>;
+}
+//# sourceMappingURL=text-extractor.d.ts.map

package/dist/extractors/text-extractor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"text-extractor.d.ts","sourceRoot":"","sources":["../../src/extractors/text-extractor.ts"],"names":[],"mappings":"AACA,OAAO,EAAE,SAAS,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAElE;;;;GAIG;AACH,qBAAa,aAAc,YAAW,SAAS;IACvC,OAAO,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,cAAc,EAAE,CAAC;CAK3D"}

package/dist/extractors/text-extractor.js

@@ -0,0 +1,52 @@
+"use strict";
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.TextExtractor = void 0;
+const fs = __importStar(require("fs"));
+/**
+ * Plain-text extractor for .txt and .md files.
+ * Returns the entire file content as a single chunk (the chunking service
+ * will split it into appropriately sized pieces).
+ */
+class TextExtractor {
+    async extract(filePath) {
+        const text = fs.readFileSync(filePath, 'utf-8');
+        if (!text.trim())
+            return [];
+        return [{ text }];
+    }
+}
+exports.TextExtractor = TextExtractor;
+//# sourceMappingURL=text-extractor.js.map

package/dist/extractors/text-extractor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"text-extractor.js","sourceRoot":"","sources":["../../src/extractors/text-extractor.ts"],"names":[],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA,uCAAyB;AAGzB;;;;GAIG;AACH,MAAa,aAAa;IACxB,KAAK,CAAC,OAAO,CAAC,QAAgB;QAC5B,MAAM,IAAI,GAAG,EAAE,CAAC,YAAY,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC;QAChD,IAAI,CAAC,IAAI,CAAC,IAAI,EAAE;YAAE,OAAO,EAAE,CAAC;QAC5B,OAAO,CAAC,EAAE,IAAI,EAAE,CAAC,CAAC;IACpB,CAAC;CACF;AAND,sCAMC"}

package/dist/extractors/xlsx-extractor.d.ts

@@ -0,0 +1,11 @@
+import { Extractor, ExtractedChunk } from './extractor.interface';
+/**
+ * XLSX extractor using the xlsx (SheetJS) package.
+ *
+ * Each worksheet is treated as a separate "page". The sheet content is
+ * serialised to CSV which captures all cell values in a readable format.
+ */
+export declare class XlsxExtractor implements Extractor {
+    extract(filePath: string): Promise<ExtractedChunk[]>;
+}
+//# sourceMappingURL=xlsx-extractor.d.ts.map

package/dist/extractors/xlsx-extractor.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"xlsx-extractor.d.ts","sourceRoot":"","sources":["../../src/extractors/xlsx-extractor.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,cAAc,EAAE,MAAM,uBAAuB,CAAC;AAElE;;;;;GAKG;AACH,qBAAa,aAAc,YAAW,SAAS;IACvC,OAAO,CAAC,QAAQ,EAAE,MAAM,GAAG,OAAO,CAAC,cAAc,EAAE,CAAC;CAkB3D"}

package/dist/extractors/xlsx-extractor.js

@@ -0,0 +1,28 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.XlsxExtractor = void 0;
+/**
+ * XLSX extractor using the xlsx (SheetJS) package.
+ *
+ * Each worksheet is treated as a separate "page". The sheet content is
+ * serialised to CSV which captures all cell values in a readable format.
+ */
+class XlsxExtractor {
+    async extract(filePath) {
+        // eslint-disable-next-line @typescript-eslint/no-require-imports
+        const XLSX = require('xlsx');
+        const workbook = XLSX.readFile(filePath, { type: 'file', cellText: true });
+        const chunks = [];
+        workbook.SheetNames.forEach((sheetName, idx) => {
+            const sheet = workbook.Sheets[sheetName];
+            const csv = XLSX.utils.sheet_to_csv(sheet, { strip: true });
+            const text = `Sheet: ${sheetName}\n${csv}`.trim();
+            if (text.length > 0) {
+                chunks.push({ text, page: idx + 1 });
+            }
+        });
+        return chunks;
+    }
+}
+exports.XlsxExtractor = XlsxExtractor;
+//# sourceMappingURL=xlsx-extractor.js.map

package/dist/extractors/xlsx-extractor.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"xlsx-extractor.js","sourceRoot":"","sources":["../../src/extractors/xlsx-extractor.ts"],"names":[],"mappings":";;;AAEA;;;;;GAKG;AACH,MAAa,aAAa;IACxB,KAAK,CAAC,OAAO,CAAC,QAAgB;QAC5B,iEAAiE;QACjE,MAAM,IAAI,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC;QAE7B,MAAM,QAAQ,GAAG,IAAI,CAAC,QAAQ,CAAC,QAAQ,EAAE,EAAE,IAAI,EAAE,MAAM,EAAE,QAAQ,EAAE,IAAI,EAAE,CAAC,CAAC;QAC3E,MAAM,MAAM,GAAqB,EAAE,CAAC;QAEpC,QAAQ,CAAC,UAAU,CAAC,OAAO,CAAC,CAAC,SAAiB,EAAE,GAAW,EAAE,EAAE;YAC7D,MAAM,KAAK,GAAG,QAAQ,CAAC,MAAM,CAAC,SAAS,CAAC,CAAC;YACzC,MAAM,GAAG,GAAW,IAAI,CAAC,KAAK,CAAC,YAAY,CAAC,KAAK,EAAE,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;YACpE,MAAM,IAAI,GAAG,UAAU,SAAS,KAAK,GAAG,EAAE,CAAC,IAAI,EAAE,CAAC;YAClD,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACpB,MAAM,CAAC,IAAI,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,GAAG,GAAG,CAAC,EAAE,CAAC,CAAC;YACvC,CAAC;QACH,CAAC,CAAC,CAAC;QAEH,OAAO,MAAM,CAAC;IAChB,CAAC;CACF;AAnBD,sCAmBC"}