@agentionai/agents 0.11.0 → 0.12.0-beta

package/dist/parsers/UnstructuredLocalParser.js

@@ -0,0 +1,118 @@
+"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.UnstructuredLocalParser = void 0;
+const DocumentParser_1 = require("./DocumentParser");
+/**
+ * Document parser that uses the **local** (open-source Python) version of
+ * Unstructured via the `@epilogo/unstructured-io-node` npm bridge.
+ *
+ * The bridge spawns a Python virtual environment and calls the Python
+ * `unstructured` library directly — no API key required, but Python 3.8+
+ * and system dependencies (poppler, tesseract, etc.) must be available.
+ *
+ * **Peer dependency:** `@epilogo/unstructured-io-node`
+ *
+ * @example
+ * ```typescript
+ * import { UnstructuredLocalParser } from "@agentionai/agents/parsers";
+ *
+ * const parser = new UnstructuredLocalParser();
+ * const doc = await parser.parse("/path/to/report.pdf", {
+ *   strategy: "hi_res",
+ *   languages: ["eng"],
+ * });
+ * console.log(doc.elements?.length, "elements");
+ *
+ * // Use with IngestionPipeline
+ * await pipeline.ingestFile("/path/to/report.pdf", parser);
+ * ```
+ */
+class UnstructuredLocalParser extends DocumentParser_1.DocumentParser {
+    constructor() {
+        super(...arguments);
+        this.name = "unstructured-local";
+    }
+    /**
+     * Parse a file using the local Python Unstructured library.
+     *
+     * On first call, `ensureEnvironmentSetup()` is invoked to download the
+     * Python venv if it does not already exist (one-time, slow operation).
+     *
+     * @param filePath - Path to the document to parse
+     * @param options  - Strategy, languages, and any other unstructured kwargs
+     */
+    async parse(filePath, options) {
+        const pkg = "@epilogo/unstructured-io-node";
+        let UnstructuredIO;
+        try {
+            // eslint-disable-next-line @typescript-eslint/no-explicit-any
+            ({ UnstructuredIO } = await Promise.resolve(`${pkg}`).then(s => __importStar(require(s))));
+        }
+        catch {
+            throw new Error("UnstructuredLocalParser requires '@epilogo/unstructured-io-node'. " +
+                "Install it with: npm install @epilogo/unstructured-io-node");
+        }
+        await UnstructuredIO.ensureEnvironmentSetup();
+        const { strategy, languages, ...rest } = options ?? {};
+        const rawElements = await UnstructuredIO.partition({
+            filename: filePath,
+            strategy: strategy ?? "auto",
+            ...(languages ? { languages } : {}),
+            ...rest,
+        });
+        const elements = this.mapRawElements(rawElements);
+        return {
+            text: this.elementsToText(elements),
+            elements,
+        };
+    }
+    mapRawElements(raw) {
+        return raw.map((el) => {
+            const e = el;
+            return {
+                type: typeof e["type"] === "string" ? e["type"] : "unknown",
+                text: typeof e["text"] === "string" ? e["text"] : "",
+                metadata: e["metadata"] != null &&
+                    typeof e["metadata"] === "object" &&
+                    !Array.isArray(e["metadata"])
+                    ? e["metadata"]
+                    : undefined,
+            };
+        });
+    }
+}
+exports.UnstructuredLocalParser = UnstructuredLocalParser;
+//# sourceMappingURL=UnstructuredLocalParser.js.map

package/dist/parsers/index.d.ts

@@ -0,0 +1,3 @@
+export type { ParsedElement, ParsedDocument, ParseOptions } from "./types";
+export { DocumentParser } from "./DocumentParser";
+//# sourceMappingURL=index.d.ts.map

package/dist/parsers/index.js

@@ -0,0 +1,6 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.DocumentParser = void 0;
+var DocumentParser_1 = require("./DocumentParser");
+Object.defineProperty(exports, "DocumentParser", { enumerable: true, get: function () { return DocumentParser_1.DocumentParser; } });
+//# sourceMappingURL=index.js.map

package/dist/parsers/types.d.ts

@@ -0,0 +1,50 @@
+/**
+ * A single structured element extracted from a document.
+ * Matches the element format returned by Unstructured and similar parsers.
+ */
+export interface ParsedElement {
+    /**
+     * Element type — e.g. "Title", "NarrativeText", "Table", "Image",
+     * "ListItem", "Header", "Footer", "Document", etc.
+     */
+    type: string;
+    /** Text content of this element */
+    text: string;
+    /**
+     * Parser-provided metadata — e.g. page_number, coordinates, languages,
+     * file_directory, filename, filetype, etc.
+     */
+    metadata?: Record<string, unknown>;
+}
+/**
+ * The result of parsing a document file.
+ */
+export interface ParsedDocument {
+    /** Full plain-text content (elements joined by double newlines) */
+    text: string;
+    /**
+     * Structured elements if the parser supports them.
+     * Absent when the parser only returns plain text.
+     */
+    elements?: ParsedElement[];
+    /** File-level metadata from the parser, when available */
+    metadata?: Record<string, unknown>;
+}
+/**
+ * Options shared across all document parsers.
+ */
+export interface ParseOptions {
+    /**
+     * Parsing strategy.
+     * - `"auto"`: Let the parser decide (default)
+     * - `"fast"`: Text extraction only, no OCR
+     * - `"hi_res"`: High-resolution layout analysis with OCR
+     * - `"ocr_only"`: Force OCR on every page
+     */
+    strategy?: "auto" | "fast" | "hi_res" | "ocr_only";
+    /** Languages to use for OCR (ISO 639-1 codes, e.g. `["eng", "fra"]`) */
+    languages?: string[];
+    /** Pass-through options specific to the underlying parser */
+    [key: string]: unknown;
+}
+//# sourceMappingURL=types.d.ts.map

package/dist/parsers/types.js

@@ -0,0 +1,3 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+//# sourceMappingURL=types.js.map

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "@agentionai/agents",
   "author": "Laurent Zuijdwijk",
-  "version": "0.11.0",
+  "version": "0.12.0-beta",
   "description": "Agent Library",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
@@ -30,6 +30,10 @@
       "types": "./dist/gemini.d.ts",
       "default": "./dist/gemini.js"
     },
+    "./ollama": {
+      "types": "./dist/agents/ollama/OllamaAgent.d.ts",
+      "default": "./dist/agents/ollama/OllamaAgent.js"
+    },
     "./embeddings": {
       "types": "./dist/embeddings/index.d.ts",
       "default": "./dist/embeddings/index.js"
@@ -61,6 +65,26 @@
     "./history/plugins": {
       "types": "./dist/history/plugins/index.d.ts",
       "default": "./dist/history/plugins/index.js"
+    },
+    "./parsers": {
+      "types": "./dist/parsers/index.d.ts",
+      "default": "./dist/parsers/index.js"
+    },
+    "./parsers/unstructured-local": {
+      "types": "./dist/parsers/UnstructuredLocalParser.d.ts",
+      "default": "./dist/parsers/UnstructuredLocalParser.js"
+    },
+    "./parsers/unstructured-api": {
+      "types": "./dist/parsers/UnstructuredAPIParser.d.ts",
+      "default": "./dist/parsers/UnstructuredAPIParser.js"
+    },
+    "./parsers/llamaindex": {
+      "types": "./dist/parsers/LlamaIndexParser.d.ts",
+      "default": "./dist/parsers/LlamaIndexParser.js"
+    },
+    "./parsers/ollama-ocr": {
+      "types": "./dist/parsers/OllamaOCRParser.d.ts",
+      "default": "./dist/parsers/OllamaOCRParser.js"
     }
   },
   "files": [
@@ -141,7 +165,12 @@
     "@modelcontextprotocol/sdk": "^1.26.0",
     "apache-arrow": "^18.0.0",
     "openai": "^6.16.0",
-    "voyageai": "^0.0.3"
+    "voyageai": "^0.0.3",
+    "@epilogo/unstructured-io-node": "*",
+    "unstructured-client": "*",
+    "llamaindex": "*",
+    "@llamaindex/readers": "*",
+    "pdf-to-img": "*"
   },
   "peerDependenciesMeta": {
     "@lancedb/lancedb": {
@@ -170,6 +199,21 @@
     },
     "@opensearch-project/opensearch": {
       "optional": true
+    },
+    "@epilogo/unstructured-io-node": {
+      "optional": true
+    },
+    "unstructured-client": {
+      "optional": true
+    },
+    "llamaindex": {
+      "optional": true
+    },
+    "@llamaindex/readers": {
+      "optional": true
+    },
+    "pdf-to-img": {
+      "optional": true
     }
   },
   "dependencies": {