@undefineds.co/xpod 0.4.15 → 0.4.16

package/node_modules/@undefineds.co/extensions/dist-cjs/adapters/llamaindex.js

@@ -0,0 +1,54 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.toLlamaIndexDocuments = toLlamaIndexDocuments;
+exports.createLlamaIndexReader = createLlamaIndexReader;
+exports.createReaderFromLlamaIndexReader = createReaderFromLlamaIndexReader;
+const reader_js_1 = require("../reader.js");
+function toLlamaIndexDocuments(result) {
+    return result.documents.map((document) => ({
+        text: document.text,
+        metadata: (0, reader_js_1.mergeDocumentMetadata)(result, document),
+    }));
+}
+function createLlamaIndexReader(reader, ctx) {
+    return {
+        loadData: async (input) => {
+            const requests = Array.isArray(input) ? input : [input];
+            const results = await Promise.all(requests.map((request) => reader.read(request, ctx)));
+            return results.flatMap((result) => toLlamaIndexDocuments(result));
+        },
+    };
+}
+function createReaderFromLlamaIndexReader(options) {
+    return {
+        id: options.id,
+        displayName: options.displayName,
+        outputs: options.outputs,
+        transports: options.transports,
+        supportedSourceSchemes: options.supportedSourceSchemes,
+        supportedMimeTypes: options.supportedMimeTypes,
+        read: async (request, ctx) => {
+            const llamaReader = await options.createReader(request, ctx);
+            const loadInput = options.createLoadDataInput?.(request, ctx) ?? request;
+            const llamaDocuments = await llamaReader.loadData(loadInput);
+            const documents = llamaDocuments.map((document) => llamaIndexDocumentToReaderDocument(document));
+            const result = {
+                reader: options.id,
+                model: request.model,
+                documents,
+                text: documents.map((document) => document.text).filter(Boolean).join('\n\n'),
+            };
+            return {
+                ...result,
+                text: (0, reader_js_1.readResultText)(result),
+            };
+        },
+    };
+}
+function llamaIndexDocumentToReaderDocument(document) {
+    return {
+        id: document.id_ ?? document.id,
+        text: document.getText?.() ?? document.text ?? document.textResource?.text ?? '',
+        metadata: document.metadata ?? {},
+    };
+}

package/node_modules/@undefineds.co/extensions/dist-cjs/index.js

@@ -0,0 +1,22 @@
+"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 __exportStar = (this && this.__exportStar) || function(m, exports) {
+    for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
+};
+Object.defineProperty(exports, "__esModule", { value: true });
+__exportStar(require("./types.js"), exports);
+__exportStar(require("./registry.js"), exports);
+__exportStar(require("./reader.js"), exports);
+__exportStar(require("./adapters/langchain.js"), exports);
+__exportStar(require("./adapters/llamaindex.js"), exports);
+__exportStar(require("./paddleocr.js"), exports);

package/node_modules/@undefineds.co/extensions/dist-cjs/package.json

@@ -0,0 +1,3 @@
+{
+  "type": "commonjs"
+}

package/node_modules/@undefineds.co/extensions/dist-cjs/paddleocr.js

@@ -0,0 +1,304 @@
+"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.PADDLEOCR_DEFAULT_ENDPOINT = void 0;
+exports.createPaddleOcrReader = createPaddleOcrReader;
+exports.createPaddleOcrExtension = createPaddleOcrExtension;
+const reader_js_1 = require("./reader.js");
+exports.PADDLEOCR_DEFAULT_ENDPOINT = 'https://paddleocr.aistudio-app.com/api/v2/ocr/jobs';
+function createPaddleOcrReader(options) {
+    const id = options.id ?? 'paddleocr';
+    const model = options.model ?? 'PP-OCRv6';
+    return {
+        id,
+        displayName: options.displayName ?? 'PaddleOCR Reader',
+        outputs: ['markdown', 'text', 'structured'],
+        transports: ['remote-http'],
+        supportedSourceSchemes: ['https', 'http', 'file'],
+        supportedMimeTypes: ['application/pdf', 'image/png', 'image/jpeg', 'image/webp', 'image/tiff'],
+        read: async (request, ctx) => {
+            const fetcher = options.fetch ?? globalThis.fetch;
+            if (!fetcher) {
+                throw new reader_js_1.ReaderError('No fetch implementation is available for PaddleOCR reader.', {
+                    code: 'INVALID_INPUT',
+                    retryable: false,
+                });
+            }
+            const payload = await createPaddlePayload(request, model, options.defaultOptions);
+            const response = await fetcher(resolvePaddleEndpoint(options), {
+                method: 'POST',
+                headers: createHeaders(options.apiKey),
+                body: JSON.stringify(payload),
+                signal: ctx.signal,
+            });
+            const body = await readJsonResponse(response);
+            if (!response.ok) {
+                throw new reader_js_1.ReaderError(`PaddleOCR request failed with HTTP ${response.status}.`, {
+                    code: response.status >= 500 || response.status === 429 ? 'PROVIDER_UNAVAILABLE' : 'PROVIDER_REJECTED',
+                    status: response.status,
+                    retryable: response.status >= 500 || response.status === 429,
+                    cause: body,
+                });
+            }
+            return paddleResponseToReadResult(id, model, request, body);
+        },
+    };
+}
+function createPaddleOcrExtension(options) {
+    return (api) => {
+        const reader = createPaddleOcrReader(options);
+        api.reader.register(reader);
+        api.config.registerSchema({
+            id: reader.id,
+            fields: [
+                { name: 'apiKey', type: 'secret', required: true, description: 'Bearer token for the PaddleOCR provider.' },
+                { name: 'model', type: 'string', required: false, description: 'PaddleOCR model or pipeline name.' },
+            ],
+        });
+        api.runtime.registerCapability({
+            id: `${reader.id}-remote-http`,
+            kind: 'remote-http',
+            readerIds: [reader.id],
+            sourceSchemes: ['https', 'http', 'file'],
+        });
+    };
+}
+function resolvePaddleEndpoint(options) {
+    return options.endpoint?.trim() || exports.PADDLEOCR_DEFAULT_ENDPOINT;
+}
+async function createPaddlePayload(request, model, defaultOptions) {
+    const input = request.input ?? sourceToInput(request.source);
+    const file = await inputToPaddleFile(input);
+    return {
+        ...defaultOptions,
+        ...request.options,
+        file,
+        fileType: inferPaddleFileType(input.mimeType ?? input.filename ?? request.source),
+        fileName: input.filename ?? filenameFromUri(request.source),
+        pageRange: request.pages,
+        model: request.model ?? model,
+    };
+}
+function sourceToInput(source) {
+    if (/^https?:\/\//u.test(source)) {
+        return {
+            kind: 'remote-url',
+            url: source,
+            filename: filenameFromUri(source),
+            mimeType: inferMimeType(source),
+        };
+    }
+    if (source.startsWith('file://')) {
+        return {
+            kind: 'local-path',
+            path: decodeURIComponent(source.slice('file://'.length)),
+            filename: filenameFromUri(source),
+            mimeType: inferMimeType(source),
+        };
+    }
+    throw new reader_js_1.ReaderError(`Unsupported PaddleOCR source URI: ${source}`, {
+        code: 'UNSUPPORTED_SOURCE',
+        retryable: false,
+    });
+}
+async function inputToPaddleFile(input) {
+    switch (input.kind) {
+        case 'remote-url':
+            return input.url;
+        case 'bytes':
+            return bytesToBase64(input.bytes);
+        case 'local-path': {
+            const fs = await Promise.resolve().then(() => __importStar(require('node:fs/promises')));
+            return bytesToBase64(await fs.readFile(input.path));
+        }
+        case 'stream':
+            return bytesToBase64(await streamToBytes(input.stream));
+    }
+}
+function createHeaders(apiKey) {
+    const headers = {
+        'Content-Type': 'application/json',
+    };
+    if (apiKey) {
+        headers.Authorization = `Bearer ${apiKey}`;
+    }
+    return headers;
+}
+async function readJsonResponse(response) {
+    const text = await response.text();
+    if (!text) {
+        return {};
+    }
+    try {
+        return JSON.parse(text);
+    }
+    catch (error) {
+        throw new reader_js_1.ReaderError('PaddleOCR returned a non-JSON response.', {
+            code: 'PARSE_FAILED',
+            status: response.status,
+            retryable: response.status >= 500,
+            cause: error,
+        });
+    }
+}
+function paddleResponseToReadResult(readerId, model, request, body) {
+    const entries = extractPaddleEntries(body);
+    const documents = entries.map((entry, index) => paddleEntryToDocument(entry, index, request, model));
+    const markdown = documents.map((document) => document.text).filter(Boolean).join('\n\n');
+    return {
+        reader: readerId,
+        model: request.model ?? model,
+        documents,
+        markdown,
+        text: markdown,
+        raw: body,
+    };
+}
+function extractPaddleEntries(body) {
+    if (!isRecord(body)) {
+        return [];
+    }
+    const candidates = [
+        body.result,
+        body.data,
+        body,
+    ];
+    for (const candidate of candidates) {
+        if (isRecord(candidate) && Array.isArray(candidate.layoutParsingResults)) {
+            return candidate.layoutParsingResults;
+        }
+        if (isRecord(candidate) && Array.isArray(candidate.results)) {
+            return candidate.results;
+        }
+    }
+    return [body];
+}
+function paddleEntryToDocument(entry, index, request, model) {
+    const record = isRecord(entry) ? entry : {};
+    const markdown = isRecord(record.markdown) ? record.markdown : undefined;
+    const text = firstString([
+        markdown?.text,
+        record.markdown,
+        record.text,
+        record.content,
+    ]);
+    const page = firstNumber([
+        record.page,
+        record.pageId,
+        record.pageIndex,
+        record.page_number,
+    ]);
+    return {
+        id: `${request.source}#paddleocr-${index + 1}`,
+        text,
+        metadata: {
+            sourceUri: request.source,
+            page,
+            model: request.model ?? model,
+        },
+        sourceUri: request.source,
+        raw: entry,
+    };
+}
+function firstString(values) {
+    for (const value of values) {
+        if (typeof value === 'string') {
+            return value;
+        }
+    }
+    return '';
+}
+function firstNumber(values) {
+    for (const value of values) {
+        if (typeof value === 'number') {
+            return value;
+        }
+    }
+    return undefined;
+}
+function isRecord(value) {
+    return typeof value === 'object' && value !== null;
+}
+function inferPaddleFileType(value) {
+    const normalized = value?.toLowerCase() ?? '';
+    if (normalized.includes('pdf') || normalized.endsWith('.pdf')) {
+        return 0;
+    }
+    return 1;
+}
+function inferMimeType(uri) {
+    const clean = uri.split(/[?#]/u)[0]?.toLowerCase() ?? '';
+    if (clean.endsWith('.pdf'))
+        return 'application/pdf';
+    if (clean.endsWith('.png'))
+        return 'image/png';
+    if (clean.endsWith('.jpg') || clean.endsWith('.jpeg'))
+        return 'image/jpeg';
+    if (clean.endsWith('.webp'))
+        return 'image/webp';
+    if (clean.endsWith('.tiff') || clean.endsWith('.tif'))
+        return 'image/tiff';
+    return undefined;
+}
+function filenameFromUri(uri) {
+    const clean = uri.split(/[?#]/u)[0] ?? '';
+    const filename = clean.split('/').filter(Boolean).at(-1);
+    return filename ? decodeURIComponent(filename) : undefined;
+}
+function bytesToBase64(bytes) {
+    return Buffer.from(bytes).toString('base64');
+}
+async function streamToBytes(stream) {
+    const reader = stream.getReader();
+    const chunks = [];
+    let total = 0;
+    while (true) {
+        const { done, value } = await reader.read();
+        if (done) {
+            break;
+        }
+        if (value) {
+            chunks.push(value);
+            total += value.byteLength;
+        }
+    }
+    const result = new Uint8Array(total);
+    let offset = 0;
+    for (const chunk of chunks) {
+        result.set(chunk, offset);
+        offset += chunk.byteLength;
+    }
+    return result;
+}

package/node_modules/@undefineds.co/extensions/dist-cjs/reader.js

@@ -0,0 +1,44 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.ReaderError = void 0;
+exports.readResultText = readResultText;
+exports.mergeDocumentMetadata = mergeDocumentMetadata;
+class ReaderError extends Error {
+    name = 'ReaderError';
+    code;
+    status;
+    retryable;
+    constructor(message, options) {
+        super(message, { cause: options.cause });
+        this.code = options.code;
+        this.status = options.status;
+        this.retryable = options.retryable ?? false;
+    }
+}
+exports.ReaderError = ReaderError;
+function readResultText(result) {
+    if (typeof result.text === 'string') {
+        return result.text;
+    }
+    if (typeof result.markdown === 'string') {
+        return result.markdown;
+    }
+    return result.documents.map((document) => document.text).filter(Boolean).join('\n\n');
+}
+function mergeDocumentMetadata(result, document) {
+    const metadata = {
+        id: document.id,
+        reader: result.reader,
+        model: result.model,
+        sourceUri: document.sourceUri,
+        pageRange: document.pageRange,
+        mimeType: document.mimeType,
+        ...document.metadata,
+    };
+    for (const [key, value] of Object.entries(metadata)) {
+        if (value === undefined) {
+            delete metadata[key];
+        }
+    }
+    return metadata;
+}

package/node_modules/@undefineds.co/extensions/dist-cjs/registry.js

@@ -0,0 +1,99 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.createExtensionRegistry = createExtensionRegistry;
+function createExtensionRegistry() {
+    return new InMemoryExtensionRegistry();
+}
+class InMemoryExtensionRegistry {
+    readers = new Map();
+    sourceAccessors = new Map();
+    runtimeCapabilities = new Map();
+    tools = new Map();
+    configSchemas = new Map();
+    install(extension, descriptor) {
+        const api = this.createApi(descriptor);
+        const result = extension(api);
+        if (isPromiseLike(result)) {
+            return result.then(() => undefined);
+        }
+    }
+    getReader(id) {
+        return this.readers.get(id);
+    }
+    listReaders() {
+        return [...this.readers.values()];
+    }
+    getSourceAccessor(id) {
+        return this.sourceAccessors.get(id);
+    }
+    listSourceAccessors() {
+        return [...this.sourceAccessors.values()];
+    }
+    getRuntimeCapability(id) {
+        return this.runtimeCapabilities.get(id);
+    }
+    listRuntimeCapabilities() {
+        return [...this.runtimeCapabilities.values()];
+    }
+    getTool(id) {
+        return this.tools.get(id);
+    }
+    listTools() {
+        return [...this.tools.values()];
+    }
+    getConfigSchema(id) {
+        return this.configSchemas.get(id);
+    }
+    listConfigSchemas() {
+        return [...this.configSchemas.values()];
+    }
+    createApi(descriptor) {
+        const registration = {
+            extensionId: descriptor.id,
+            extensionVersion: descriptor.version,
+        };
+        return {
+            reader: {
+                register: (definition) => {
+                    assertUnique(this.readers, 'reader', definition.id);
+                    this.readers.set(definition.id, { ...definition, ...registration });
+                },
+            },
+            source: {
+                register: (definition) => {
+                    assertUnique(this.sourceAccessors, 'source accessor', definition.id);
+                    this.sourceAccessors.set(definition.id, { ...definition, ...registration });
+                },
+            },
+            runtime: {
+                registerCapability: (definition) => {
+                    assertUnique(this.runtimeCapabilities, 'runtime capability', definition.id);
+                    this.runtimeCapabilities.set(definition.id, { ...definition, ...registration });
+                },
+            },
+            tools: {
+                register: (definition) => {
+                    assertUnique(this.tools, 'tool', definition.id);
+                    this.tools.set(definition.id, { ...definition, ...registration });
+                },
+            },
+            config: {
+                registerSchema: (definition) => {
+                    assertUnique(this.configSchemas, 'config schema', definition.id);
+                    this.configSchemas.set(definition.id, { ...definition, ...registration });
+                },
+            },
+        };
+    }
+}
+function assertUnique(map, label, id) {
+    if (map.has(id)) {
+        throw new Error(`Duplicate ${label} id ${id}`);
+    }
+}
+function isPromiseLike(value) {
+    return typeof value === 'object'
+        && value !== null
+        && 'then' in value
+        && typeof value.then === 'function';
+}

package/node_modules/@undefineds.co/extensions/dist-cjs/types.js

@@ -0,0 +1,2 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });