langchain 0.0.191 → 0.0.193

package/dist/document_loaders/web/apify_dataset.js

@@ -7,6 +7,27 @@ import { getEnvironmentVariable } from "../../util/env.js";
  * A class that extends the BaseDocumentLoader and implements the
  * DocumentLoader interface. It represents a document loader that loads
  * documents from an Apify dataset.
+ * @example
+ * ```typescript
+ * const loader = new ApifyDatasetLoader("your-dataset-id", {
+ *   datasetMappingFunction: (item) =>
+ *     new Document({
+ *       pageContent: item.text || "",
+ *       metadata: { source: item.url },
+ *     }),
+ *   clientOptions: {
+ *     token: "your-apify-token",
+ *   },
+ * });
+ *
+ * const docs = await loader.load();
+ *
+ * const chain = new RetrievalQAChain();
+ * const res = await chain.invoke({ query: "What is LangChain?" });
+ *
+ * console.log(res.text);
+ * console.log(res.sourceDocuments.map((d) => d.metadata.source));
+ * ```
  */
 export class ApifyDatasetLoader extends BaseDocumentLoader {
     constructor(datasetId, config) {
@@ -52,8 +73,11 @@ export class ApifyDatasetLoader extends BaseDocumentLoader {
      * @returns An array of Document instances.
      */
     async load() {
-        const datasetItems = (await this.apifyClient.dataset(this.datasetId).listItems({ clean: true })).items;
-        return await Promise.all(datasetItems.map((item) => this.caller.call(async () => this.datasetMappingFunction(item))));
+        const dataset = await this.apifyClient
+            .dataset(this.datasetId)
+            .listItems({ clean: true });
+        const documentList = await Promise.all(dataset.items.map((item) => this.caller.call(async () => this.datasetMappingFunction(item))));
+        return documentList.flat();
     }
     /**
      * Create an ApifyDatasetLoader by calling an Actor on the Apify platform and waiting for its results to be ready.

package/dist/document_loaders/web/assemblyai.cjs

@@ -74,6 +74,15 @@ class CreateTranscriptLoader extends AssemblyAILoader {
 }
 /**
  * Creates and loads the transcript as a document using AssemblyAI.
+ * @example
+ * ```typescript
+ * const loader = new AudioTranscriptLoader(
+ *   { audio_url: "https:
+ *   { apiKey: "ASSEMBLYAI_API_KEY" },
+ * );
+ * const docs = await loader.load();
+ * console.dir(docs, { depth: Infinity });
+ * ```
  */
 class AudioTranscriptLoader extends CreateTranscriptLoader {
     /**
@@ -131,6 +140,21 @@ class AudioTranscriptSentencesLoader extends CreateTranscriptLoader {
 exports.AudioTranscriptSentencesLoader = AudioTranscriptSentencesLoader;
 /**
  * Creates a transcript and loads subtitles for the transcript as `srt` or `vtt` format.
+ * @example
+ * ```typescript
+ * const loader = new AudioSubtitleLoader(
+ *   {
+ *     audio_url: "https:
+ *   },
+ *   "srt",
+ *   {
+ *     apiKey: "<ASSEMBLYAI_API_KEY>",
+ *   },
+ * );
+ *
+ *
+ * const docs = await loader.load();
+ * ```
  */
 class AudioSubtitleLoader extends CreateTranscriptLoader {
     /**

package/dist/document_loaders/web/assemblyai.d.ts

@@ -29,6 +29,15 @@ declare abstract class CreateTranscriptLoader extends AssemblyAILoader {
 }
 /**
  * Creates and loads the transcript as a document using AssemblyAI.
+ * @example
+ * ```typescript
+ * const loader = new AudioTranscriptLoader(
+ *   { audio_url: "https:
+ *   { apiKey: "ASSEMBLYAI_API_KEY" },
+ * );
+ * const docs = await loader.load();
+ * console.dir(docs, { depth: Infinity });
+ * ```
  */
 export declare class AudioTranscriptLoader extends CreateTranscriptLoader {
     /**
@@ -61,6 +70,21 @@ export declare class AudioTranscriptSentencesLoader extends CreateTranscriptLoad
 }
 /**
  * Creates a transcript and loads subtitles for the transcript as `srt` or `vtt` format.
+ * @example
+ * ```typescript
+ * const loader = new AudioSubtitleLoader(
+ *   {
+ *     audio_url: "https:
+ *   },
+ *   "srt",
+ *   {
+ *     apiKey: "<ASSEMBLYAI_API_KEY>",
+ *   },
+ * );
+ *
+ *
+ * const docs = await loader.load();
+ * ```
  */
 export declare class AudioSubtitleLoader extends CreateTranscriptLoader {
     private subtitleFormat;

package/dist/document_loaders/web/assemblyai.js

@@ -71,6 +71,15 @@ class CreateTranscriptLoader extends AssemblyAILoader {
 }
 /**
  * Creates and loads the transcript as a document using AssemblyAI.
+ * @example
+ * ```typescript
+ * const loader = new AudioTranscriptLoader(
+ *   { audio_url: "https:
+ *   { apiKey: "ASSEMBLYAI_API_KEY" },
+ * );
+ * const docs = await loader.load();
+ * console.dir(docs, { depth: Infinity });
+ * ```
  */
 export class AudioTranscriptLoader extends CreateTranscriptLoader {
     /**
@@ -125,6 +134,21 @@ export class AudioTranscriptSentencesLoader extends CreateTranscriptLoader {
 }
 /**
  * Creates a transcript and loads subtitles for the transcript as `srt` or `vtt` format.
+ * @example
+ * ```typescript
+ * const loader = new AudioSubtitleLoader(
+ *   {
+ *     audio_url: "https:
+ *   },
+ *   "srt",
+ *   {
+ *     apiKey: "<ASSEMBLYAI_API_KEY>",
+ *   },
+ * );
+ *
+ *
+ * const docs = await loader.load();
+ * ```
  */
 export class AudioSubtitleLoader extends CreateTranscriptLoader {
     /**

package/dist/document_loaders/web/azure_blob_storage_file.cjs

@@ -34,6 +34,17 @@ const unstructured_js_1 = require("../fs/unstructured.cjs");
  * Class representing a document loader that loads a specific file from
  * Azure Blob Storage. It extends the BaseDocumentLoader class and
  * implements the DocumentLoader interface.
+ * @example
+ * ```typescript
+ * const loader = new AzureBlobStorageFileLoader({
+ *   azureConfig: {
+ *     connectionString: "{connectionString}",
+ *     container: "{containerName}",
+ *     blobName: "{blobName}",
+ *   },
+ * });
+ * const docs = await loader.load();
+ * ```
  */
 class AzureBlobStorageFileLoader extends base_js_1.BaseDocumentLoader {
     constructor({ azureConfig, unstructuredConfig, }) {

package/dist/document_loaders/web/azure_blob_storage_file.d.ts

@@ -22,6 +22,17 @@ interface AzureBlobStorageFileLoaderConfig {
  * Class representing a document loader that loads a specific file from
  * Azure Blob Storage. It extends the BaseDocumentLoader class and
  * implements the DocumentLoader interface.
+ * @example
+ * ```typescript
+ * const loader = new AzureBlobStorageFileLoader({
+ *   azureConfig: {
+ *     connectionString: "{connectionString}",
+ *     container: "{containerName}",
+ *     blobName: "{blobName}",
+ *   },
+ * });
+ * const docs = await loader.load();
+ * ```
  */
 export declare class AzureBlobStorageFileLoader extends BaseDocumentLoader {
     private readonly connectionString;

package/dist/document_loaders/web/azure_blob_storage_file.js

@@ -8,6 +8,17 @@ import { UnstructuredLoader, } from "../fs/unstructured.js";
  * Class representing a document loader that loads a specific file from
  * Azure Blob Storage. It extends the BaseDocumentLoader class and
  * implements the DocumentLoader interface.
+ * @example
+ * ```typescript
+ * const loader = new AzureBlobStorageFileLoader({
+ *   azureConfig: {
+ *     connectionString: "{connectionString}",
+ *     container: "{containerName}",
+ *     blobName: "{blobName}",
+ *   },
+ * });
+ * const docs = await loader.load();
+ * ```
  */
 export class AzureBlobStorageFileLoader extends BaseDocumentLoader {
     constructor({ azureConfig, unstructuredConfig, }) {

package/dist/document_loaders/web/cheerio.cjs

@@ -8,6 +8,12 @@ const async_caller_js_1 = require("../../util/async_caller.cjs");
  * A class that extends the BaseDocumentLoader and implements the
  * DocumentLoader interface. It represents a document loader for loading
  * web-based documents using Cheerio.
+ * @example
+ * ```typescript
+ * const loader = new CheerioWebBaseLoader("https:exampleurl.com");
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 class CheerioWebBaseLoader extends base_js_1.BaseDocumentLoader {
     constructor(webPath, fields) {

package/dist/document_loaders/web/cheerio.d.ts

@@ -27,6 +27,12 @@ export interface WebBaseLoaderParams extends AsyncCallerParams {
  * A class that extends the BaseDocumentLoader and implements the
  * DocumentLoader interface. It represents a document loader for loading
  * web-based documents using Cheerio.
+ * @example
+ * ```typescript
+ * const loader = new CheerioWebBaseLoader("https:exampleurl.com");
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 export declare class CheerioWebBaseLoader extends BaseDocumentLoader implements DocumentLoader {
     webPath: string;

package/dist/document_loaders/web/cheerio.js

@@ -5,6 +5,12 @@ import { AsyncCaller } from "../../util/async_caller.js";
  * A class that extends the BaseDocumentLoader and implements the
  * DocumentLoader interface. It represents a document loader for loading
  * web-based documents using Cheerio.
+ * @example
+ * ```typescript
+ * const loader = new CheerioWebBaseLoader("https:exampleurl.com");
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 export class CheerioWebBaseLoader extends BaseDocumentLoader {
     constructor(webPath, fields) {

package/dist/document_loaders/web/college_confidential.cjs

@@ -6,6 +6,12 @@ const cheerio_js_1 = require("./cheerio.cjs");
 /**
  * A document loader specifically designed for loading documents from the
  * College Confidential website. It extends the CheerioWebBaseLoader.
+ * @example
+ * ```typescript
+ * const loader = new CollegeConfidentialLoader("https:exampleurl.com");
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 class CollegeConfidentialLoader extends cheerio_js_1.CheerioWebBaseLoader {
     constructor(webPath) {

package/dist/document_loaders/web/college_confidential.d.ts

@@ -3,6 +3,12 @@ import { CheerioWebBaseLoader } from "./cheerio.js";
 /**
  * A document loader specifically designed for loading documents from the
  * College Confidential website. It extends the CheerioWebBaseLoader.
+ * @example
+ * ```typescript
+ * const loader = new CollegeConfidentialLoader("https:exampleurl.com");
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 export declare class CollegeConfidentialLoader extends CheerioWebBaseLoader {
     constructor(webPath: string);

package/dist/document_loaders/web/college_confidential.js

@@ -3,6 +3,12 @@ import { CheerioWebBaseLoader } from "./cheerio.js";
 /**
  * A document loader specifically designed for loading documents from the
  * College Confidential website. It extends the CheerioWebBaseLoader.
+ * @example
+ * ```typescript
+ * const loader = new CollegeConfidentialLoader("https:exampleurl.com");
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 export class CollegeConfidentialLoader extends CheerioWebBaseLoader {
     constructor(webPath) {

package/dist/document_loaders/web/confluence.cjs

@@ -6,6 +6,17 @@ const document_js_1 = require("../../document.cjs");
 const base_js_1 = require("../base.cjs");
 /**
  * Class representing a document loader for loading pages from Confluence.
+ * @example
+ * ```typescript
+ * const loader = new ConfluencePagesLoader({
+ *   baseUrl: "https:
+ *   spaceKey: "~EXAMPLE362906de5d343d49dcdbae5dEXAMPLE",
+ *   username: "your-username",
+ *   accessToken: "your-access-token",
+ * });
+ * const documents = await loader.load();
+ * console.log(documents);
+ * ```
  */
 class ConfluencePagesLoader extends base_js_1.BaseDocumentLoader {
     constructor({ baseUrl, spaceKey, username, accessToken, limit = 25, }) {

package/dist/document_loaders/web/confluence.d.ts

@@ -32,6 +32,17 @@ export interface ConfluenceAPIResponse {
 }
 /**
  * Class representing a document loader for loading pages from Confluence.
+ * @example
+ * ```typescript
+ * const loader = new ConfluencePagesLoader({
+ *   baseUrl: "https:
+ *   spaceKey: "~EXAMPLE362906de5d343d49dcdbae5dEXAMPLE",
+ *   username: "your-username",
+ *   accessToken: "your-access-token",
+ * });
+ * const documents = await loader.load();
+ * console.log(documents);
+ * ```
  */
 export declare class ConfluencePagesLoader extends BaseDocumentLoader {
     readonly baseUrl: string;

package/dist/document_loaders/web/confluence.js

@@ -3,6 +3,17 @@ import { Document } from "../../document.js";
 import { BaseDocumentLoader } from "../base.js";
 /**
  * Class representing a document loader for loading pages from Confluence.
+ * @example
+ * ```typescript
+ * const loader = new ConfluencePagesLoader({
+ *   baseUrl: "https:
+ *   spaceKey: "~EXAMPLE362906de5d343d49dcdbae5dEXAMPLE",
+ *   username: "your-username",
+ *   accessToken: "your-access-token",
+ * });
+ * const documents = await loader.load();
+ * console.log(documents);
+ * ```
  */
 export class ConfluencePagesLoader extends BaseDocumentLoader {
     constructor({ baseUrl, spaceKey, username, accessToken, limit = 25, }) {

package/dist/document_loaders/web/figma.cjs

@@ -9,6 +9,15 @@ const env_js_1 = require("../../util/env.cjs");
  * extends the BaseDocumentLoader and implements the FigmaLoaderParams
  * interface. The constructor takes a config object as a parameter, which
  * contains the access token, an array of node IDs, and the file key.
+ * @example
+ * ```typescript
+ * const loader = new FigmaFileLoader({
+ *   accessToken: "FIGMA_ACCESS_TOKEN",
+ *   nodeIds: ["id1", "id2", "id3"],
+ *   fileKey: "key",
+ * });
+ * const docs = await loader.load();
+ * ```
  */
 class FigmaFileLoader extends base_js_1.BaseDocumentLoader {
     constructor({ accessToken = (0, env_js_1.getEnvironmentVariable)("FIGMA_ACCESS_TOKEN"), nodeIds, fileKey, }) {

package/dist/document_loaders/web/figma.d.ts

@@ -39,6 +39,15 @@ export interface FigmaLoaderParams {
  * extends the BaseDocumentLoader and implements the FigmaLoaderParams
  * interface. The constructor takes a config object as a parameter, which
  * contains the access token, an array of node IDs, and the file key.
+ * @example
+ * ```typescript
+ * const loader = new FigmaFileLoader({
+ *   accessToken: "FIGMA_ACCESS_TOKEN",
+ *   nodeIds: ["id1", "id2", "id3"],
+ *   fileKey: "key",
+ * });
+ * const docs = await loader.load();
+ * ```
  */
 export declare class FigmaFileLoader extends BaseDocumentLoader implements FigmaLoaderParams {
     accessToken?: string;

package/dist/document_loaders/web/figma.js

@@ -6,6 +6,15 @@ import { getEnvironmentVariable } from "../../util/env.js";
  * extends the BaseDocumentLoader and implements the FigmaLoaderParams
  * interface. The constructor takes a config object as a parameter, which
  * contains the access token, an array of node IDs, and the file key.
+ * @example
+ * ```typescript
+ * const loader = new FigmaFileLoader({
+ *   accessToken: "FIGMA_ACCESS_TOKEN",
+ *   nodeIds: ["id1", "id2", "id3"],
+ *   fileKey: "key",
+ * });
+ * const docs = await loader.load();
+ * ```
  */
 export class FigmaFileLoader extends BaseDocumentLoader {
     constructor({ accessToken = getEnvironmentVariable("FIGMA_ACCESS_TOKEN"), nodeIds, fileKey, }) {

package/dist/document_loaders/web/notionapi.cjs

@@ -24,6 +24,22 @@ exports.isDatabase = isDatabase;
 /**
  * A class that extends the BaseDocumentLoader class. It represents a
  * document loader for loading documents from Notion using the Notion API.
+ * @example
+ * ```typescript
+ * const pageLoader = new NotionAPILoader({
+ *   clientOptions: { auth: "<NOTION_INTEGRATION_TOKEN>" },
+ *   id: "<PAGE_ID>",
+ *   type: "page",
+ * });
+ * const pageDocs = await pageLoader.loadAndSplit();
+ * const dbLoader = new NotionAPILoader({
+ *   clientOptions: { auth: "<NOTION_INTEGRATION_TOKEN>" },
+ *   id: "<DATABASE_ID>",
+ *   type: "database",
+ *   propertiesAsHeader: true,
+ * });
+ * const dbDocs = await dbLoader.load();
+ * ```
  */
 class NotionAPILoader extends base_js_1.BaseDocumentLoader {
     constructor(options) {

package/dist/document_loaders/web/notionapi.d.ts

@@ -34,6 +34,22 @@ export type NotionAPILoaderOptions = {
 /**
  * A class that extends the BaseDocumentLoader class. It represents a
  * document loader for loading documents from Notion using the Notion API.
+ * @example
+ * ```typescript
+ * const pageLoader = new NotionAPILoader({
+ *   clientOptions: { auth: "<NOTION_INTEGRATION_TOKEN>" },
+ *   id: "<PAGE_ID>",
+ *   type: "page",
+ * });
+ * const pageDocs = await pageLoader.loadAndSplit();
+ * const dbLoader = new NotionAPILoader({
+ *   clientOptions: { auth: "<NOTION_INTEGRATION_TOKEN>" },
+ *   id: "<DATABASE_ID>",
+ *   type: "database",
+ *   propertiesAsHeader: true,
+ * });
+ * const dbDocs = await dbLoader.load();
+ * ```
  */
 export declare class NotionAPILoader extends BaseDocumentLoader {
     private caller;

package/dist/document_loaders/web/notionapi.js

@@ -13,6 +13,22 @@ export const isDatabase = (res) => isDatabaseResponse(res) && isFullDatabase(res
 /**
  * A class that extends the BaseDocumentLoader class. It represents a
  * document loader for loading documents from Notion using the Notion API.
+ * @example
+ * ```typescript
+ * const pageLoader = new NotionAPILoader({
+ *   clientOptions: { auth: "<NOTION_INTEGRATION_TOKEN>" },
+ *   id: "<PAGE_ID>",
+ *   type: "page",
+ * });
+ * const pageDocs = await pageLoader.loadAndSplit();
+ * const dbLoader = new NotionAPILoader({
+ *   clientOptions: { auth: "<NOTION_INTEGRATION_TOKEN>" },
+ *   id: "<DATABASE_ID>",
+ *   type: "database",
+ *   propertiesAsHeader: true,
+ * });
+ * const dbDocs = await dbLoader.load();
+ * ```
  */
 export class NotionAPILoader extends BaseDocumentLoader {
     constructor(options) {

package/dist/document_loaders/web/notiondb.cjs

@@ -4,7 +4,17 @@ exports.NotionDBLoader = void 0;
 const base_js_1 = require("../base.cjs");
 const env_js_1 = require("../../util/env.cjs");
 const NOTION_BASE_URL = "https://api.notion.com/v1";
-/** @deprecated use the `NotionAPILoader` class instead. */
+/** @deprecated use the `NotionAPILoader` class instead.
+ * @example
+ * ```typescript
+ * const loader = new NotionDBLoader({
+ *   pageSizeLimit: 10,
+ *   databaseId: "{databaseId}",
+ *   notionIntegrationToken: "{notionIntegrationToken}",
+ * });
+ * const docs = await loader.load();
+ * ```
+ */
 class NotionDBLoader extends base_js_1.BaseDocumentLoader {
     constructor({ databaseId, notionApiVersion = "2022-06-28", notionIntegrationToken = (0, env_js_1.getEnvironmentVariable)("NOTION_INTEGRATION_TOKEN"), pageSizeLimit = 50, }) {
         super();

package/dist/document_loaders/web/notiondb.d.ts

@@ -11,7 +11,17 @@ export interface NotionDBLoaderParams {
     notionApiVersion?: string;
     pageSizeLimit?: number;
 }
-/** @deprecated use the `NotionAPILoader` class instead. */
+/** @deprecated use the `NotionAPILoader` class instead.
+ * @example
+ * ```typescript
+ * const loader = new NotionDBLoader({
+ *   pageSizeLimit: 10,
+ *   databaseId: "{databaseId}",
+ *   notionIntegrationToken: "{notionIntegrationToken}",
+ * });
+ * const docs = await loader.load();
+ * ```
+ */
 export declare class NotionDBLoader extends BaseDocumentLoader implements NotionDBLoaderParams {
     integrationToken: string;
     databaseId: string;

package/dist/document_loaders/web/notiondb.js

@@ -1,7 +1,17 @@
 import { BaseDocumentLoader } from "../base.js";
 import { getEnvironmentVariable } from "../../util/env.js";
 const NOTION_BASE_URL = "https://api.notion.com/v1";
-/** @deprecated use the `NotionAPILoader` class instead. */
+/** @deprecated use the `NotionAPILoader` class instead.
+ * @example
+ * ```typescript
+ * const loader = new NotionDBLoader({
+ *   pageSizeLimit: 10,
+ *   databaseId: "{databaseId}",
+ *   notionIntegrationToken: "{notionIntegrationToken}",
+ * });
+ * const docs = await loader.load();
+ * ```
+ */
 export class NotionDBLoader extends BaseDocumentLoader {
     constructor({ databaseId, notionApiVersion = "2022-06-28", notionIntegrationToken = getEnvironmentVariable("NOTION_INTEGRATION_TOKEN"), pageSizeLimit = 50, }) {
         super();

package/dist/document_loaders/web/pdf.cjs

@@ -6,9 +6,15 @@ const base_js_1 = require("../base.cjs");
 const document_js_2 = require("../../util/document.cjs");
 /**
  * A document loader for loading data from PDFs.
+ * @example
+ * ```typescript
+ * const loader = new WebPDFLoader(new Blob());
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 class WebPDFLoader extends base_js_1.BaseDocumentLoader {
-    constructor(blob, { splitPages = true, pdfjs = PDFLoaderImports } = {}) {
+    constructor(blob, { splitPages = true, pdfjs = PDFLoaderImports, parsedItemSeparator = " ", } = {}) {
         super();
         Object.defineProperty(this, "blob", {
             enumerable: true,
@@ -28,9 +34,16 @@ class WebPDFLoader extends base_js_1.BaseDocumentLoader {
             writable: true,
             value: void 0
         });
+        Object.defineProperty(this, "parsedItemSeparator", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
         this.blob = blob;
         this.splitPages = splitPages ?? this.splitPages;
         this.pdfjs = pdfjs;
+        this.parsedItemSeparator = parsedItemSeparator;
     }
     /**
      * Loads the contents of the PDF as documents.
@@ -68,7 +81,7 @@ class WebPDFLoader extends base_js_1.BaseDocumentLoader {
                     lastY = item.transform[5];
                 }
             }
-            const text = textItems.join(" ");
+            const text = textItems.join(this.parsedItemSeparator);
             documents.push(new document_js_1.Document({
                 pageContent: text,
                 metadata: {

package/dist/document_loaders/web/pdf.d.ts

@@ -3,14 +3,22 @@ import { Document } from "../../document.js";
 import { BaseDocumentLoader } from "../base.js";
 /**
  * A document loader for loading data from PDFs.
+ * @example
+ * ```typescript
+ * const loader = new WebPDFLoader(new Blob());
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 export declare class WebPDFLoader extends BaseDocumentLoader {
     protected blob: Blob;
     protected splitPages: boolean;
     private pdfjs;
-    constructor(blob: Blob, { splitPages, pdfjs }?: {
+    protected parsedItemSeparator: string;
+    constructor(blob: Blob, { splitPages, pdfjs, parsedItemSeparator, }?: {
         splitPages?: boolean | undefined;
         pdfjs?: typeof PDFLoaderImports | undefined;
+        parsedItemSeparator?: string | undefined;
     });
     /**
      * Loads the contents of the PDF as documents.

package/dist/document_loaders/web/pdf.js

@@ -3,9 +3,15 @@ import { BaseDocumentLoader } from "../base.js";
 import { formatDocumentsAsString } from "../../util/document.js";
 /**
  * A document loader for loading data from PDFs.
+ * @example
+ * ```typescript
+ * const loader = new WebPDFLoader(new Blob());
+ * const docs = await loader.load();
+ * console.log({ docs });
+ * ```
  */
 export class WebPDFLoader extends BaseDocumentLoader {
-    constructor(blob, { splitPages = true, pdfjs = PDFLoaderImports } = {}) {
+    constructor(blob, { splitPages = true, pdfjs = PDFLoaderImports, parsedItemSeparator = " ", } = {}) {
         super();
         Object.defineProperty(this, "blob", {
             enumerable: true,
@@ -25,9 +31,16 @@ export class WebPDFLoader extends BaseDocumentLoader {
             writable: true,
             value: void 0
         });
+        Object.defineProperty(this, "parsedItemSeparator", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
         this.blob = blob;
         this.splitPages = splitPages ?? this.splitPages;
         this.pdfjs = pdfjs;
+        this.parsedItemSeparator = parsedItemSeparator;
     }
     /**
      * Loads the contents of the PDF as documents.
@@ -65,7 +78,7 @@ export class WebPDFLoader extends BaseDocumentLoader {
                     lastY = item.transform[5];
                 }
             }
-            const text = textItems.join(" ");
+            const text = textItems.join(this.parsedItemSeparator);
             documents.push(new Document({
                 pageContent: text,
                 metadata: {

package/dist/document_loaders/web/puppeteer.cjs

@@ -7,6 +7,18 @@ const base_js_1 = require("../base.cjs");
  * Class that extends the BaseDocumentLoader class and implements the
  * DocumentLoader interface. It represents a document loader for scraping
  * web pages using Puppeteer.
+ * @example
+ * ```typescript
+ * const loader = new PuppeteerWebBaseLoader("https:exampleurl.com", {
+ *   launchOptions: {
+ *     headless: true,
+ *   },
+ *   gotoOptions: {
+ *     waitUntil: "domcontentloaded",
+ *   },
+ * });
+ * const screenshot = await loader.screenshot();
+ * ```
  */
 class PuppeteerWebBaseLoader extends base_js_1.BaseDocumentLoader {
     constructor(webPath, options) {