@promptbook/cli 0.114.0-2 → 0.114.0-4

package/src/remote-server/createRemoteClient.ts

@@ -1,10 +1,19 @@
 import type { Socket } from 'socket.io-client';
-import { io } from 'socket.io-client';
 import { spaceTrim } from 'spacetrim';
 import { CONNECTION_RETRIES_LIMIT, CONNECTION_TIMEOUT_MS } from '../config';
+import { createLazyModuleLoader } from '../utils/misc/createLazyModuleLoader';
 import { isValidUrl } from '../utils/validators/url/isValidUrl';
 import type { RemoteClientOptions } from './types/RemoteClientOptions';
 
+/**
+ * Loads the Socket.io client (`socket.io-client`) on demand
+ *
+ * Note: [🐌] Loaded lazily to keep the startup of the `ptbk` CLI utility fast
+ *
+ * @private internal utility of `createRemoteClient`
+ */
+const loadSocketIoClientModule = createLazyModuleLoader(() => import('socket.io-client'));
+
 /**
  * Creates a connection to the remote proxy server.
  *
@@ -44,6 +53,8 @@ export async function createRemoteClient<TCustomOptions = undefined>(
         );
     }
 
+    const { io } = await loadSocketIoClientModule();
+
     return new Promise((resolve, reject) => {
         const socket = io(remoteServerUrl, {
             retries: CONNECTION_RETRIES_LIMIT,

package/src/{commitments/USE_BROWSER → scrapers/_common/utils}/fetchUrlContent.ts

@@ -1,9 +1,9 @@
 import { spaceTrim } from 'spacetrim';
-import type { LlmExecutionTools } from '../../execution/LlmExecutionTools';
-import { $provideFilesystemForNode } from '../../scrapers/_common/register/$provideFilesystemForNode';
-import type { ScraperSourceHandler } from '../../scrapers/_common/Scraper';
-import { promptbookFetch } from '../../scrapers/_common/utils/promptbookFetch';
-import { WebsiteScraper } from '../../scrapers/website/WebsiteScraper';
+import type { LlmExecutionTools } from '../../../execution/LlmExecutionTools';
+import { $provideFilesystemForNode } from '../register/$provideFilesystemForNode';
+import type { ScraperSourceHandler } from '../Scraper';
+import { WebsiteScraper } from '../../website/WebsiteScraper';
+import { promptbookFetch } from './promptbookFetch';
 
 /**
  * Fetches and scrapes content from a URL (SERVER-SIDE ONLY)
@@ -15,13 +15,12 @@ import { WebsiteScraper } from '../../scrapers/website/WebsiteScraper';
  * 4. Returns the scraped markdown content
  *
  * WARNING: This function should NOT be used directly in browser environments.
- * For browser environments, use fetchUrlContentViaBrowser which proxies through
- * the Agents Server API endpoint at /api/scrape
+ * For browser environments, proxy the request through the Agents Server API endpoint at /api/scrape
  *
  * @param url The URL to fetch and scrape
  * @returns Markdown content from the URL
  *
- * @private internal utility for USE BROWSER commitment
+ * @private internal scraping utility
  */
 export async function fetchUrlContent(url: string): Promise<string> {
     try {
@@ -137,4 +136,4 @@ export async function fetchUrlContent(url: string): Promise<string> {
     }
 }
 
-// Note: [🟢] Code for Node browser-fetch helper [fetchUrlContent](src/commitments/USE_BROWSER/fetchUrlContent.ts) should never be published into packages that could be imported into browser environment
+// Note: [🟢] Code for Node fetch-and-scrape helper [fetchUrlContent](src/scrapers/_common/utils/fetchUrlContent.ts) should never be published into packages that could be imported into browser environment

package/src/scrapers/website/WebsiteScraper.ts

@@ -4,8 +4,6 @@ import type { Converter } from '../_common/Converter';
 import type { Scraper, ScraperSourceHandler } from '../_common/Scraper';
 // TODO: [🏳‍🌈] Finally take pick of .json vs .ts
 // import PipelineCollection from '../../../books/books';
-import { Readability } from '@mozilla/readability';
-import { JSDOM } from 'jsdom';
 import type { Converter as ShowdownConverter } from 'showdown';
 import { DEFAULT_INTERMEDIATE_FILES_STRATEGY, DEFAULT_IS_VERBOSE, DEFAULT_SCRAPE_CACHE_DIRNAME } from '../../config';
 import { EnvironmentMismatchError } from '../../errors/EnvironmentMismatchError';
@@ -13,6 +11,7 @@ import { KnowledgeScrapeError } from '../../errors/KnowledgeScrapeError';
 import { UnexpectedError } from '../../errors/UnexpectedError';
 import type { ExecutionTools } from '../../execution/ExecutionTools';
 import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { createLazyModuleLoader } from '../../utils/misc/createLazyModuleLoader';
 import { MarkdownScraper } from '../markdown/MarkdownScraper';
 import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
 import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
@@ -20,6 +19,25 @@ import { getScraperIntermediateSource } from '../_common/utils/getScraperInterme
 import { websiteScraperMetadata } from './register-metadata';
 import { createShowdownConverter } from './utils/createShowdownConverter';
 
+/**
+ * Loads `jsdom` on demand
+ *
+ * Note: [🐌] `jsdom` is by far the heaviest dependency of Promptbook, loading it eagerly would slow down every single
+ *       run of the `ptbk` CLI utility even when no website is scraped
+ *
+ * @private internal utility of `WebsiteScraper`
+ */
+const loadJsdomModule = createLazyModuleLoader(() => import('jsdom'));
+
+/**
+ * Loads `@mozilla/readability` on demand
+ *
+ * Note: [🐌] Loaded lazily to keep the startup of the `ptbk` CLI utility fast
+ *
+ * @private internal utility of `WebsiteScraper`
+ */
+const loadReadabilityModule = createLazyModuleLoader(() => import('@mozilla/readability'));
+
 /**
  * Scraper for websites
  *
@@ -77,6 +95,8 @@ export class WebsiteScraper implements Converter, Scraper {
             throw new EnvironmentMismatchError('Can not scrape websites without filesystem tools');
         }
 
+        const [{ JSDOM }, { Readability }] = await Promise.all([loadJsdomModule(), loadReadabilityModule()]);
+
         const jsdom = new JSDOM(await source.asText(), {
             url: source.url,
         });

package/src/utils/misc/createLazyModuleLoader.ts

@@ -0,0 +1,27 @@
+/**
+ * Creates a loader which imports one module on the first call and reuses the very same module afterwards
+ *
+ * Note: [🐌] Heavy third-party dependencies are imported lazily to keep the startup of the Promptbook CLI fast.
+ *       A statically imported dependency is loaded every single time the bundle is loaded, even when the running
+ *       command never touches it. A lazily imported dependency is loaded only when the feature is really used.
+ *
+ * @example
+ * const loadJsdomModule = createLazyModuleLoader(() => import('jsdom'));
+ * const { JSDOM } = await loadJsdomModule();
+ *
+ * @private internal utility of Promptbook
+ */
+export function createLazyModuleLoader<TModule>(importModule: () => Promise<TModule>): () => Promise<TModule> {
+    let importedModulePromise: Promise<TModule> | null = null;
+
+    return function loadModule(): Promise<TModule> {
+        if (importedModulePromise === null) {
+            importedModulePromise = importModule();
+        }
+
+        return importedModulePromise;
+    };
+}
+
+// Note: [🐌] Do not convert the lazy `import(...)` calls back to static `import` statements, it would bring back the
+//       slow startup of the `ptbk` CLI utility

package/src/version.ts

@@ -16,11 +16,11 @@ export const BOOK_LANGUAGE_VERSION: string_semantic_version = '2.0.0';
  * @generated
  * @see https://github.com/webgptorg/promptbook
  */
-export const PROMPTBOOK_ENGINE_VERSION: string_promptbook_version = '0.114.0-2';
+export const PROMPTBOOK_ENGINE_VERSION: string_promptbook_version = '0.114.0-4';
 
 /**
  * Represents the version string of the Promptbook engine.
- * It follows semantic versioning (e.g., `0.113.1-0`).
+ * It follows semantic versioning (e.g., `0.114.0-3`).
  *
  * @generated
  */

package/src/versions.txt

@@ -1180,3 +1180,5 @@
 0.113.0
 0.113.1-0
 0.114.0-2
+0.114.0-3
+0.114.0-4