@crawlee/browser 4.0.0-beta.82 → 4.0.0-beta.84

package/internals/browser-crawler.d.ts

@@ -224,17 +224,13 @@ export interface BrowserCrawlerOptions<Page extends CommonPage = CommonPage, Res
  * @category Crawlers
  */
 export declare abstract class BrowserCrawler<Page extends CommonPage = CommonPage, Response extends BaseResponse = BaseResponse, InternalBrowserPoolOptions extends BrowserPoolOptions = BrowserPoolOptions, LaunchOptions extends Dictionary | undefined = Dictionary, Context extends BrowserCrawlingContext<Page, Response, Dictionary> = BrowserCrawlingContext<Page, Response, Dictionary>, ContextExtension = Dictionary<never>, ExtendedContext extends Context = Context & ContextExtension, GoToOptions extends Dictionary = Dictionary> extends BasicCrawler<Context, ContextExtension, ExtendedContext> {
+    /** Backs the {@link BrowserCrawler.browserPool|`browserPool`} getter. */
+    private browserPoolDep;
     /**
      * A reference to the underlying browser pool that manages the crawler's browsers. Typed as
      * {@link IBrowserPool} so custom implementations can be plugged in via the `browserPool` constructor option.
      */
-    browserPool: IBrowserPool<Page>;
-    /**
-     * Set when the crawler constructed its own pool (a {@link BrowserPool}, or a {@link RemoteBrowserPool}
-     * built from the `remoteBrowser` option). Holds the same instance as `browserPool` but is the only reference
-     * the crawler tears down — a user-supplied `browserPool` is never owned and never destroyed by the crawler.
-     */
-    private ownedBrowserPool?;
+    get browserPool(): IBrowserPool<Page>;
     launchContext: BrowserLaunchContext<LaunchOptions, unknown>;
     protected readonly ignoreShadowRoots: boolean;
     protected readonly ignoreIframes: boolean;

package/internals/browser-crawler.js

@@ -1,4 +1,4 @@
-import { BasicCrawler, browserPoolCookieToToughCookie, ContextPipeline, cookieStringToToughCookie, enqueueLinks, handleRequestTimeout, NavigationSkippedError, RequestState, resolveBaseUrlForEnqueueLinksFiltering, SessionError, toughCookieToBrowserPoolCookie, tryAbsoluteURL, validators, } from '@crawlee/basic';
+import { BasicCrawler, browserPoolCookieToToughCookie, ContextPipeline, cookieStringToToughCookie, enqueueLinks, handleRequestTimeout, NavigationSkippedError, OwnedOrInjected, RequestState, resolveBaseUrlForEnqueueLinksFiltering, SessionError, toughCookieToBrowserPoolCookie, tryAbsoluteURL, validators, } from '@crawlee/basic';
 import { BrowserPool, RemoteBrowserPool } from '@crawlee/browser-pool';
 import { CLOUDFLARE_RETRY_CSS_SELECTORS, RETRY_CSS_SELECTORS, sleep } from '@crawlee/utils';
 import ow from 'ow';
@@ -48,17 +48,15 @@ const readContextField = (ctx, key) => ctx[key];
  * @category Crawlers
  */
 export class BrowserCrawler extends BasicCrawler {
+    /** Backs the {@link BrowserCrawler.browserPool|`browserPool`} getter. */
+    browserPoolDep;
     /**
      * A reference to the underlying browser pool that manages the crawler's browsers. Typed as
      * {@link IBrowserPool} so custom implementations can be plugged in via the `browserPool` constructor option.
      */
-    browserPool;
-    /**
-     * Set when the crawler constructed its own pool (a {@link BrowserPool}, or a {@link RemoteBrowserPool}
-     * built from the `remoteBrowser` option). Holds the same instance as `browserPool` but is the only reference
-     * the crawler tears down — a user-supplied `browserPool` is never owned and never destroyed by the crawler.
-     */
-    ownedBrowserPool;
+    get browserPool() {
+        return this.browserPoolDep.value;
+    }
     launchContext;
     ignoreShadowRoots;
     ignoreIframes;
@@ -120,35 +118,34 @@ export class BrowserCrawler extends BasicCrawler {
             this.launchContext.launchOptions.headless = headless;
         }
         this.saveResponseCookies = saveResponseCookies;
-        // `browserPool` wins over `remoteBrowser` — a passed-in pool is used as-is, the sugar is ignored.
-        if (browserPool) {
-            this.browserPool = browserPool;
-            return;
-        }
-        const resolvedBrowserPoolOptions = browserPoolOptions ?? {};
-        if (launchContext?.userAgent) {
-            if (resolvedBrowserPoolOptions.useFingerprints)
-                this.log.info('Custom user agent provided, disabling automatic browser fingerprint injection!');
-            resolvedBrowserPoolOptions.useFingerprints = false;
-        }
-        if (remoteBrowser) {
-            // The crawler already built the right plugin for its browser — hand it to a RemoteBrowserPool so the
-            // remote connection is always for the matching browser (no plugin to construct, no way to mismatch).
-            const { browserPlugins, ...remoteBrowserPoolOptions } = resolvedBrowserPoolOptions;
-            const remotePool = new RemoteBrowserPool({
-                browserPlugins: browserPlugins,
-                ...remoteBrowser,
-                browserPoolOptions: remoteBrowserPoolOptions,
+        // `browserPool` wins over `remoteBrowser` — a passed-in pool is used as-is (borrowed), the sugar is ignored.
+        // The default is only built when no pool was injected, so all the option/launchContext fiddling below stays
+        // inside the factory.
+        this.browserPoolDep = OwnedOrInjected.resolve(browserPool, () => {
+            const resolvedBrowserPoolOptions = browserPoolOptions ?? {};
+            if (launchContext?.userAgent) {
+                if (resolvedBrowserPoolOptions.useFingerprints)
+                    this.log.info('Custom user agent provided, disabling automatic browser fingerprint injection!');
+                resolvedBrowserPoolOptions.useFingerprints = false;
+            }
+            if (remoteBrowser) {
+                // The crawler already built the right plugin for its browser — hand it to a RemoteBrowserPool so the
+                // remote connection is always for the matching browser (no plugin to construct, no way to mismatch).
+                const { browserPlugins, ...remoteBrowserPoolOptions } = resolvedBrowserPoolOptions;
+                return new RemoteBrowserPool({
+                    browserPlugins: browserPlugins,
+                    ...remoteBrowser,
+                    browserPoolOptions: remoteBrowserPoolOptions,
+                });
+            }
+            // Double cast: `BrowserPool` implements `IBrowserPool<PageReturn>`, where `PageReturn` is derived from the
+            // plugin/controller generics and doesn't overlap with the crawler's free `Page` type param, so TS won't
+            // narrow it directly. The concrete pool does satisfy the `Page`/`destroy` contract at runtime — this is the
+            // long-standing `Page` variance gap, not a `destroy`-related hole.
+            return new BrowserPool({
+                ...resolvedBrowserPoolOptions,
             });
-            this.ownedBrowserPool = remotePool;
-            this.browserPool = remotePool;
-            return;
-        }
-        const ownedBrowserPool = new BrowserPool({
-            ...resolvedBrowserPoolOptions,
         });
-        this.ownedBrowserPool = ownedBrowserPool;
-        this.browserPool = ownedBrowserPool;
     }
     buildContextPipeline() {
         return ContextPipeline.create().compose({
@@ -363,7 +360,7 @@ export class BrowserCrawler extends BasicCrawler {
      * @ignore
      */
     async teardown() {
-        await this.ownedBrowserPool?.destroy();
+        await this.browserPoolDep.ifOwned((pool) => pool.destroy());
         await super.teardown();
     }
 }

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@crawlee/browser",
-    "version": "4.0.0-beta.82",
+    "version": "4.0.0-beta.84",
     "description": "The scalable web crawling and scraping library for JavaScript/Node.js. Enables development of data extraction and web automation jobs (not only) with headless Chrome and Puppeteer.",
     "engines": {
         "node": ">=22.0.0"
@@ -48,10 +48,10 @@
     },
     "dependencies": {
         "@apify/timeout": "^0.3.2",
-        "@crawlee/basic": "4.0.0-beta.82",
-        "@crawlee/browser-pool": "4.0.0-beta.82",
-        "@crawlee/types": "4.0.0-beta.82",
-        "@crawlee/utils": "4.0.0-beta.82",
+        "@crawlee/basic": "4.0.0-beta.84",
+        "@crawlee/browser-pool": "4.0.0-beta.84",
+        "@crawlee/types": "4.0.0-beta.84",
+        "@crawlee/utils": "4.0.0-beta.84",
         "ow": "^2.0.0",
         "tslib": "^2.8.1",
         "type-fest": "^4.41.0"
@@ -75,5 +75,5 @@
             }
         }
     },
-    "gitHead": "eb1096f7c7743d124375ef011fbbadb19476822e"
+    "gitHead": "5f8f4ae742975dc914c865c9766b93d274d7468f"
 }