@crawlee/playwright 4.0.0-beta.99 → 4.0.0-rc.1

package/internals/playwright-crawler.d.ts

@@ -1,48 +1,25 @@
-import type { BrowserCrawlerOptions, BrowserCrawlingContext, BrowserHook, GetUserDataFromRequest, RequestHandler, RouterHandler, RouterRoutes, RouteSchemas, RoutesFromSchemas } from '@crawlee/browser';
+import type { BrowserCrawlerOptions, BrowserCrawlingContext, BrowserHook, GetUserDataFromRequest, RouterHandler, RouterRoutes, RouteSchemas, RoutesFromSchemas } from '@crawlee/browser';
 import { BrowserCrawler } from '@crawlee/browser';
-import type { PlaywrightPlugin } from '@crawlee/browser-pool';
 import type { Dictionary } from '@crawlee/types';
 // @ts-ignore optional peer dependency or compatibility with es2022
-import type { Download, LaunchOptions, Page, Response } from 'playwright';
-import type { EnqueueLinksByClickingElementsOptions } from './enqueue-links/click-elements.js';
+import type { Page, Response } from 'playwright';
+import { z } from 'zod';
 import type { PlaywrightLaunchContext } from './playwright-launcher.js';
-import type { BlockRequestsOptions, DirectNavigationOptions, HandleCloudflareChallengeOptions, InfiniteScrollOptions, InjectFileOptions, PlaywrightContextUtils, SaveSnapshotOptions } from './utils/playwright-utils.js';
+import type { DirectNavigationOptions, HandleCloudflareChallengeOptions, PlaywrightContextUtils } from './utils/playwright-utils.js';
 export type PlaywrightGotoOptions = NonNullable<Parameters<Page['goto']>[1]>;
-export interface PlaywrightCrawlingContext<UserData extends Dictionary = Dictionary> extends BrowserCrawlingContext<Page, Response, UserData, PlaywrightGotoOptions>, PlaywrightContextUtils {
+export interface PlaywrightCrawlingContext<UserData extends Dictionary = any> extends BrowserCrawlingContext<Page, Response, UserData, PlaywrightGotoOptions>, PlaywrightContextUtils {
 }
-// @ts-ignore optional peer dependency or compatibility with es2022
-export interface PlaywrightHook extends BrowserHook<PlaywrightCrawlingContext> {
-}
-export interface PlaywrightCrawlerOptions<ContextExtension = Dictionary<never>, ExtendedContext extends PlaywrightCrawlingContext = PlaywrightCrawlingContext & ContextExtension, Routes extends Record<keyof Routes, Dictionary> = Record<string, GetUserDataFromRequest<PlaywrightCrawlingContext['request']>>> extends BrowserCrawlerOptions<Page, Response, PlaywrightCrawlingContext, ContextExtension, ExtendedContext, {
-    browserPlugins: [PlaywrightPlugin];
-}, Routes> {
+export type PlaywrightHook<UserData extends Dictionary = any> = BrowserHook<PlaywrightCrawlingContext<UserData>>;
+export interface PlaywrightCrawlerOptions<ContextExtension = Dictionary<never>, ExtendedContext extends PlaywrightCrawlingContext = PlaywrightCrawlingContext & ContextExtension, Routes extends Record<keyof Routes, Dictionary> = Record<string, GetUserDataFromRequest<PlaywrightCrawlingContext['request']>>, StatisticStateExtension extends object = {}> extends BrowserCrawlerOptions<Page, Response, PlaywrightCrawlingContext, ContextExtension, ExtendedContext, Routes, StatisticStateExtension> {
     /**
      * The same options as used by {@link launchPlaywright}.
      */
     launchContext?: PlaywrightLaunchContext;
     /**
-     * Function that is called to process each request.
-     *
-     * The function receives the {@link PlaywrightCrawlingContext} as an argument, where:
-     * - `request` is an instance of the {@link Request} object with details about the URL to open, HTTP method etc.
-     * - `page` is an instance of the `Playwright`
-     * [`Page`](https://playwright.dev/docs/api/class-page)
-     * - `response` is an instance of the `Playwright`
-     * [`Response`](https://playwright.dev/docs/api/class-response),
-     * which is the main resource response as returned by `page.goto(request.url)`.
-     *
-     * The function must return a promise, which is then awaited by the crawler.
-     *
-     * If the function throws an exception, the crawler will try to re-crawl the
-     * request later, up to `option.maxRequestRetries` times.
-     * If all the retries fail, the crawler calls the function
-     * provided to the `failedRequestHandler` parameter.
-     * To make this work, you should **always**
-     * let your function throw exceptions rather than catch them.
-     * The exceptions are logged to the request using the
-     * {@link Request.pushErrorMessage} function.
+     * Whether to run browser in headless mode. Defaults to `true`.
+     * Can be also set via {@link Configuration}.
      */
-    requestHandler?: RouterHandler<ExtendedContext, Routes> | RequestHandler<ExtendedContext>;
+    headless?: boolean;
     /**
      * Async functions that are sequentially evaluated before the navigation. Good for setting additional cookies
      * or browser properties before navigation. The function receives the `crawlingContext`; the options object
@@ -59,7 +36,7 @@ export interface PlaywrightCrawlerOptions<ContextExtension = Dictionary<never>,
      * ]
      * ```
      */
-    preNavigationHooks?: BrowserHook<PlaywrightCrawlingContext, ContextExtension>[];
+    preNavigationHooks?: BrowserHook<PlaywrightCrawlingContext<GetUserDataFromRequest<ExtendedContext['request']>>, ContextExtension>[];
     /**
      * Async functions that are sequentially evaluated after the navigation. Good for checking if the navigation was successful.
      * The function accepts `crawlingContext` as the only parameter. A hook may optionally return a partial object
@@ -76,7 +53,7 @@ export interface PlaywrightCrawlerOptions<ContextExtension = Dictionary<never>,
      * ]
      * ```
      */
-    postNavigationHooks?: BrowserHook<PlaywrightCrawlingContext, ContextExtension>[];
+    postNavigationHooks?: BrowserHook<PlaywrightCrawlingContext<GetUserDataFromRequest<ExtendedContext['request']>>, ContextExtension>[];
 }
 /**
  * Provides a simple framework for parallel crawling of web pages
@@ -143,129 +120,135 @@ export interface PlaywrightCrawlerOptions<ContextExtension = Dictionary<never>,
  * ```
  * @category Crawlers
  */
-export declare class PlaywrightCrawler<ContextExtension = Dictionary<never>, ExtendedContext extends PlaywrightCrawlingContext = PlaywrightCrawlingContext & ContextExtension, Routes extends Record<keyof Routes, Dictionary> = Record<string, GetUserDataFromRequest<PlaywrightCrawlingContext['request']>>> extends BrowserCrawler<Page, Response, {
-    browserPlugins: [PlaywrightPlugin];
-}, LaunchOptions, PlaywrightCrawlingContext, ContextExtension, ExtendedContext, Routes> {
+export declare class PlaywrightCrawler<ContextExtension = Dictionary<never>, ExtendedContext extends PlaywrightCrawlingContext = PlaywrightCrawlingContext & ContextExtension, Routes extends Record<keyof Routes, Dictionary> = Record<string, GetUserDataFromRequest<PlaywrightCrawlingContext['request']>>, StatisticStateExtension extends object = {}> extends BrowserCrawler<Page, Response, PlaywrightCrawlingContext, ContextExtension, ExtendedContext, Routes, StatisticStateExtension> {
+    #private;
+    /**
+     * @internal
+     */
     protected static optionsShape: {
-// @ts-ignore optional peer dependency or compatibility with es2022
-        browserPoolOptions: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        launcher: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        ignoreIframes: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        ignoreShadowRoots: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        navigationTimeoutSecs: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        preNavigationHooks: import("ow").ArrayPredicate<unknown> & import("ow").BasePredicate<unknown[] | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        postNavigationHooks: import("ow").ArrayPredicate<unknown> & import("ow").BasePredicate<unknown[] | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        launchContext: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        headless: import("ow").AnyPredicate<string | boolean>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        browserPool: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        remoteBrowser: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        saveResponseCookies: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        proxyConfiguration: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        contextPipelineBuilder: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        extendContext: import("ow").Predicate<Function> & import("ow").BasePredicate<Function | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        requestList: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        requestQueue: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        requestHandler: import("ow").Predicate<Function> & import("ow").BasePredicate<Function | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        requestHandlerTimeoutSecs: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        errorHandler: import("ow").Predicate<Function> & import("ow").BasePredicate<Function | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        failedRequestHandler: import("ow").Predicate<Function> & import("ow").BasePredicate<Function | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        maxRequestRetries: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        sameDomainDelaySecs: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        maxRequestsPerCrawl: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        maxCrawlDepth: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        taskLoopOptions: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        concurrencySystem: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        sessionPool: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        statusMessageLoggingInterval: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        statusMessageCallback: import("ow").Predicate<Function> & import("ow").BasePredicate<Function | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        additionalHttpErrorStatusCodes: import("ow").ArrayPredicate<number>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        ignoreHttpErrorStatusCodes: import("ow").ArrayPredicate<number>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        blockedStatusCodes: import("ow").ArrayPredicate<number>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        retryOnBlocked: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        respectRobotsTxtFile: import("ow").AnyPredicate<boolean | object>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        onSkippedRequest: import("ow").Predicate<Function> & import("ow").BasePredicate<Function | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        httpClient: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        configuration: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        storageBackend: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        eventManager: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        logger: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        minConcurrency: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        maxConcurrency: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        maxRequestsPerMinute: import("ow").NumberPredicate & import("ow").BasePredicate<number | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        keepAlive: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        statisticsOptions: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        id: import("ow").StringPredicate & import("ow").BasePredicate<string | undefined>;
+        contextPipelineBuilder: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        extendContext: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        requestList: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        requestQueue: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        requestManager: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        requestHandler: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        requestHandlerTimeoutSecs: z.ZodOptional<z.ZodCustom<number, number>>;
+        errorHandler: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        failedRequestHandler: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        maxRequestRetries: z.ZodDefault<z.ZodCustom<number, number>>;
+        sameDomainDelaySecs: z.ZodDefault<z.ZodCustom<number, number>>;
+        maxRequestsPerCrawl: z.ZodOptional<z.ZodCustom<number, number>>;
+        maxCrawlDepth: z.ZodOptional<z.ZodCustom<number, number>>;
+        taskLoopOptions: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        concurrencySystem: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        sessionPool: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        statusMessageLoggingInterval: z.ZodDefault<z.ZodCustom<number, number>>;
+        statusMessageCallback: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        additionalHttpErrorStatusCodes: z.ZodDefault<z.ZodArray<z.ZodCustom<number, number>>>;
+        ignoreHttpErrorStatusCodes: z.ZodDefault<z.ZodArray<z.ZodCustom<number, number>>>;
+        blockedStatusCodes: z.ZodOptional<z.ZodArray<z.ZodCustom<number, number>>>;
+        retryOnBlocked: z.ZodDefault<z.ZodBoolean>;
+        respectRobotsTxtFile: z.ZodDefault<z.ZodUnion<readonly [z.ZodBoolean, z.ZodCustom<Dictionary, Dictionary>]>>;
+        transactionalStorage: z.ZodOptional<z.ZodUnion<readonly [z.ZodBoolean, z.ZodObject<{
+            requestQueue: z.ZodOptional<z.ZodEnum<{
+                deferred: "deferred";
+                writeThrough: "writeThrough";
+            }>>;
+        }, z.core.$strict>]>>;
+        onSkippedRequest: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+// @ts-ignore optional peer dependency or compatibility with es2022
+        httpClient: z.ZodOptional<z.ZodInstanceOf<import("@crawlee/http-client").BaseHttpClient>>;
+// @ts-ignore optional peer dependency or compatibility with es2022
+        configuration: z.ZodOptional<z.ZodInstanceOf<import("@crawlee/browser").Configuration>>;
+        storageBackend: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+// @ts-ignore optional peer dependency or compatibility with es2022
+        eventManager: z.ZodOptional<z.ZodInstanceOf<import("@crawlee/browser").EventManager>>;
+        logger: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        minConcurrency: z.ZodOptional<z.ZodCustom<number, number>>;
+        maxConcurrency: z.ZodOptional<z.ZodCustom<number, number>>;
+        initialConcurrency: z.ZodOptional<z.ZodCustom<number, number>>;
+        maxRequestsPerMinute: z.ZodOptional<z.ZodCustom<number, number>>;
+        keepAlive: z.ZodOptional<z.ZodBoolean>;
+        statistics: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        id: z.ZodOptional<z.ZodString>;
+        navigationTimeoutSecs: z.ZodDefault<z.ZodCustom<number, number>>;
+        preNavigationHooks: z.ZodDefault<z.ZodCustom<unknown[], unknown[]>>;
+        postNavigationHooks: z.ZodDefault<z.ZodCustom<unknown[], unknown[]>>;
+        browserPool: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        browserPoolBuilder: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        remoteBrowser: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        saveResponseCookies: z.ZodDefault<z.ZodBoolean>;
+        proxyConfiguration: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        ignoreIframes: z.ZodDefault<z.ZodBoolean>;
+        ignoreShadowRoots: z.ZodDefault<z.ZodBoolean>;
+        launchContext: z.ZodDefault<z.ZodCustom<Dictionary, Dictionary>>;
+        headless: z.ZodOptional<z.ZodBoolean>;
     };
+    /** @internal */
+    protected static optionsSchema: z.ZodObject<{
+        contextPipelineBuilder: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        extendContext: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        requestList: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        requestQueue: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        requestManager: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        requestHandler: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        requestHandlerTimeoutSecs: z.ZodOptional<z.ZodCustom<number, number>>;
+        errorHandler: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        failedRequestHandler: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        maxRequestRetries: z.ZodDefault<z.ZodCustom<number, number>>;
+        sameDomainDelaySecs: z.ZodDefault<z.ZodCustom<number, number>>;
+        maxRequestsPerCrawl: z.ZodOptional<z.ZodCustom<number, number>>;
+        maxCrawlDepth: z.ZodOptional<z.ZodCustom<number, number>>;
+        taskLoopOptions: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        concurrencySystem: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        sessionPool: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        statusMessageLoggingInterval: z.ZodDefault<z.ZodCustom<number, number>>;
+        statusMessageCallback: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        additionalHttpErrorStatusCodes: z.ZodDefault<z.ZodArray<z.ZodCustom<number, number>>>;
+        ignoreHttpErrorStatusCodes: z.ZodDefault<z.ZodArray<z.ZodCustom<number, number>>>;
+        blockedStatusCodes: z.ZodOptional<z.ZodArray<z.ZodCustom<number, number>>>;
+        retryOnBlocked: z.ZodDefault<z.ZodBoolean>;
+        respectRobotsTxtFile: z.ZodDefault<z.ZodUnion<readonly [z.ZodBoolean, z.ZodCustom<Dictionary, Dictionary>]>>;
+        transactionalStorage: z.ZodOptional<z.ZodUnion<readonly [z.ZodBoolean, z.ZodObject<{
+            requestQueue: z.ZodOptional<z.ZodEnum<{
+                deferred: "deferred";
+                writeThrough: "writeThrough";
+            }>>;
+        }, z.core.$strict>]>>;
+        onSkippedRequest: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+// @ts-ignore optional peer dependency or compatibility with es2022
+        httpClient: z.ZodOptional<z.ZodInstanceOf<import("@crawlee/http-client").BaseHttpClient>>;
+// @ts-ignore optional peer dependency or compatibility with es2022
+        configuration: z.ZodOptional<z.ZodInstanceOf<import("@crawlee/browser").Configuration>>;
+        storageBackend: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+// @ts-ignore optional peer dependency or compatibility with es2022
+        eventManager: z.ZodOptional<z.ZodInstanceOf<import("@crawlee/browser").EventManager>>;
+        logger: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        minConcurrency: z.ZodOptional<z.ZodCustom<number, number>>;
+        maxConcurrency: z.ZodOptional<z.ZodCustom<number, number>>;
+        initialConcurrency: z.ZodOptional<z.ZodCustom<number, number>>;
+        maxRequestsPerMinute: z.ZodOptional<z.ZodCustom<number, number>>;
+        keepAlive: z.ZodOptional<z.ZodBoolean>;
+        statistics: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        id: z.ZodOptional<z.ZodString>;
+        navigationTimeoutSecs: z.ZodDefault<z.ZodCustom<number, number>>;
+        preNavigationHooks: z.ZodDefault<z.ZodCustom<unknown[], unknown[]>>;
+        postNavigationHooks: z.ZodDefault<z.ZodCustom<unknown[], unknown[]>>;
+        browserPool: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        browserPoolBuilder: z.ZodOptional<z.ZodCustom<(...args: any[]) => unknown, (...args: any[]) => unknown>>;
+        remoteBrowser: z.ZodOptional<z.ZodCustom<Dictionary, Dictionary>>;
+        saveResponseCookies: z.ZodDefault<z.ZodBoolean>;
+        proxyConfiguration: z.ZodOptional<z.ZodType<Dictionary<any>, unknown, z.core.$ZodTypeInternals<Dictionary<any>, unknown>>>;
+        ignoreIframes: z.ZodDefault<z.ZodBoolean>;
+        ignoreShadowRoots: z.ZodDefault<z.ZodBoolean>;
+        launchContext: z.ZodDefault<z.ZodCustom<Dictionary, Dictionary>>;
+        headless: z.ZodOptional<z.ZodBoolean>;
+    }, z.core.$strict>;
     /**
      * All `PlaywrightCrawler` parameters are passed via an options object.
      */
-    constructor(options?: PlaywrightCrawlerOptions<ContextExtension, ExtendedContext, Routes>);
-// @ts-ignore optional peer dependency or compatibility with es2022
-    protected buildContextPipeline(): import("@crawlee/browser").ContextPipeline<import("@crawlee/browser").CrawlingContext<Dictionary>, BrowserCrawlingContext<Page, Response, Dictionary, Dictionary> & {
-        injectFile: (filePath: string, options?: InjectFileOptions) => Promise<unknown>;
-        injectJQuery: () => Promise<void>;
-        blockRequests: (options?: BlockRequestsOptions) => Promise<void>;
-        waitForSelector: (selector: string, timeoutMs?: number) => Promise<void>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        parseWithCheerio: (selector?: string, timeoutMs?: number) => Promise<import("@crawlee/browser").CheerioAPI>;
-        infiniteScroll: (options?: InfiniteScrollOptions) => Promise<void>;
-        listDownloads: () => Promise<Download[]>;
-        saveSnapshot: (options?: SaveSnapshotOptions) => Promise<void>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        enqueueLinksByClickingElements: (options: Omit<EnqueueLinksByClickingElementsOptions, "page" | "requestManager">) => Promise<import("@crawlee/types").BatchAddRequestsResult>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        compileScript: (scriptString: string, ctx?: Dictionary) => import("./utils/playwright-utils.js").CompiledScriptFunction;
-        closeCookieModals: () => Promise<void>;
-        handleCloudflareChallenge: (options?: HandleCloudflareChallengeOptions) => Promise<Response | undefined>;
-    }>;
-    protected _navigationHandler(crawlingContext: PlaywrightCrawlingContext, gotoOptions: DirectNavigationOptions): Promise<Response | null>;
-    private enhanceContext;
+    constructor(options?: PlaywrightCrawlerOptions<ContextExtension, ExtendedContext, Routes, StatisticStateExtension>);
+    protected navigationHandler(crawlingContext: PlaywrightCrawlingContext, gotoOptions: DirectNavigationOptions): Promise<Response | null>;
 }
 /**
  * Returns a `postNavigationHooks`-ready hook that runs {@link PlaywrightContextUtils.handleCloudflareChallenge}

package/internals/playwright-crawler.js

@@ -1,7 +1,8 @@
 import { BrowserCrawler, RequestState, Router, serviceLocator } from '@crawlee/browser';
-import ow from 'ow';
-import { PlaywrightLauncher } from './playwright-launcher.js';
-import { gotoExtended, playwrightUtils } from './utils/playwright-utils.js';
+import { assertBrowserPoolNotConfigured, parseArgument, schemas } from '@crawlee/utils/internal';
+import { z } from 'zod';
+import { playwrightBrowserPool, remotePlaywrightBrowserPool } from './playwright-browser-pool.js';
+import { blockRequests, compileScript, enqueueLinksByClickingElements, gotoExtended, handleCloudflareChallenge, infiniteScroll, injectFile, injectJQuery, parseWithCheerio, saveSnapshot, } from './utils/playwright-utils.js';
 /**
  * Provides a simple framework for parallel crawling of web pages
  * using headless Chromium, Firefox and Webkit browsers with [Playwright](https://github.com/microsoft/playwright).
@@ -68,51 +69,49 @@ import { gotoExtended, playwrightUtils } from './utils/playwright-utils.js';
  * @category Crawlers
  */
 export class PlaywrightCrawler extends BrowserCrawler {
+    /**
+     * @internal
+     */
     static optionsShape = {
         ...BrowserCrawler.optionsShape,
-        browserPoolOptions: ow.optional.object,
-        launcher: ow.optional.object,
-        ignoreIframes: ow.optional.boolean,
-        ignoreShadowRoots: ow.optional.boolean,
+        launchContext: schemas.anyObject.default(() => ({})),
+        headless: z.boolean().optional(),
     };
+    /** @internal */
+    static optionsSchema = z.strictObject(PlaywrightCrawler.optionsShape);
     /**
      * All `PlaywrightCrawler` parameters are passed via an options object.
      */
     constructor(options = {}) {
-        ow(options, 'PlaywrightCrawlerOptions', ow.object.exactShape(PlaywrightCrawler.optionsShape));
-        const { launchContext = {}, headless, contextPipelineBuilder, ...browserCrawlerOptions } = options;
-        const browserPoolOptions = {
-            ...options.browserPoolOptions,
-        };
+        const parsedOptions = parseArgument(options, PlaywrightCrawler.optionsSchema, 'PlaywrightCrawlerOptions');
+        const { launchContext, headless, configuration, ...browserCrawlerOptions } = parsedOptions;
         if (launchContext.proxyUrl) {
             throw new Error('PlaywrightCrawlerOptions.launchContext.proxyUrl is not allowed in PlaywrightCrawler.' +
                 'Use PlaywrightCrawlerOptions.proxyConfiguration');
         }
-        // `browserPlugins` is working when it's not overridden by `launchContext`,
-        // which for crawlers it is always overridden. Hence the error to use the other option.
-        if (browserPoolOptions.browserPlugins) {
-            throw new Error('browserPoolOptions.browserPlugins is disallowed. Use launchContext.launcher instead.');
-        }
-        if (headless != null) {
-            launchContext.launchOptions ??= {};
-            launchContext.launchOptions.headless = headless;
+        if (options.browserPool) {
+            // The raw options, not the parsed ones: `launchContext` has a default, so by now it is always set.
+            assertBrowserPoolNotConfigured(new.target.name, {
+                launchContext: options.launchContext,
+                headless: options.headless,
+            });
         }
-        const playwrightLauncher = new PlaywrightLauncher(launchContext, options.configuration);
-        browserPoolOptions.browserPlugins = [playwrightLauncher.createBrowserPlugin()];
         super({
             ...browserCrawlerOptions,
-            launchContext,
-            browserPoolOptions,
-            contextPipelineBuilder: contextPipelineBuilder ?? (() => this.buildContextPipeline()),
+            configuration,
+            browserPoolBuilder: (remoteBrowser) => remoteBrowser
+                ? remotePlaywrightBrowserPool({ ...remoteBrowser, launchContext, headless, configuration })
+                : playwrightBrowserPool({ launchContext, headless, configuration }),
+            contextPipelineBuilder: () => this.#buildContextPipeline(),
         });
     }
-    buildContextPipeline() {
-        return super.buildContextPipeline().compose({ action: this.enhanceContext.bind(this) });
+    #buildContextPipeline() {
+        return this.buildContextPipeline().compose(this.#enhanceContext.bind(this));
     }
-    async _navigationHandler(crawlingContext, gotoOptions) {
+    async navigationHandler(crawlingContext, gotoOptions) {
         return gotoExtended(crawlingContext.page, crawlingContext.request, gotoOptions);
     }
-    async enhanceContext(context) {
+    async #enhanceContext(context) {
         const waitForSelector = async (selector, timeoutMs = 5_000) => {
             const locator = context.page.locator(selector).first();
             await locator.waitFor({ timeout: timeoutMs, state: 'attached' });
@@ -120,38 +119,37 @@ export class PlaywrightCrawler extends BrowserCrawler {
         const downloads = [];
         context.page.on('download', (download) => downloads.push(download));
         return {
-            injectFile: async (filePath, options) => playwrightUtils.injectFile(context.page, filePath, options),
+            injectFile: async (filePath, options) => injectFile(context.page, filePath, options),
             injectJQuery: async () => {
                 if (context.request.state === RequestState.BEFORE_NAV) {
                     context.log.warning('Using injectJQuery() in preNavigationHooks leads to unstable results. Use it in a postNavigationHook or a requestHandler instead.');
-                    await playwrightUtils.injectJQuery(context.page);
+                    await injectJQuery(context.page);
                     return;
                 }
-                await playwrightUtils.injectJQuery(context.page, { surviveNavigations: false });
+                await injectJQuery(context.page, { surviveNavigations: false });
             },
-            blockRequests: async (options) => playwrightUtils.blockRequests(context.page, options),
+            blockRequests: async (options) => blockRequests(context.page, options),
             waitForSelector,
             parseWithCheerio: async (selector, timeoutMs = 5_000) => {
                 if (selector) {
                     await waitForSelector(selector, timeoutMs);
                 }
-                return playwrightUtils.parseWithCheerio(context.page, this.ignoreShadowRoots, this.ignoreIframes);
+                return parseWithCheerio(context.page, this.ignoreShadowRoots, this.ignoreIframes);
             },
-            infiniteScroll: async (options) => playwrightUtils.infiniteScroll(context.page, options),
+            infiniteScroll: async (options) => infiniteScroll(context.page, options),
             listDownloads: async () => downloads,
-            saveSnapshot: async (options) => playwrightUtils.saveSnapshot(context.page, {
+            saveSnapshot: async (options) => saveSnapshot(context.page, {
                 ...options,
                 configuration: serviceLocator.getConfiguration(),
             }),
-            enqueueLinksByClickingElements: async (options) => playwrightUtils.enqueueLinksByClickingElements({
+            enqueueLinksByClickingElements: async (options) => enqueueLinksByClickingElements({
                 ...options,
                 page: context.page,
                 requestManager: this.requestManager,
             }),
-            compileScript: (scriptString, ctx) => playwrightUtils.compileScript(scriptString, ctx),
-            closeCookieModals: async () => playwrightUtils.closeCookieModals(context.page),
+            compileScript: (scriptString, ctx) => compileScript(scriptString, ctx),
             handleCloudflareChallenge: async (options) => {
-                return playwrightUtils.handleCloudflareChallenge(context.page, context.request.url, options);
+                return handleCloudflareChallenge(context.page, context.request.url, options);
             },
         };
     }

package/internals/playwright-launcher.d.ts

@@ -3,6 +3,7 @@ import { BrowserLauncher, Configuration } from '@crawlee/browser';
 import { PlaywrightPlugin } from '@crawlee/browser-pool';
 // @ts-ignore optional peer dependency or compatibility with es2022
 import type { Browser, BrowserType, LaunchOptions } from 'playwright';
+import { z } from 'zod';
 /**
  * Apify extends the launch options of Playwright.
  * You can use any of the Playwright compatible
@@ -71,28 +72,36 @@ export interface PlaywrightLaunchContext extends BrowserLaunchContext<LaunchOpti
  */
 export declare class PlaywrightLauncher extends BrowserLauncher<PlaywrightPlugin> {
     readonly configuration: Configuration;
+    /**
+     * @internal
+     */
     protected static optionsShape: {
+        proxyUrl: z.ZodOptional<z.ZodURL>;
+        useChrome: z.ZodOptional<z.ZodBoolean>;
+        useIncognitoPages: z.ZodOptional<z.ZodBoolean>;
+        browserPerProxy: z.ZodOptional<z.ZodBoolean>;
+        ignoreProxyCertificate: z.ZodOptional<z.ZodBoolean>;
+        userDataDir: z.ZodOptional<z.ZodString>;
 // @ts-ignore optional peer dependency or compatibility with es2022
-        launcher: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        launchContextOptions: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        proxyUrl: import("ow").StringPredicate & import("ow").BasePredicate<string | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        useChrome: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
+        launchOptions: z.ZodOptional<z.ZodCustom<import("@crawlee/types").Dictionary, import("@crawlee/types").Dictionary>>;
+        userAgent: z.ZodOptional<z.ZodString>;
 // @ts-ignore optional peer dependency or compatibility with es2022
-        useIncognitoPages: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        browserPerProxy: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        ignoreProxyCertificate: import("ow").BooleanPredicate & import("ow").BasePredicate<boolean | undefined>;
-// @ts-ignore optional peer dependency or compatibility with es2022
-        userDataDir: import("ow").StringPredicate & import("ow").BasePredicate<string | undefined>;
+        launcher: z.ZodOptional<z.ZodCustom<import("@crawlee/types").Dictionary, import("@crawlee/types").Dictionary>>;
+    };
+    /** @internal */
+    protected static optionsSchema: z.ZodObject<{
+        proxyUrl: z.ZodOptional<z.ZodURL>;
+        useChrome: z.ZodOptional<z.ZodBoolean>;
+        useIncognitoPages: z.ZodOptional<z.ZodBoolean>;
+        browserPerProxy: z.ZodOptional<z.ZodBoolean>;
+        ignoreProxyCertificate: z.ZodOptional<z.ZodBoolean>;
+        userDataDir: z.ZodOptional<z.ZodString>;
 // @ts-ignore optional peer dependency or compatibility with es2022
-        launchOptions: import("ow").ObjectPredicate<object> & import("ow").BasePredicate<object | undefined>;
+        launchOptions: z.ZodOptional<z.ZodCustom<import("@crawlee/types").Dictionary, import("@crawlee/types").Dictionary>>;
+        userAgent: z.ZodOptional<z.ZodString>;
 // @ts-ignore optional peer dependency or compatibility with es2022
-        userAgent: import("ow").StringPredicate & import("ow").BasePredicate<string | undefined>;
-    };
+        launcher: z.ZodOptional<z.ZodCustom<import("@crawlee/types").Dictionary, import("@crawlee/types").Dictionary>>;
+    }, z.core.$strict>;
     /**
      * All `PlaywrightLauncher` parameters are passed via this launchContext object.
      */

package/internals/playwright-launcher.js

@@ -1,29 +1,35 @@
 import { BrowserLauncher, Configuration } from '@crawlee/browser';
 import { PlaywrightPlugin } from '@crawlee/browser-pool';
-import ow from 'ow';
+import { parseArgument, schemas } from '@crawlee/utils/internal';
+import { z } from 'zod';
 /**
  * `PlaywrightLauncher` is based on the `BrowserLauncher`. It launches `playwright` browser instance.
  * @ignore
  */
 export class PlaywrightLauncher extends BrowserLauncher {
     configuration;
+    /**
+     * @internal
+     */
     static optionsShape = {
         ...BrowserLauncher.optionsShape,
-        launcher: ow.optional.object,
-        launchContextOptions: ow.optional.object,
+        // Passthrough schema — the launcher module object must keep its prototype through parsing.
+        launcher: schemas.anyObject.optional(),
     };
+    /** @internal */
+    static optionsSchema = z.strictObject(PlaywrightLauncher.optionsShape);
     /**
      * All `PlaywrightLauncher` parameters are passed via this launchContext object.
      */
     constructor(launchContext = {}, configuration = Configuration.getGlobalConfiguration()) {
-        ow(launchContext, 'PlaywrightLauncherOptions', ow.object.exactShape(PlaywrightLauncher.optionsShape));
-        const { launcher = BrowserLauncher.requireLauncherOrThrow('playwright', 'apify/actor-node-playwright-*').chromium, } = launchContext;
-        const { launchOptions = {}, ...rest } = launchContext;
+        const parsedContext = parseArgument(launchContext, PlaywrightLauncher.optionsSchema, 'PlaywrightLaunchContext');
+        const { launcher = BrowserLauncher.requireLauncherOrThrow('playwright', 'apify/actor-node-playwright-*').chromium, } = parsedContext;
+        const { launchOptions = {}, ...rest } = parsedContext;
         super({
             ...rest,
             launchOptions: {
                 ...launchOptions,
-                executablePath: getDefaultExecutablePath(launchContext, configuration),
+                executablePath: getDefaultExecutablePath(parsedContext, configuration),
             },
             launcher,
         }, configuration);