@crawlee/playwright 4.0.0-beta.119 → 4.0.0-beta.120

package/internals/adaptive-playwright-crawler.d.ts

@@ -1,7 +1,7 @@
 import type { BrowserHook, LoadedRequest, Request, RouterHandler, RouteSchemas, RoutesFromSchemas } from '@crawlee/browser';
 import type { BasicCrawlerOptions } from '@crawlee/basic';
 import { BasicCrawler } from '@crawlee/basic';
-import type { ContextPipeline, CrawlingContext, EnqueueLinksOptions, GetUserDataFromRequest, RouterRoutes, StatisticsOptions, StatisticState, StorageTransactionView } from '@crawlee/core';
+import type { ContextPipeline, CrawlingContext, EnqueueLinksOptions, GetUserDataFromRequest, RouterRoutes, StatisticPersistedState, StatisticState, StorageTransactionView } from '@crawlee/core';
 import { Statistics } from '@crawlee/core';
 import type { Dictionary, Awaitable } from '@crawlee/types';
 import { type CheerioRoot } from '@crawlee/utils/internal';
@@ -16,11 +16,15 @@ interface AdaptivePlaywrightCrawlerStatisticState extends StatisticState {
     browserRequestHandlerRuns?: number;
     renderingTypeMispredictions?: number;
 }
+interface AdaptivePlaywrightCrawlerPersistedStatisticState extends StatisticPersistedState {
+    httpOnlyRequestHandlerRuns?: number;
+    browserRequestHandlerRuns?: number;
+    renderingTypeMispredictions?: number;
+}
 declare class AdaptivePlaywrightCrawlerStatistics extends Statistics {
-    state: AdaptivePlaywrightCrawlerStatisticState;
-    constructor(options?: StatisticsOptions);
-    reset(): void;
-    protected maybeLoadStatistics(): Promise<void>;
+    get state(): AdaptivePlaywrightCrawlerStatisticState;
+    protected defaultState(): AdaptivePlaywrightCrawlerStatisticState;
+    protected deserializeState(persistedState: AdaptivePlaywrightCrawlerPersistedStatisticState): AdaptivePlaywrightCrawlerStatisticState;
     trackHttpOnlyRequestHandlerRun(): void;
     trackBrowserRequestHandlerRun(): void;
     trackRenderingTypeMisprediction(): void;

package/internals/adaptive-playwright-crawler.js

@@ -9,26 +9,24 @@ import { addTimeoutToPromise } from '@apify/timeout';
 import { PlaywrightCrawler } from './playwright-crawler.js';
 import { RenderingTypePredictor, } from './utils/rendering-type-prediction.js';
 class AdaptivePlaywrightCrawlerStatistics extends Statistics {
-    state = null; // this needs to be assigned for a valid override, but the initialization is done by a reset() call from the parent constructor
-    constructor(options = {}) {
-        super(options);
-        this.reset();
+    get state() {
+        return super.state;
     }
-    reset() {
-        super.reset();
-        this.state.httpOnlyRequestHandlerRuns = 0;
-        this.state.browserRequestHandlerRuns = 0;
-        this.state.renderingTypeMispredictions = 0;
+    defaultState() {
+        return {
+            ...super.defaultState(),
+            httpOnlyRequestHandlerRuns: 0,
+            browserRequestHandlerRuns: 0,
+            renderingTypeMispredictions: 0,
+        };
     }
-    async maybeLoadStatistics() {
-        await super.maybeLoadStatistics();
-        const savedState = await this.keyValueStore?.getValue(this.persistStateKey);
-        if (!savedState) {
-            return;
-        }
-        this.state.httpOnlyRequestHandlerRuns = savedState.httpOnlyRequestHandlerRuns;
-        this.state.browserRequestHandlerRuns = savedState.browserRequestHandlerRuns;
-        this.state.renderingTypeMispredictions = savedState.renderingTypeMispredictions;
+    deserializeState(persistedState) {
+        return {
+            ...super.deserializeState(persistedState),
+            httpOnlyRequestHandlerRuns: persistedState.httpOnlyRequestHandlerRuns,
+            browserRequestHandlerRuns: persistedState.browserRequestHandlerRuns,
+            renderingTypeMispredictions: persistedState.renderingTypeMispredictions,
+        };
     }
     trackHttpOnlyRequestHandlerRun() {
         this.state.httpOnlyRequestHandlerRuns ??= 0;

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@crawlee/playwright",
-    "version": "4.0.0-beta.119",
+    "version": "4.0.0-beta.120",
     "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"
@@ -49,13 +49,13 @@
     "dependencies": {
         "@apify/datastructures": "^2.0.3",
         "@apify/timeout": "^0.4.4",
-        "@crawlee/basic": "4.0.0-beta.119",
-        "@crawlee/browser": "4.0.0-beta.119",
-        "@crawlee/browser-pool": "4.0.0-beta.119",
-        "@crawlee/cheerio": "4.0.0-beta.119",
-        "@crawlee/core": "4.0.0-beta.119",
-        "@crawlee/types": "4.0.0-beta.119",
-        "@crawlee/utils": "4.0.0-beta.119",
+        "@crawlee/basic": "4.0.0-beta.120",
+        "@crawlee/browser": "4.0.0-beta.120",
+        "@crawlee/browser-pool": "4.0.0-beta.120",
+        "@crawlee/cheerio": "4.0.0-beta.120",
+        "@crawlee/core": "4.0.0-beta.120",
+        "@crawlee/types": "4.0.0-beta.120",
+        "@crawlee/utils": "4.0.0-beta.120",
         "cheerio": "^1.0.0",
         "jquery": "^3.7.1",
         "ml-logistic-regression": "^2.0.0",
@@ -85,5 +85,5 @@
             }
         }
     },
-    "gitHead": "552fe2371a3c6e9e9f3514010536ea6abdc27eb4"
+    "gitHead": "da5d427c4c1c9dcaea95b151e9c4c7310100885d"
 }