@crawlee/cheerio 4.0.0-beta.230 → 4.0.0-beta.232

package/internals/cheerio-crawler.d.ts

@@ -1,6 +1,8 @@
 import type { CrawlingContext, DOMCrawlingContext, ErrorHandler, GetUserDataFromRequest, HttpCrawlerOptions, RequestHandler, RouterHandler, RouterRoutes, RouteSchemas, RoutesFromSchemas } from '@crawlee/http';
 import { DOMCrawler } from '@crawlee/http';
 import type { Dictionary } from '@crawlee/types';
+import type { Cheerio } from 'cheerio';
+import type { AnyNode } from 'domhandler';
 import type { CheerioParseResult } from './cheerio-parser.js';
 export type CheerioErrorHandler<UserData extends Dictionary = any, // with default to Dictionary we cant use a typed router in untyped crawler
 JSONData extends Dictionary = any, // with default to Dictionary we cant use a typed router in untyped crawler
@@ -10,7 +12,7 @@ JSONData extends Dictionary = any, // with default to Dictionary we cant use a t
 Routes extends Record<keyof Routes, Dictionary> = Record<string, UserData>, StatisticStateExtension extends object = {}> extends Omit<HttpCrawlerOptions<CheerioCrawlingContext<UserData, JSONData>, ContextExtension, ExtendedContext, Routes, StatisticStateExtension>, 'contextPipelineBuilder'> {
 }
 export interface CheerioCrawlingContext<UserData extends Dictionary = any, // with default to Dictionary we cant use a typed router in untyped crawler
-JSONData extends Dictionary = any> extends DOMCrawlingContext<CheerioParseResult, UserData, JSONData> {
+JSONData extends Dictionary = any> extends DOMCrawlingContext<CheerioParseResult, Cheerio<AnyNode>, Cheerio<AnyNode>, UserData, JSONData> {
 }
 export type CheerioRequestHandler<UserData extends Dictionary = any, // with default to Dictionary we cant use a typed router in untyped crawler
 JSONData extends Dictionary = any> = RequestHandler<CheerioCrawlingContext<UserData, JSONData>>;
@@ -93,7 +95,7 @@ JSONData extends Dictionary = any> = RequestHandler<CheerioCrawlingContext<UserD
  * ```
  * @category Crawlers
  */
-export declare class CheerioCrawler<ContextExtension = Dictionary<never>, ExtendedContext extends CheerioCrawlingContext = CheerioCrawlingContext & ContextExtension, Routes extends Record<keyof Routes, Dictionary> = Record<string, GetUserDataFromRequest<CheerioCrawlingContext['request']>>, StatisticStateExtension extends object = {}> extends DOMCrawler<CheerioParseResult, ContextExtension, ExtendedContext, Routes, StatisticStateExtension> {
+export declare class CheerioCrawler<ContextExtension = Dictionary<never>, ExtendedContext extends CheerioCrawlingContext = CheerioCrawlingContext & ContextExtension, Routes extends Record<keyof Routes, Dictionary> = Record<string, GetUserDataFromRequest<CheerioCrawlingContext['request']>>, StatisticStateExtension extends object = {}> extends DOMCrawler<CheerioParseResult, Cheerio<AnyNode>, Cheerio<AnyNode>, ContextExtension, ExtendedContext, Routes, StatisticStateExtension> {
     /**
      * All `CheerioCrawler` parameters are passed via an options object.
      */

package/internals/cheerio-parser.d.ts

@@ -1,5 +1,6 @@
 import type { DOMParser } from '@crawlee/http';
-import type { CheerioAPI } from 'cheerio';
+import type { Cheerio, CheerioAPI } from 'cheerio';
+import type { AnyNode } from 'domhandler';
 export interface CheerioParseResult {
     $: CheerioAPI;
     body: string;
@@ -8,4 +9,4 @@ export interface CheerioParseResult {
  * A {@link DOMParser} backed by [cheerio](https://www.npmjs.com/package/cheerio). Pass it to a
  * {@link DOMCrawler} to get the crawling context {@link CheerioCrawler} provides.
  */
-export declare function cheerioParser(): DOMParser<CheerioParseResult>;
+export declare function cheerioParser(): DOMParser<CheerioParseResult, Cheerio<AnyNode>, Cheerio<AnyNode>>;

package/internals/cheerio-parser.js

@@ -21,7 +21,11 @@ export function cheerioParser() {
             return { $, body };
         },
         extractLinks: ({ $ }, selector, baseUrl) => extractUrlsFromCheerio($, selector, baseUrl),
-        select: ({ $ }, selector) => $(selector).get(),
+        querySelectorAll: ({ $ }, selector) => $(selector),
+        querySelectorOne: ({ $ }, selector) => {
+            const match = $(selector).first();
+            return match.length ? match : undefined;
+        },
         toCheerio: ({ $ }) => $,
     };
 }

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "@crawlee/cheerio",
-    "version": "4.0.0-beta.230",
+    "version": "4.0.0-beta.232",
     "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": ">=22.13.0"
@@ -47,10 +47,11 @@
         "access": "public"
     },
     "dependencies": {
-        "@crawlee/http": "4.0.0-beta.230",
-        "@crawlee/types": "4.0.0-beta.230",
-        "@crawlee/utils": "4.0.0-beta.230",
+        "@crawlee/http": "4.0.0-beta.232",
+        "@crawlee/types": "4.0.0-beta.232",
+        "@crawlee/utils": "4.0.0-beta.232",
         "cheerio": "^1.0.0",
+        "domhandler": "^6.0.0",
         "htmlparser2": "^10.0.0",
         "tslib": "^2.8.1"
     },
@@ -61,5 +62,5 @@
             }
         }
     },
-    "gitHead": "9af978adbda8c2a750920e35b50ae19d6e940ddd"
+    "gitHead": "5eab8f877242d46d7cae13b640d8e1adbf194911"
 }