@crawlee/cheerio 3.7.1-beta.4 → 3.7.1-beta.6
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
|
@@ -1,5 +1,8 @@
|
|
|
1
|
+
// @ts-ignore optional peer dependency or compatibility with es2022
|
|
1
2
|
/// <reference types="node" />
|
|
3
|
+
// @ts-ignore optional peer dependency or compatibility with es2022
|
|
2
4
|
/// <reference types="node/http" />
|
|
5
|
+
// @ts-ignore optional peer dependency or compatibility with es2022
|
|
3
6
|
/// <reference types="node_modules/got/dist/source/core/timed-out" />
|
|
4
7
|
import type { IncomingMessage } from 'http';
|
|
5
8
|
import type { EnqueueLinksOptions, ErrorHandler, GetUserDataFromRequest, HttpCrawlerOptions, InternalHttpCrawlingContext, InternalHttpHook, RequestHandler, RouterRoutes, Configuration, RequestProvider } from '@crawlee/http';
|
|
@@ -123,6 +126,7 @@ export declare class CheerioCrawler extends HttpCrawler<CheerioCrawlingContext>
|
|
|
123
126
|
dom: unknown;
|
|
124
127
|
$: cheerio.CheerioAPI;
|
|
125
128
|
readonly body: string;
|
|
129
|
+
// @ts-ignore optional peer dependency or compatibility with es2022
|
|
126
130
|
enqueueLinks: (enqueueOptions?: EnqueueLinksOptions) => Promise<import("@crawlee/types").BatchAddRequestsResult>;
|
|
127
131
|
}>;
|
|
128
132
|
protected _parseHtmlToDom(response: IncomingMessage, isXml: boolean): Promise<unknown>;
|
|
@@ -136,6 +140,7 @@ interface EnqueueLinksInternalOptions {
|
|
|
136
140
|
finalRequestUrl?: string;
|
|
137
141
|
}
|
|
138
142
|
/** @internal */
|
|
143
|
+
// @ts-ignore optional peer dependency or compatibility with es2022
|
|
139
144
|
export declare function cheerioCrawlerEnqueueLinks({ options, $, requestQueue, originalRequestUrl, finalRequestUrl }: EnqueueLinksInternalOptions): Promise<import("@crawlee/types").BatchAddRequestsResult>;
|
|
140
145
|
/**
|
|
141
146
|
* Creates new {@apilink Router} instance that works based on request labels.
|
|
@@ -161,6 +166,7 @@ export declare function cheerioCrawlerEnqueueLinks({ options, $, requestQueue, o
|
|
|
161
166
|
* await crawler.run();
|
|
162
167
|
* ```
|
|
163
168
|
*/
|
|
169
|
+
// @ts-ignore optional peer dependency or compatibility with es2022
|
|
164
170
|
export declare function createCheerioRouter<Context extends CheerioCrawlingContext = CheerioCrawlingContext, UserData extends Dictionary = GetUserDataFromRequest<Context['request']>>(routes?: RouterRoutes<Context, UserData>): import("@crawlee/http").RouterHandler<Context>;
|
|
165
171
|
export {};
|
|
166
172
|
//# sourceMappingURL=cheerio-crawler.d.ts.map
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@crawlee/cheerio",
|
|
3
|
-
"version": "3.7.1-beta.
|
|
3
|
+
"version": "3.7.1-beta.6",
|
|
4
4
|
"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.",
|
|
5
5
|
"engines": {
|
|
6
6
|
"node": ">=16.0.0"
|
|
@@ -53,8 +53,8 @@
|
|
|
53
53
|
"access": "public"
|
|
54
54
|
},
|
|
55
55
|
"dependencies": {
|
|
56
|
-
"@crawlee/http": "3.7.1-beta.
|
|
57
|
-
"@crawlee/types": "3.7.1-beta.
|
|
56
|
+
"@crawlee/http": "3.7.1-beta.6",
|
|
57
|
+
"@crawlee/types": "3.7.1-beta.6",
|
|
58
58
|
"cheerio": "^1.0.0-rc.12",
|
|
59
59
|
"htmlparser2": "^9.0.0",
|
|
60
60
|
"tslib": "^2.4.0"
|
|
@@ -66,5 +66,5 @@
|
|
|
66
66
|
}
|
|
67
67
|
}
|
|
68
68
|
},
|
|
69
|
-
"gitHead": "
|
|
69
|
+
"gitHead": "01664e58799374a6acc2aa9db397566c1acc69d4"
|
|
70
70
|
}
|