🐛 fix: 修正返回结果导致插件无法正常识别的问题

src/plugins/webCrawler/runner.ts

@@ -1,6 +1,6 @@
 import { PluginRunner } from '@/plugins/type';
 
-import { Result } from './type';
+import { ParserResponse, Result } from './type';
 
 const BASE_URL = process.env.BROWSERLESS_URL ?? 'https://chrome.browserless.io';
 const BROWSERLESS_TOKEN = process.env.BROWSERLESS_TOKEN;
@@ -34,10 +34,12 @@ const runner: PluginRunner<{ url: string }, Result> = async ({ url }) => {
       method: 'POST',
     });
 
-    return await parseRes.json();
+    const { title, textContent, siteName } = (await parseRes.json()) as ParserResponse;
+
+    return { content: textContent, title, url, website: siteName };
   } catch (error) {
     console.error(error);
-    return { content: '抓取失败', errorMessage: (error as any).message };
+    return { content: '抓取失败', errorMessage: (error as any).message, url };
   }
 };
 

src/plugins/webCrawler/type.ts

@@ -1,5 +1,35 @@
 export type Result = {
-  content?: string;
+  content: string;
   title?: string;
-  url?: string;
+  url: string;
+  website?: string;
 };
+
+export interface ParserResponse {
+  /** author metadata */
+  byline: string;
+
+  /** HTML string of processed article content */
+  content: string;
+
+  /** content direction */
+  dir: string;
+
+  /** article description, or short excerpt from the content */
+  excerpt: string;
+
+  /** content language */
+  lang: string;
+
+  /** length of an article, in characters */
+  length: number;
+
+  /** name of the site */
+  siteName: string;
+
+  /** text content of the article, with all the HTML tags removed */
+  textContent: string;
+
+  /** article title */
+  title: string;
+}