yuanhau/news-analyze
1
0
Fork 0
mirror of https://github.com/hpware/news-analyze.git synced 2025-06-23 21:14:23 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

The Line Today scaper finally WORKS!! Now we just need make a UI for

Browse source 
it...
This commit is contained in:
yuanhau 2025-05-18 11:13:25 +08:00
parent 0bcb646cc0
commit fe5e2d996e
2 changed files with 7 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

1
server/api/news/get/lt/[slug].ts
View file

@ -2,6 +2,5 @@ import lineToday from "~/server/scrape/line_today";
export default defineEventHandler(async (event) => {
const slug = getRouterParam(event, "slug");
const data = await lineToday(slug);
console.log(data);
return data;
});

14
server/scrape/line_today.ts
View file

@ -1,5 +1,4 @@
import JSSoup from "jssoup";
//import cheerio from "cheerio";
import * as cheerio from "cheerio";
async function lineToday(slug: string) {
const url = "https://today.line.me/tw/v2/article/" + slug;
@ -21,16 +20,17 @@ async function lineToday(slug: string) {
const data = await fetchPageCode.text();
// 加 await? no.
// AHHH I NEED TO CHANGE TO SOMETHING ELSE.
const soup = new JSSoup(data, false);
const titlesoup = soup.find("h1", "entityTitle");
const title = titlesoup.text.replaceAll("\n", "");
const html = cheerio.load(data);
const title = html("h1.entityTitle").text().replaceAll("\n", "");
const article = soup.find("article", "news-content");
const paragraph = article.text;
const paragraph = html("article.news-content").text();
return {
title: title,
paragraph: paragraph,
};
}
// Texting on console only!
//console.log(await lineToday("oqmazXP"));
export default lineToday;