rsshub/lib/routes/nga/post.ts

Make RSS Great Again!

import { Route } from '@/types';
import got from '@/utils/got';
import { load } from 'cheerio';
import iconv from 'iconv-lite';
import { parseDate } from '@/utils/parse-date';
import timezone from '@/utils/timezone';
import { config } from '@/config';

export const route: Route = {
    path: '/post/:tid/:authorId?',
    categories: ['bbs'],
    example: '/nga/post/18449558',
    parameters: { tid: '帖子 id, 可在帖子 URL 找到', authorId: '作者 id' },
    features: {
        requireConfig: false,
        requirePuppeteer: false,
        antiCrawler: false,
        supportBT: false,
        supportPodcast: false,
        supportScihub: false,
    },
    name: '帖子',
    maintainers: ['xyqfer', 'syrinka'],
    handler,
};

async function handler(ctx) {
    const getPageUrl = (tid, authorId, page = 1, hash = '') => `https://nga.178.com/read.php?tid=${tid}&page=${page}${authorId ? `&authorid=${authorId}` : ''}&rand=${Math.random() * 1000}#${hash}`;
    const getPage = async (tid, authorId, pageId = 1) => {
        const link = getPageUrl(tid, authorId, pageId);
        const timestamp = Math.floor(Date.now() / 1000);
        let cookieString = `guestJs=${timestamp};`;
        if (config.nga.uid && config.nga.cid) {
            cookieString = `ngaPassportUid=${config.nga.uid}; ngaPassportCid=${config.nga.cid};`;
        }
        const response = await got(link, {
            responseType: 'buffer',
            headers: {
                Cookie: cookieString,
            },
        });

        const htmlString = iconv.decode(response.data, 'gbk');
        return load(htmlString);
    };

    const getLastPageId = async (tid, authorId) => {
        const $ = await getPage(tid, authorId);
        const nav = $('#pagebtop');
        const match = nav.html().match(/{0:'\/read\.php\?tid=(\d+).*?',1:(\d+),.*?}/);
        return match ? match[2] : 1;
    };

    const deepReplace = (str, pattern, replace) => {
        // 对于可能存在嵌套的样式一路 replace 到最深处
        while (pattern.test(str)) {
            str = str.replace(pattern, replace);
        }
        return str;
    };

    const formatContent = (str) => {
        // 简单样式
        str = deepReplace(str, /\[(b|u|i|del|code|sub|sup)](.+?)\[\/\1]/g, '<$1>$2</$1>');
        str = str
            .replaceAll(/\[dice](.+?)\[\/dice]/g, '<b>ROLL : $1</b>')
            .replaceAll(/\[color=(.+?)](.+?)\[\/color]/g, '<span style="color:$1;">$2</span>')
            .replaceAll(/\[font=(.+?)](.+?)\[\/font]/g, '<span style="font-family:$1;">$2</span>')
            .replaceAll(/\[size=(.+?)](.+?)\[\/size]/g, '<span style="font-size:$1;">$2</span>')
            .replaceAll(/\[align=(.+?)](.+?)\[\/align]/g, '<span style="text-align:$1;">$2</span>');
        // 列表
        str = deepReplace(str, /\[\*](.+?)(?=\[\*]|\[\/list])/g, '<li>$1</li>');
        str = deepReplace(str, /\[list](.+?)\[\/list]/g, '<ul>$1</ul>');
        // 图片
        str = str.replaceAll(/\[img](.+?)\[\/img]/g, (m, src) => `<img src='${src[0] === '.' ? 'https://img.nga.178.com/attachments' + src.slice(1) : src}'></img>`);
        // 折叠
        str = deepReplace(str, /\[collapse(?:=(.+?))?](.+?)\[\/collapse]/g, '<details><summary>$1</summary>$2</details>');
        // 引用
        str = deepReplace(str, /\[quote](.+?)\[\/quote]/g, '<blockquote>$1</blockquote>')
            .replaceAll(/\[@(.+?)]/g, '<a href="https://nga.178.com/nuke.php?func=ucp&username=$1">@$1</a>')
            .replaceAll(/\[uid=(\d+)](.+?)\[\/uid]/g, '<a href="https://nga.178.com/nuke.php?func=ucp&uid=$1">@$2</a>')
            .replaceAll(/\[tid=(\d+)](.+?)\[\/tid]/g, '<a href="https://nga.178.com/read.php?tid=$1">$2</a>')
            .replaceAll(/\[pid=(\d+),(\d+),(\d+)](.+?)\[\/pid]/g, (m, pid, tid, page, str) => {
                const url = `https://nga.178.com/read.php?tid=${tid}&page=${page}#pid${pid}Anchor`;
                return `<a href="${url}">${str}</a>`;
            });
        // 链接
        str = str.replaceAll(/\[url=(.+?)](.+?)\[\/url]/g, '<a href="$1">$2</a>');
        // 分割线
        str = str.replaceAll(/\[h](.+?)\[\/h]/g, '<h4 style="font-size:1.17em;font-weight:bold;border-bottom:1px solid #aaa;clear:both;margin:1.33em 0 0.2em 0;">$1</h4>');
        return str;
    };

    const tid = ctx.req.param('tid');
    const authorId = ctx.req.param('authorId') || undefined;
    const pageId = await getLastPageId(tid, authorId);

    const $ = await getPage(tid, authorId, pageId);
    const title = $('title').text() || '';
    const posterMap = JSON.parse(
        $('script')
            .text()
            .match(/commonui\.userInfo\.setAll\((.*)\)$/m)[1]
    );
    const authorName = authorId ? posterMap[authorId].username : undefined;

    const items = $('#m_posts_c')
        .children()
        .filter('table')
        .toArray()
        .map((post_) => {
            const post = $(post_);
            const posterId = post
                .find('.posterinfo a')
                .first()
                .attr('href')
                .match(/&uid=(-?\d+)$/)[1];
            const poster = authorName || posterMap[posterId].username;
            const content = post.find('.postcontent').first();
            const description = formatContent(content.html());
            const postId = content.attr('id');
            const link = getPageUrl(tid, authorId, pageId, postId);
            const pubDate = timezone(parseDate(post.find('.postInfo > span').first().text(), 'YYYY-MM-DD HH:mm'), +8);

            return {
                title: load(description).text(),
                author: poster,
                link,
                description,
                pubDate,
                guid: postId,
            };
        });

    const rssTitle = authorName ? `NGA ${authorName} ${title}` : `NGA ${title}`;

    return {
        title: rssTitle,
        link: getPageUrl(tid, authorId, pageId),
        item: items,
    };
}