feat(route): add 公众号 360 (#9457)

Signed-off-by: Rongrong <15956627+Rongronggg9@users.noreply.github.com>

docs/new-media.md

@@ -1855,6 +1855,32 @@ area 分区选项
 
 <Route author="luyuhuang" example="/ofweek/news" path="/ofweek/news"/>
 
+## 公众号 360
+
+### 公众号
+
+<Route author="Rongronggg9" example="/gzh360/gzh/北京青年报" path="/gzh360/gzh/:name" :paramsDesc="['公众号名，也可以是公众号 360 的内部 id']" radar="1" />
+
+### 分类
+
+<Route author="Rongronggg9" example="/gzh360/category/5d357964e2eb992114a3d588" path="/gzh360/category/:id?" :paramsDesc="['分类 id，见下表']" radar="1">
+
+| `id`                       | 分类  |   | `id`                       | 分类 |
+| -------------------------- | --- | - | -------------------------- | -- |
+|                            | 首页  |   | `5d357ae6e2eb992114a3d592` | 育儿 |
+| `5d357964e2eb992114a3d588` | 热门  |   | `5d357b00e2eb992114a3d593` | 旅游 |
+| `5d3579a2e2eb992114a3d589` | 搞笑  |   | `5d357b17e2eb992114a3d594` | 职场 |
+| `5d3579b0e2eb992114a3d58a` | 健康  |   | `5d357b34e2eb992114a3d595` | 美食 |
+| `5d3579bae2eb992114a3d58b` | 私房话 |   | `5d357b4ae2eb992114a3d596` | 历史 |
+| `5d357a10e2eb992114a3d58c` | 八卦精 |   | `5d357b60e2eb992114a3d597` | 教育 |
+| `5d357a4ae2eb992114a3d58d` | 科技咖 |   | `5d357b76e2eb992114a3d598` | 星座 |
+| `5d357a72e2eb992114a3d58e` | 财经迷 |   | `5d357b8de2eb992114a3d599` | 体育 |
+| `5d357a8be2eb992114a3d58f` | 汽车控 |   | `5d357b9be2eb992114a3d59a` | 军事 |
+| `5d357aa1e2eb992114a3d590` | 生活家 |   | `5d357bc2e2eb992114a3d59b` | 游戏 |
+| `5d357ab6e2eb992114a3d591` | 时尚圈 |   | `5d357bd4e2eb992114a3d59c` | 萌宠 |
+
+</Route>
+
 ## 谷歌新闻
 
 ### 新闻
@@ -3372,6 +3398,10 @@ column 为 third 时可选的 category:
 
 <Route author="TonyRL Rongronggg9" example="/wechat/feeddd/6131e1441269c358aa0e2141" path="/wechat/feeddd/:id" :paramsDesc="['公众号 id, 打开 `https://feeddd.org/feeds` 或 `https://cdn.jsdelivr.net/gh/feeddd/feeds/feeds_all_rss.txt`, 在 URL 中找到 id; 注意不是公众号页的 id, 而是订阅的 id']"/>
 
+### 公众号 (公众号 360 来源)
+
+见 [#公众号 360](#gong-zhong-hao-360)
+
 ### 公众号栏目 (非推送 & 历史消息)
 
 <Route author="MisteryMonster" example="/wechat/mp/homepage/MzA3MDM3NjE5NQ==/16" path="/wechat/mp/homepage/:biz/:hid/:cid?" :paramsDesc="['公众号id', '分页id', '页内栏目']" radar="1" rssbud="1" anticrawler="1">

lib/v2/gzh360/category.js

@@ -0,0 +1,8 @@
+const universal = require('./universal');
+
+module.exports = async (ctx) => {
+    const path = 'category';
+    const id = ctx.params.id ?? '';
+    const titleHeader = '公众号 360 - ';
+    await universal(ctx, path, id, titleHeader);
+};

lib/v2/gzh360/gzh.js

@@ -0,0 +1,18 @@
+const { mpIdEncode, mpIdDecode } = require('./utils');
+const universal = require('./universal');
+
+module.exports = async (ctx) => {
+    const path = 'gzh_articles';
+
+    let name = ctx.params.name;
+    let id = name;
+
+    // in order to support RSSHub Radar, we need to accept both name and id
+    try {
+        name = mpIdDecode(name);
+    } catch {
+        id = mpIdEncode(name);
+    }
+
+    await universal(ctx, path, id, '', name, true, true);
+};

lib/v2/gzh360/maintainer.js

@@ -0,0 +1,4 @@
+module.exports = {
+    '/category/:id?': ['Rongronggg9'],
+    '/gzh/:name?': ['Rongronggg9'],
+};

lib/v2/gzh360/radar.js

@@ -0,0 +1,19 @@
+module.exports = {
+    'gzh360.com': {
+        _name: '公众号360',
+        web: [
+            {
+                title: '公众号',
+                docs: 'https://docs.rsshub.app/new-media.html#gong-zhong-hao-360',
+                source: ['/gzh_articles', '/gzh', '/'],
+                target: (params, url) => `/gzh360/gzh/${new URL(url).searchParams.get('id') ?? ''}`,
+            },
+            {
+                title: '分类',
+                docs: 'https://docs.rsshub.app/new-media.html#gong-zhong-hao-360',
+                source: ['/category', '/'],
+                target: (params, url) => `/gzh360/category/${new URL(url).searchParams.get('id') ?? ''}`,
+            },
+        ],
+    },
+};

lib/v2/gzh360/router.js

@@ -0,0 +1,4 @@
+module.exports = function (router) {
+    router.get('/category/:id?', require('./category'));
+    router.get('/gzh/:name', require('./gzh'));
+};

lib/v2/gzh360/universal.js

@@ -0,0 +1,53 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const { parseDate } = require('@/utils/parse-date');
+const { finishArticleItem } = require('./utils');
+const config = require('@/config').value;
+
+const getInitEntry = async (url) =>
+    await got(url)
+        .then((_r) => _r.data)
+        .catch((err) => {
+            if (err.response.statusCode === 404) {
+                throw new Error('This category / WeChat Official Account is not found on GZH360.');
+            }
+            throw err;
+        });
+
+module.exports = async (ctx, path, id, titleHeader = '', custom_title = null, skipAuthor = false, cacheInitEntry = false) => {
+    const rootUrl = 'http://web.gzh360.com';
+    const currentUrl = `${rootUrl}/${id ? `${path}?id=${id}` : ''}`;
+
+    const respData = cacheInitEntry ? await ctx.cache.tryGet(currentUrl, async () => await getInitEntry(currentUrl), config.cache.routeExpire, false) : await getInitEntry(currentUrl);
+    const $ = cheerio.load(respData);
+
+    const title = id ? $('head > title').text().split(' - ', 1)[0] : '首页';
+
+    let items = $('div.content div.news_desc')
+        .map((_, item) => {
+            item = $(item);
+            const link = item.find('h3 > a');
+            const href = link.attr('href');
+            const title = link.text();
+            const pubDate = item.find('span.datecss > span').attr('data-timestamp');
+            let author;
+            if (!skipAuthor) {
+                author = item.find('span.fromcss > a').text(); // only some pages have this
+            }
+            return {
+                link: `${href.startsWith('http') ? '' : rootUrl}${href}`,
+                title,
+                pubDate: parseDate(pubDate),
+                author,
+            };
+        })
+        .get();
+
+    items = await Promise.all(items.map((item) => finishArticleItem(ctx, item, skipAuthor)));
+
+    ctx.state.data = {
+        title: `${titleHeader}${custom_title ?? title}`,
+        link: currentUrl,
+        item: items.filter((item) => item),
+    };
+};

lib/v2/gzh360/utils.js

@@ -0,0 +1,61 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const { parseDate } = require('@/utils/parse-date');
+
+const invalidIdError = new RangeError('Invalid id');
+
+const mpIdEncode = (name) => {
+    const onePassed = Buffer.from(name).toString('base64'); // one-pass uses the standard base64 alphabet
+    // two-passed base64 is always url-safe (`7f+/` never appear), you can prove it by yourself
+    return Buffer.from(onePassed).toString('base64');
+};
+
+const mpIdDecode = (id) => {
+    // verify that the decoded name can be a valid WeChat mp name: https://kf.qq.com/faq/120911VrYVrA141110r2MRJV.html
+    // "公众号名称/昵称可设置4-30个字符（1个汉字算2字符）": 4 ASCII characters => 12, 15 Chinese characters => 80
+    // and the id is a valid two-passed base64 string (`7f+/` never appear)
+    if (id.length < 12 || id.length > 80 || id.length % 4 !== 0 || !/^[a-eg-zA-Z0-68-9]+={0,2}$/.test(id)) {
+        throw invalidIdError;
+    }
+    const deSecondPassed = Buffer.from(id, 'base64').toString();
+    // verify that it is a valid base64 string using standard base64 alphabet
+    if (deSecondPassed.length % 4 !== 0 || !/^[a-zA-Z0-9+/]+={0,2}$/.test(deSecondPassed)) {
+        throw invalidIdError;
+    }
+    const deFirstPassed = Buffer.from(deSecondPassed, 'base64').toString();
+    // "空格不可在最前或者最后，且空格不可连续"
+    if (deFirstPassed.length < 2 || deFirstPassed.startsWith(' ') || deFirstPassed.endsWith(' ') || deFirstPassed.includes('  ')) {
+        throw invalidIdError;
+    }
+    return deFirstPassed;
+};
+
+const finishArticleItem = async (ctx, item, skipAuthor = false) => {
+    // the website is slow and unstable, so we need to ignore errors to avoid breaking the whole route
+    // we can't cache the item instead of the webpage because we need to retry when the last request failed
+    const article = await ctx.cache.tryGet(
+        item.link,
+        async () =>
+            await got(item.link)
+                .then((_r) => _r.data)
+                .catch(() => null) // it is safe do that in tryGet because a false value always lead to a cache miss
+    );
+
+    if (article) {
+        const $ = cheerio.load(article);
+        if (!skipAuthor) {
+            item.author = item.author || $('[id=webbdgzh]+a').text();
+        }
+        item.title = item.title || $('div.desc > h1').text();
+        item.pubDate = item.pubDate || parseDate($('div.desc span[data-timestamp]').attr('data-timestamp'));
+        item.description = $('div.rich_media_content').html(); // sometimes it is an empty string due to the website's fault
+    }
+
+    return item;
+};
+
+module.exports = {
+    mpIdEncode,
+    mpIdDecode,
+    finishArticleItem,
+};