Merge pull request #948 from DIYgod/master

[pull] master from diygod:master

docs/blog.md

@@ -152,6 +152,60 @@ username 为博主用户名，而非`xxx.hashnode.dev`中`xxx`所代表的 blog
 
 <RouteEn author="5upernova-heng" example="/macmenubar/recently/developer-apps,system-tools" path="/macmenubar/recently/:category?" :paramsDesc="['分类名，多个使用逗号隔开，留空则为全部。分类名可在 URL 中找到']" radar="1" />
 
+## Medium
+
+### List
+
+<Route author="ImSingee" example="/medium/list/imsingee/f2d8d48096a9" path="/medium/list/:user/:catalogId" :paramsDesc="['用户名', 'List 的 ID']">
+
+List ID 取的是网址中最后一部分 `-` 后面的内容，例如 `https://medium.com/@imsingee/list/collection-7e67004f23f9` 的用户名为 imsingee、ID 为 `7e67004f23f9`
+
+::: warning 注意
+
+想要获取 Private 的 List 则只支持自建
+
+:::
+
+</Route>
+
+### 个性推荐 - For You
+
+<Route author="ImSingee" example="/medium/for-you/imsingee" path="/medium/for-you/:user" :paramsDesc="['用户名']" selfhost="1">
+
+::: warning 注意
+
+个性推荐需要登录后的 Cookie 值，所以只能自建，详情见部署页面的配置模块。
+
+:::
+
+</Route>
+
+### 个性推荐 - Following
+
+<Route author="ImSingee" example="/medium/following/imsingee" path="/medium/following/:user" :paramsDesc="['用户名']" selfhost="1">
+
+::: warning 注意
+
+个性推荐需要登录后的 Cookie 值，所以只能自建，详情见部署页面的配置模块。
+
+:::
+
+</Route>
+
+### 个性推荐 - Tag
+
+<Route author="ImSingee" example="/medium/tag/imsingee/cybersecurity" path="/medium/tag/:user/:tag" :paramsDesc="['用户名', '订阅的 Tag']" selfhost="1">
+
+Tag 有很多，可从首页点进 Tag 以后的 URL 获取，例如 `https://medium.com/?tag=web3` 则 tag 为 `web3`
+
+::: warning 注意
+
+个性推荐需要登录后的 Cookie 值，所以只能自建，详情见部署页面的配置模块。
+
+:::
+
+</Route>
+
 ## Miris Whispers
 
 ### 博客

docs/en/blog.md

@@ -70,6 +70,60 @@ pageClass: routes
 
 <RouteEn author="5upernova-heng" example="/macmenubar/recently/developer-apps,system-tools" path="/macmenubar/recently/:category?" :paramsDesc="['Category path name, seperate by comma, default is all categories. Category path name can be found in url']" radar="1" />
 
+## Medium
+
+### List
+
+<RouteEn author="ImSingee" example="/medium/list/imsingee/f2d8d48096a9" path="/medium/list/:user/:catalogId" :paramsDesc="['Username', 'List ID']">
+
+The List ID is the last part of the URL after `-`, for example, the username in "https://medium.com/@imsingee/list/collection-7e67004f23f9" is `imsingee`, and the ID is `7e67004f23f9`.
+
+::: warning Note
+
+To access private lists, only self-hosting is supported.
+
+:::
+
+</RouteEn>
+
+### Personalized Recommendations - For You
+
+<RouteEn author="ImSingee" example="/medium/for-you/imsingee" path="/medium/for-you/:user" :paramsDesc="['Username']" selfhost="1">
+
+::: warning Note
+
+Personalized recommendations require the cookie value after logging in, so only self-hosting is supported. See the configuration module on the deployment page for details.
+
+:::
+
+</RouteEn>
+
+### Personalized Recommendations - Following
+
+<RouteEn author="ImSingee" example="/medium/following/imsingee" path="/medium/following/:user" :paramsDesc="['Username']" selfhost="1">
+
+::: warning Note
+
+Personalized recommendations require the cookie value after logging in, so only self-hosting is supported. See the configuration module on the deployment page for details.
+
+:::
+
+</RouteEn>
+
+### Personalized Recommendations - Tag
+
+<RouteEn author="ImSingee" example="/medium/tag/imsingee/cybersecurity" path="/medium/tag/:user/:tag" :paramsDesc="['Username', 'Subscribed Tag']" selfhost="1">
+
+There are many tags, which can be obtained by clicking on a tag from the homepage and looking at the URL. For example, if the URL is `https://medium.com/?tag=web3`, then the tag is `web3`.
+
+::: warning Note
+
+Personalized recommendations require the cookie value after logging in, so only self-hosting is supported. See the configuration module on the deployment page for details.
+
+:::
+
+</RouteEn>
+
 ## Miris Whispers
 
 ### Blog

docs/en/install/README.md

@@ -758,6 +758,11 @@ See docs of the specified route and `lib/config.js` for detailed information.
     -   `MASTODON_API_ACCESS_TOKEN`: user access token
     -   `MASTODON_API_ACCT_DOMAIN`: acct domain for particular instance
 
+-  Medium related routes: Open the console, copy the cookie (in theory, only uid and sid are required)
+	
+    -  `MEDIUM_ARTICLE_COOKIE`: Cookie used when requesting the full article, can access the full text of paid content when there is an active Member subscription.
+	-  `MEDIUM_COOKIE_{username}`: Cookie of the user corresponding to the username, required for personalized recommendation related routes.
+
 -   nhentai torrent: [Registration](https://nhentai.net/register/)
 
     -   `NHENTAI_USERNAME`: nhentai username or email

docs/en/joinus/debug.md

@@ -3,7 +3,9 @@ sidebarDepth: 0
 ---
 # Debugging
 
-When debugging your code, you can use more than just `console.log` or attaching the node process to a debugger. Another option is to use `ctx.state.json` to provide a custom object for debugging.
+When debugging your code, you can use more than just `console.log` or attaching the node process to a debugger. You can also use the following methods for debugging.
+
+Note: The following methods are only effective when the instance is running with `debugInfo=true`.
 
 ## Using `ctx.state.json`
 
@@ -24,3 +26,9 @@ ctx.state.json = {
 ```
 
 In the example above, we're passing the `info` object to `ctx.state.json`, which we can then access using the corresponding route + `.debug.json`.
+
+## debug.html
+
+In order to quickly test if the `description` in `ctx.state.data` is correct, you can use the `.debug.html` file suffix to obtain the HTML of the corresponding entry. The link can be directly opened in the browser to preview the rendering result.
+
+Usage: Access the corresponding route + `.{index}.debug.html`, where `{index}` is the item number (starting from 0) in your `ctx.state.data.item`. And the data corresponds to the `ctx.state.data.item[index].description` information will be returned as route result.

docs/en/parameter.md

@@ -148,7 +148,7 @@ E.g.
 -   JSON Feed - [https://rsshub.app/twitter/user/DIYgod.json](https://rsshub.app/twitter/user/DIYgod.json)
 -   Apply filters or URL query - [https://rsshub.app/dribbble/popular.atom?filterout=Blue|Yellow|Black](https://rsshub.app/dribbble/popular.atom?filterout=Blue|Yellow|Black)
 
-### Debug
+### debug.json
 
 If the RSSHub instance is running with `debugInfo=true` enabled, suffixing a route with `.debug.json` will result in the value of `ctx.state.json` being returned.
 
@@ -158,6 +158,15 @@ For example：
 
 -   `/furstar/characters/cn.debug.json`
 
+### debug.html
+
+By adding `.{index}.debug.html` (where `{index}` is a number starting from 0) at the end of the route and running the instance with `debugInfo=true`, RSSHub will return the content set in the plugin's `ctx.state.data.item[index].description`. You can access this page with a browser to quickly view the extracted information.
+
+Example:
+
+-  `/furstar/characters/cn.0.debug.html`
+
+
 ## Brief introduction
 
 Set the parameter `brief` to generate a brief pure-text introduction with a limited number of characters ( ≥ `100`).

docs/install/README.md

@@ -783,6 +783,11 @@ RSSHub 支持使用访问密钥 / 码，白名单和黑名单三种方式进行
     -   `MASTODON_API_ACCESS_TOKEN`: 用户 access token, 申请应用后，在应用配置页可以看到申请者的 access token
     -   `MASTODON_API_ACCT_DOMAIN`: 该实例本地用户 acct 标识的域名
 
+-   Medium 相关路由：打开控制台，复制 Cookie（理论上只需要 uid 和 sid 即可）
+
+    -   `MEDIUM_ARTICLE_COOKIE`：请求全文时使用的 Cookie，存在活跃的 Member 订阅时可获取付费内容全文
+    -   `MEDIUM_COOKIE_{username}`：对应 username 的用户的 Cookie，个性推荐相关路由需要
+
 -   MiniFlux 全部路由：
 
     -   `MINIFLUX_INSTANCE`： 用户所用的实例，默认为 MiniFlux 官方提供的 [付费服务地址](https://reader.miniflux.app)

docs/joinus/debug.md

@@ -3,7 +3,9 @@ sidebarDepth: 0
 ---
 # 调试
 
-当调试代码时，除了使用 `console.log` 或将 node 进程附加到调试器，您还可以将自定义对象提供给 `ctx.state.json` 来进行调试。
+当调试代码时，除了使用 `console.log` 或将 node 进程附加到调试器，您还可以使用如下方式进行调试。
+
+注意：需要实例运行在 `debugInfo=true` 的情况下以下方式才有效
 
 ## 使用 `ctx.state.json`
 
@@ -24,3 +26,9 @@ ctx.state.json = {
 ```
 
 在上面的示例中，我们将 `info` 对象传递给 `ctx.state.json`，然后可以使用相应的路由 + `.debug.json` 来访问它。
+
+## debug.html
+
+为了快速测试 `ctx.state.data` 中的 description 是否正确，你可以利用 `debug.html` 机制来获取相关条目的 HTML，该链接可以直接在浏览器中打开以预览渲染结果。
+
+使用方式：访问相应的路由 + `.{index}.debug.html`，其中 `{index}` 为你的 `ctx.state.data.item` 中的项目序号（从 0 开始），即返回对应路由结果中的 `ctx.state.data.item[index].description` 信息。

docs/parameter.md

@@ -147,7 +147,7 @@ RSSHub 同时支持 RSS 2.0、Atom 和 JSON Feed 输出格式，在路由末尾
 -   JSON Feed - <https://rsshub.app/twitter/user/DIYgod.json>
 -   和 filter 或其他 URL query 一起使用 - `https://rsshub.app/bilibili/user/coin/2267573.atom?filter=微小微|赤九玖|暴走大事件`
 
-### debug
+### debug.json
 
 在路由末尾添加 `.debug.json`且实例运行在`debugInfo=true`的情况下，RSShub 将会返回插件设置在`ctx.state.json`的内容
 
@@ -157,6 +157,14 @@ RSSHub 同时支持 RSS 2.0、Atom 和 JSON Feed 输出格式，在路由末尾
 
 -   `/furstar/characters/cn.debug.json`
 
+### debug.html
+
+在路由末尾添加 `.{index}.debug.html` （`{index}` 为数字，为从 0 开始的下标）且实例运行在 `debugInfo=true` 的情况下，RSShub 将会返回插件设置在 `ctx.state.data.item[index].description` 的内容，你可用浏览器访问该页面来快速查看提取的信息的展示结果。
+
+举例：
+
+-   `/furstar/characters/cn.0.debug.html`
+
 ## 输出简讯
 
 可以使用 `brief` 参数输出特定字数 ( ≥ `100` 字 ) 的纯文本内容

docs/university.md

@@ -3750,3 +3750,17 @@ jsjxy.hbut.edu.cn 证书链不全，自建 RSSHub 可设置环境变量 NODE_TLS
 ### 数据科学与计算机学院动态
 
 <Route author="Neutrino3316 MegrezZhu nczitzk" example="/sysu/cse" path="/sysu/cse"/>
+
+### 粤港澳发展研究院
+
+<Route author="TonyRL" example="/sysu/ygafz" path="/sysu/ygafz/:type?" :paramsDesc="['分类，见下表，默认为 `notice`']" radar="1" puppeteer="1">
+
+| 人才招聘   | 人才培养      | 新闻动态 | 通知公告 | 专家观点 |
+| ---------- | ------------- | -------- | -------- | -------- |
+| jobopening | personnelplan | news     | notice   | opinion  |
+
+| 研究成果 | 研究论文 | 学术著作 | 形势政策 |
+| -------- | -------- | -------- | -------- |
+| results  | papers   | writings | policy   |
+
+</Route>

lib/config.js

@@ -9,6 +9,7 @@ const calculateValue = () => {
     const twitter_tokens = {};
     const email_config = {};
     const discuz_cookies = {};
+    const medium_cookies = {};
 
     for (const name in envs) {
         if (name.startsWith('BILIBILI_COOKIE_')) {
@@ -23,6 +24,9 @@ const calculateValue = () => {
         } else if (name.startsWith('DISCUZ_COOKIE_')) {
             const cid = name.slice(14);
             discuz_cookies[cid] = envs[name];
+        } else if (name.startsWith('MEDIUM_COOKIE_')) {
+            const username = name.slice(14).toLowerCase();
+            medium_cookies[username] = envs[name];
         }
     }
 
@@ -208,6 +212,10 @@ const calculateValue = () => {
             accessToken: envs.MASTODON_API_ACCESS_TOKEN,
             acctDomain: envs.MASTODON_API_ACCT_DOMAIN,
         },
+        medium: {
+            cookies: medium_cookies,
+            articleCookie: envs.MEDIUM_ARTICLE_COOKIE || '',
+        },
         miniflux: {
             instance: envs.MINIFLUX_INSTANCE || 'https://reader.miniflux.app',
             token: envs.MINIFLUX_TOKEN || '',

lib/middleware/template.js

@@ -1,7 +1,7 @@
 const { art, json } = require('@/utils/render');
 const path = require('path');
 const config = require('@/config').value;
-const typeRegex = /\.(atom|rss|debug\.json|json)$/;
+const typeRegex = /\.(atom|rss|debug\.json|json|\d+\.debug\.html)$/;
 const { collapseWhitespace, convertDateToISO8601 } = require('@/utils/common-utils');
 
 module.exports = async (ctx, next) => {
@@ -26,6 +26,19 @@ module.exports = async (ctx, next) => {
             ctx.body = JSON.stringify({ message: 'plugin does not set debug json' });
         }
     }
+    if (outputType.endsWith('.debug.html') && config.debugInfo) {
+        ctx.set({
+            'Content-Type': 'text/html; charset=UTF-8',
+        });
+
+        const index = parseInt(outputType.match(/(\d+)\.debug\.html$/)[1]);
+        if (!(ctx.state.data && ctx.state.data.item && ctx.state.data.item[index])) {
+            ctx.body = `ctx.state.data.item[${index}] not found`;
+        } else {
+            ctx.body = ctx.state.data.item[index].description;
+        }
+    }
+
     if (outputType === 'json') {
         ctx.set({ 'Content-Type': 'application/feed+json; charset=UTF-8' });
     }

lib/v2/gov/ccdi/index.js

@@ -7,7 +7,7 @@ module.exports = async (ctx) => {
     pathname = pathname === '' ? defaultPath : pathname.endsWith('/') ? pathname : pathname + '/';
     const currentUrl = `${rootUrl}${pathname}`;
 
-    const { list, title } = await parseNewsList(currentUrl, '.list_news_dl li', ctx);
+    const { list, title } = await parseNewsList(currentUrl, '.list_news_dl2 li', ctx);
     const items = await Promise.all(list.map((item) => parseArticle(item, ctx)));
 
     ctx.state.data = {

lib/v2/gov/ccdi/utils.js

@@ -38,7 +38,7 @@ const parseNewsList = async (url, selector, ctx) => {
             return {
                 title: item.find('a').first().text().trim(),
                 link: new URL(item.find('a').first().attr('href'), url).href,
-                pubDate: parseDate(item.find('span').text(), 'YYYY-MM-DD'),
+                pubDate: parseDate(item.find('.more').text(), 'YYYY-MM-DD'),
             };
         });
     const title = $('.other_Location')

lib/v2/medium/following.js

@@ -0,0 +1,31 @@
+const config = require('@/config').value;
+
+const parseArticle = require('./parse-article.js');
+const { getFollowingFeedQuery } = require('./graphql.js');
+
+module.exports = async (ctx) => {
+    const user = ctx.params.user;
+
+    const cookie = config.medium.cookies[user];
+    if (cookie === undefined) {
+        throw Error(`缺少 Medium 用户 ${user} 登录后的 Cookie 值`);
+    }
+
+    const posts = await getFollowingFeedQuery(user, cookie);
+    ctx.state.json = posts;
+
+    if (!posts) {
+        // login failed
+        throw Error(`Medium 用户 ${user} 的 Cookie 无效或已过期`);
+    }
+
+    const urls = posts.items.map((data) => data.post.mediumUrl);
+
+    const parsedArticles = await Promise.all(urls.map((url) => parseArticle(ctx, url)));
+
+    ctx.state.data = {
+        title: `${user} Medium Following`,
+        link: 'https://medium.com/?feed=following',
+        item: parsedArticles,
+    };
+};

lib/v2/medium/for-you.js

@@ -0,0 +1,31 @@
+const config = require('@/config').value;
+
+const parseArticle = require('./parse-article.js');
+const { getWebInlineRecommendedFeedQuery } = require('./graphql.js');
+
+module.exports = async (ctx) => {
+    const user = ctx.params.user;
+
+    const cookie = config.medium.cookies[user];
+    if (cookie === undefined) {
+        throw Error(`缺少 Medium 用户 ${user} 登录后的 Cookie 值`);
+    }
+
+    const posts = await getWebInlineRecommendedFeedQuery(user, cookie);
+    ctx.state.json = posts;
+
+    if (!posts) {
+        // login failed
+        throw Error(`Medium 用户 ${user} 的 Cookie 无效或已过期`);
+    }
+
+    const urls = posts.items.map((data) => data.post.mediumUrl);
+
+    const parsedArticles = await Promise.all(urls.map((url) => parseArticle(ctx, url)));
+
+    ctx.state.data = {
+        title: `${user} Medium For You`,
+        link: 'https://medium.com/',
+        item: parsedArticles,
+    };
+};