feat: 为html/json转化为rss提供通用支持

docs/en/other.md

@@ -316,6 +316,80 @@ please refer to the [Notion API documentation](https://developers.notion.com/ref
 
 <RouteEn author="sbilly" example="/sans/summit_archive" path="/sans/summit_archive" />
 
+## Transformation
+
+Pass URL and transformation rules to convert HTML/JSON into RSS.
+
+### HTML
+
+Specify options (in the format of query string) in parameter `routeParams` parameter to extract data from HTML.
+
+| Key            | Meaning                                            | Accepted Values | Default                 |
+| -------------- | -------------------------------------------------- | --------------- | ----------------------- |
+| `title`        | The title of the RSS                               | `string`        | Extract from `<title>`  |
+| `item`         | The HTML elements as `item` using CSS selector     | `string`        | html                    |
+| `itemTitle`    | The HTML elements as `title` in `item` using CSS selector | `string` | `item` element          |
+| `itemTitleAttr` | The attributes of `title` element as title        | `string`        | Element text            |
+| `itemLink`     | The HTML elements as `link` in `item` using CSS selector | `string`  | `item` element          |
+| `itemLinkAttr` | The attributes of `link` element as link           | `string`        | `href`                  |
+| `itemDesc`     | The HTML elements as `descrption` in `item` using CSS selector | `string` | `item` element     |
+| `itemDescAttr` | The attributes of `descrption` element as description | `string`     | Element html            |
+
+<RouteEn author="ttttmr" example="/rsshub/transform/html/https%3A%2F%2Fwechat2rss.xlab.app%2Fposts%2Flist%2F/item=div%5Bclass%3D%27post%2Dcontent%27%5D%20p%20a" path="/rsshub/transform/html/:url/:routeParams" :paramsDesc="['`encodeURIComponent`ed URL address', 'Transformation rules, requires URL encode']" selfhost="1">
+
+Parameters parsing in the above example:
+
+| Parameter    | Value                                     |
+| ------------ | ----------------------------------------- |
+| `url`        | `https://wechat2rss.xlab.app/posts/list/` |
+| `routeParams`| `item=div[class='post-content'] p a`      |
+
+Parsing of `routeParams` parameter:
+
+| Parameter | Value                           |
+| --------- | ------------------------------- |
+| `item`    | `div[class='post-content'] p a` |
+
+</RouteEn>
+
+### JSON
+
+Specify options (in the format of query string) in parameter `routeParams` parameter to extract data from JSON.
+
+| Key         | Meaning                                  | Accepted Values | Default                                    |
+| ----------  | ----------------------------- ---------- | --------------- | ----------  ------------------------------ |
+| `title`     | The title of the RSS                     | `string`        | Extracted from home page of current domain |
+| `item`      | The JSON Path as `item` element          | `string`        | Entire JSON response                       |
+| `itemTitle` | The JSON Path as `title` in `item`       | `string`        | None                                       |
+| `itemLink`  | The JSON Path as `link` in `item`        | `string`        | None                                       |
+| `itemDesc`  | The JSON Path as `description` in `item` | `string`        | None                                       |
+
+::: tip Note
+
+JSON Path only supports format like `a.b.c`. if you need to access arrays, like `a[0].b`, you can write it as `a.0.b`.
+
+:::
+
+<RouteEn author="ttttmr" example="/rsshub/transform/json/https%3A%2F%2Fapi.github.com%2Frepos%2Fginuerzh%2Fgost%2Freleases/title=Gost%20releases&itemTitle=tag_name&itemLink=html_url&itemDesc=body" path="/rsshub/transform/json/:url/:routeParams" :paramsDesc="['`encodeURIComponent`ed URL address', 'Transformation rules, requires URL encode']" selfhost="1">
+
+Parameters parsing in the above example:
+
+| Parameter     | Value                                           |
+| ------------- | ----------------------------------------------- |
+| `url`         | `https://api.github.com/repos/ginuerzh/gost/releases` |
+| `routeParams` | `title=Gost releases&itemTitle=tag_name&itemLink=html_url&itemDesc=body` |
+
+Parsing of `routeParams` parameter:
+
+| Parameter    | Value            |
+| ------------ | ---------------- |
+| `title`      | `Gost releases`  |
+| `itemTitle`  | `tag_name`       |
+| `itemLink`   | `html_url`       |
+| `itemDesc`   | `body`           |
+
+</RouteEn>
+
 ## Trending Search Keyword Aggregator
 
 ### Aggregated Keyword Tracker

docs/other.md

@@ -1121,6 +1121,80 @@ type 为 all 时，category 参数不支持 cost 和 free
 
 <Route author="Fatpandac" example="/ems/apple/EZ319397281CN" path="/ems/apple/:id" :paramsDesc="['苹果邮件编号']"/>
 
+## 转换
+
+传递 URL 和转化规则，将 HTML/JSON 转换为 RSS
+
+### HTML
+
+在 `routeParams` 参数中以 query string 格式指定选项，可以控制提取数据
+
+| 键              | 含义                                                            | 接受的值 | 默认值                   |
+| --------------- | --------------------------------------------------------------- | -------- | ------------------------ |
+| `title`         | 指定 RSS 的标题                                                 | `string` | 从当前网页中取 `<title>` |
+| `item`          | 通过 CSS 选择器查找 HTML 元素作为 `item` 元素                   | `string` | html                     |
+| `itemTitle`     | 在 `item` 中通过 CSS 选择器查找 HTML 元素作为 `title` 元素      | `string` | `item` 元素              |
+| `itemTitleAttr` | 获取 `title` 元素属性作为标题                                   | `string` | 元素 text                |
+| `itemLink`      | 在 `item` 中通过 CSS 选择器查找 HTML 元素作为 `link` 元素       | `string` | `item` 元素              |
+| `itemLinkAttr`  | 获取 `link` 元素属性作为链接                                    | `string` | `href`                   |
+| `itemDesc`      | 在 `item` 中通过 CSS 选择器查找 HTML 元素作为 `descrption` 元素 | `string` | `item` 元素              |
+| `itemDescAttr`  | 获取 `descrption` 元素属性作为描述                              | `string` | 元素 html                |
+
+<Route author="ttttmr" example="/rsshub/transform/html/https%3A%2F%2Fwechat2rss.xlab.app%2Fposts%2Flist%2F/item=div%5Bclass%3D%27post%2Dcontent%27%5D%20p%20a" path="/rsshub/transform/html/:url/:routeParams" :paramsDesc="['URL地址，需经 URL 编码', '转换规则，需经 URL 编码']" selfhost="1">
+
+上述例子中参数解析如下
+
+| 参数           | 值                                        |
+| -------------- | ----------------------------------------- |
+| `:url`         | `https://wechat2rss.xlab.app/posts/list/` |
+| `:routeParams` | `item=div[class='post-content'] p a`      |
+
+`routeParams`参数解析如下
+
+| 参数   | 值                              |
+| ------ | ------------------------------- |
+| `item` | `div[class='post-content'] p a` |
+
+</Route>
+
+### JSON
+
+在 `routeParams` 参数中以 query string 格式指定选项，可以控制提取数据
+
+| 键          | 含义                                    | 接受的值 | 默认值                               |
+| ----------- | --------------------------------------- | -------- | ------------------------------------ |
+| `title`     | 指定 RSS 的标题                         | `string` | 从当前域名的根路径网页中取 `<title>` |
+| `item`      | 通过 JSON Path 查找作为 `item` 元素     | `string` | 整个响应 JSON                        |
+| `itemTitle` | 在 `item` 中通过 JSON Path 查找作为标题 | `string` | 无                                   |
+| `itemLink`  | 在 `item` 中通过 JSON Path 查找作为链接 | `string` | 无                                   |
+| `itemDesc`  | 在 `item` 中通过 JSON Path 查找作为描述 | `string` | 无                                   |
+
+::: tip 注意
+
+JSON Path 目前只支持例如 `a.b.c` 的形式，如果需要从数组中读取，例如 `a[0].b`，可以写成 `a.0.b`
+
+:::
+
+<Route author="ttttmr" example="/rsshub/transform/json/https%3A%2F%2Fapi.github.com%2Frepos%2Fginuerzh%2Fgost%2Freleases/title=Gost%20releases&itemTitle=tag_name&itemLink=html_url&itemDesc=body" path="/rsshub/transform/json/:url/:routeParams" :paramsDesc="['URL地址，需经 URL 编码', '转换规则，需经 URL 编码']" selfhost="1">
+
+上述例子中参数解析如下
+
+| 参数           | 值                                                                       |
+| -------------- | ------------------------------------------------------------------------ |
+| `:url`         | `https://api.github.com/repos/ginuerzh/gost/releases`                    |
+| `:routeParams` | `title=Gost releases&itemTitle=tag_name&itemLink=html_url&itemDesc=body` |
+
+`routeParams` 参数解析如下
+
+| 参数        | 值              |
+| ----------- | --------------- |
+| `title`     | `Gost releases` |
+| `itemTitle` | `tag_name`      |
+| `itemLink`  | `html_url`      |
+| `itemDesc`  | `body`          |
+
+</Route>
+
 ## 自如
 
 ### 房源

lib/maintainer.js

@@ -5,7 +5,7 @@ const { join } = require('path');
 // Presence Check
 for (const dir of fs.readdirSync(dirname)) {
     const dirPath = join(dirname, dir);
-    if (!fs.existsSync(join(dirPath, 'maintainer.js'))) {
+    if (fs.existsSync(join(dirPath, 'router.js')) && !fs.existsSync(join(dirPath, 'maintainer.js'))) {
         throw Error(`No maintainer.js in "${dirPath}".`);
     }
 }

lib/v2/altervista/radar.js

@@ -0,0 +1,13 @@
+module.exports = {
+    'altervista.org': {
+        _name: 'Altervista',
+        hyp3rlinx: [
+            {
+                title: 'hyp3rlinx blog',
+                docs: 'https://docs.rsshub.app/',
+                source: ['/'],
+                target: '/rsshub/transform/html/http%3A%2F%2Fhyp3rlinx.altervista.org%2F/item=table[border=%221%22]%20tr%20td%20a',
+            },
+        ],
+    },
+};

lib/v2/rsshub/maintainer.js

@@ -1,4 +1,6 @@
 module.exports = {
     '/routes/:lang?': ['DIYgod'],
     '/rsshub/sponsors': ['DIYgod'],
+    '/transform/html/:url/:routeParams': ['ttttmr'],
+    '/transform/json/:url/:routeParams': ['ttttmr'],
 };

lib/v2/rsshub/router.js

@@ -1,6 +1,6 @@
 module.exports = (router) => {
-    router.get('/rss', require('./routes')); // 弃用
-
     router.get('/routes/:lang?', require('./routes'));
     router.get('/sponsors', require('./sponsors'));
+    router.get('/transform/html/:url/:routeParams', require('./transform/html'));
+    router.get('/transform/json/:url/:routeParams', require('./transform/json'));
 };

lib/v2/rsshub/transform/html.js

@@ -0,0 +1,75 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const config = require('@/config').value;
+
+module.exports = async (ctx) => {
+    if (!config.feature.allow_user_supply_unsafe_domain) {
+        ctx.throw(403, `This RSS is disabled unless 'ALLOW_USER_SUPPLY_UNSAFE_DOMAIN' is set to 'true'.`);
+    }
+    const { url } = ctx.params;
+    const response = await got({
+        method: 'get',
+        url,
+    });
+
+    const routeParams = new URLSearchParams(ctx.params.routeParams);
+    const $ = cheerio.load(response.data);
+    const rssTitle = routeParams.get('title') ? routeParams.get('title') : $('title').text();
+    const item = routeParams.get('item') ? routeParams.get('item') : 'html';
+    const items = $(item)
+        .toArray()
+        .map((item) => {
+            try {
+                item = $(item);
+
+                let title;
+                const titleEle = routeParams.get('itemTitle') ? item.find(routeParams.get('itemTitle')) : item;
+                if (routeParams.get('itemTitleAttr')) {
+                    title = titleEle.attr(routeParams.get('itemTitleAttr'));
+                } else {
+                    title = titleEle.text();
+                }
+
+                let link;
+                const linkEle = routeParams.get('itemLink') ? item.find(routeParams.get('itemLink')) : item;
+                if (routeParams.get('itemLinkAttr')) {
+                    link = linkEle.attr(routeParams.get('itemLinkAttr'));
+                } else {
+                    if (linkEle.is('a')) {
+                        link = linkEle.attr('href');
+                    } else {
+                        link = linkEle.find('a').attr('href');
+                    }
+                }
+                // 补全绝对链接
+                link = link.trim();
+                if (link && !link.startsWith('http')) {
+                    link = `${new URL(url).origin}${link}`;
+                }
+
+                let desc;
+                const descEle = routeParams.get('itemDesc') ? item.find(routeParams.get('itemDesc')) : item;
+                if (routeParams.get('itemDescAttr')) {
+                    desc = descEle.attr(routeParams.get('itemDescAttr'));
+                } else {
+                    desc = descEle.html();
+                }
+
+                return {
+                    title,
+                    link,
+                    description: desc,
+                };
+            } catch (e) {
+                return null;
+            }
+        })
+        .filter(Boolean);
+
+    ctx.state.data = {
+        title: rssTitle,
+        link: url,
+        description: `Proxy ${url}`,
+        item: items,
+    };
+};

lib/v2/rsshub/transform/json.js

@@ -0,0 +1,57 @@
+const got = require('@/utils/got');
+const cheerio = require('cheerio');
+const config = require('@/config').value;
+
+function jsonGet(obj, attr) {
+    if (typeof attr !== 'string') {
+        return obj;
+    }
+    // a.b.c
+    // a.b[0].c => a.b.0.c
+    attr.split('.').forEach((key) => {
+        obj = obj[key];
+    });
+    return obj;
+}
+
+module.exports = async (ctx) => {
+    if (!config.feature.allow_user_supply_unsafe_domain) {
+        ctx.throw(403, `This RSS is disabled unless 'ALLOW_USER_SUPPLY_UNSAFE_DOMAIN' is set to 'true'.`);
+    }
+    const { url } = ctx.params;
+    const response = await got({
+        method: 'get',
+        url,
+    });
+
+    const routeParams = new URLSearchParams(ctx.params.routeParams);
+    let rssTitle = routeParams.get('title');
+    if (!rssTitle) {
+        const resp = await got({
+            method: 'get',
+            url: new URL(url).origin,
+        });
+        const $ = cheerio.load(resp.data);
+        rssTitle = $('title').text();
+    }
+
+    const items = jsonGet(response.data, routeParams.get('item')).map((item) => {
+        let link = jsonGet(item, routeParams.get('itemLink')).trim();
+        // 补全绝对链接
+        if (link && !link.startsWith('http')) {
+            link = `${new URL(url).origin}${link}`;
+        }
+        return {
+            title: jsonGet(item, routeParams.get('itemTitle')),
+            link,
+            description: routeParams.get('itemDesc') ? jsonGet(item, routeParams.get('itemDesc')) : '',
+        };
+    });
+
+    ctx.state.data = {
+        title: rssTitle,
+        link: url,
+        description: `Proxy ${url}`,
+        item: items,
+    };
+};

lib/v2/sec/radar.js

@@ -0,0 +1,13 @@
+module.exports = {
+    'sec.today': {
+        _name: '每日安全',
+        '.': [
+            {
+                title: '动态',
+                docs: 'https://docs.rsshub.app/',
+                source: ['/pulses', '/'],
+                target: '/rsshub/transform/html/https%3A%2F%2Fsec.today%2Fpulses%2F/item=div[class="card-body"]',
+            },
+        ],
+    },
+};