From afe6a78939dca0c1ef7f4bea89a1bd777c854b0b Mon Sep 17 00:00:00 2001
From: Andvari <31068367+dzx-dzx@users.noreply.github.com>
Date: Mon, 16 Oct 2023 20:46:58 +0800
Subject: [PATCH] feat(route/caixin): Fetch all blog articles. (#13522)

* feat(route/caixin): Fetch all blog articles.

* .

* Update doc.

* Update lib/v2/caixin/blog.js

* .

---------
---
 lib/v2/caixin/blog.js       | 120 +++++++++++++++++++++---------------
 lib/v2/caixin/router.js     |   2 +-
 website/docs/routes/blog.md |   2 +-
 3 files changed, 73 insertions(+), 51 deletions(-)

diff --git a/lib/v2/caixin/blog.js b/lib/v2/caixin/blog.js
index b4aa1cc082..6f9c6dc053 100644
--- a/lib/v2/caixin/blog.js
+++ b/lib/v2/caixin/blog.js
@@ -5,54 +5,76 @@ const { parseDate } = require('@/utils/parse-date');
 const { parseBlogArticle } = require('./utils');
 
 module.exports = async (ctx) => {
-    const { column } = ctx.params;
-    if (!isValidHost(column)) {
-        throw Error('Invalid column');
+    const { column, limit = 20 } = ctx.params;
+    if (!column) {
+        const { data } = await got('https://blog.caixin.com/blog-api/post/index', {
+            searchParams: {
+                page: 1,
+                size: limit,
+            },
+        }).json();
+        const posts = data.map((item) => ({
+            title: item.title,
+            description: item.brief,
+            author: item.authorName,
+            link: item.postUrl.replace('http://', 'https://'),
+            pubDate: parseDate(item.publishTime, 'x'),
+        }));
+        const items = await Promise.all(posts.map((item) => parseBlogArticle(item, ctx.cache.tryGet)));
+
+        ctx.state.data = {
+            title: `财新博客 - 全部`,
+            link: 'https://blog.caixin.com',
+            // description: introduce,
+            // image: avatar,
+            item: items,
+        };
+    } else {
+        if (!isValidHost(column)) {
+            throw Error('Invalid column');
+        }
+        const link = `https://${column}.blog.caixin.com`;
+        const { data: response } = await got(link);
+        const $ = cheerio.load(response);
+        const user = $('div.indexMainConri > script[type="text/javascript"]')
+            .text()
+            .substring('window.user = '.length + 1)
+            .split(';')[0]
+            .replaceAll(/\s/g, '');
+        const authorId = user.match(/id:"(\d+)"/)[1];
+        const authorName = user.match(/name:"(.*?)"/)[1];
+        const avatar = user.match(/avatar:"(.*?)"/)[1];
+        const introduce = user.match(/introduce:"(.*?)"/)[1];
+
+        const {
+            data: { data },
+        } = await got('https://blog.caixin.com/blog-api/post/posts', {
+            searchParams: {
+                page: 1,
+                size: ctx.query.limit ? parseInt(ctx.query.limit) : 20,
+                content: '',
+                authorId,
+                sort: 'publishTime',
+                direction: 'DESC',
+            },
+        });
+
+        const posts = data.map((item) => ({
+            title: item.title,
+            description: item.brief,
+            author: item.displayName,
+            link: item.guid.replace('http://', 'https://'),
+            pubDate: parseDate(item.publishTime, 'x'),
+        }));
+
+        const items = await Promise.all(posts.map((item) => parseBlogArticle(item, ctx.cache.tryGet)));
+
+        ctx.state.data = {
+            title: `财新博客 - ${authorName}`,
+            link,
+            description: introduce,
+            image: avatar,
+            item: items,
+        };
     }
-    const link = `https://${column}.blog.caixin.com`;
-    const { data: response } = await got(link);
-    const $ = cheerio.load(response);
-    const authorId = $('script[type="text/javascript"]')
-        .text()
-        .match(/window\.authorId = (\d+);/)[1];
-    const authorName = $('script[type="text/javascript"]')
-        .text()
-        .match(/window\.authorName = "(.*)";/)[1];
-    const avatar = $('script[type="text/javascript"]')
-        .text()
-        .match(/avatar: "(.*?)",/)[1];
-    const introduce = $('script[type="text/javascript"]')
-        .text()
-        .match(/introduce: "(.*?)"\n/)[1];
-
-    const {
-        data: { data },
-    } = await got('https://blog.caixin.com/blog-api/post/posts', {
-        searchParams: {
-            page: 1,
-            size: ctx.query.limit ? parseInt(ctx.query.limit) : 20,
-            content: '',
-            authorId,
-            sort: 'publishTime',
-            direction: 'DESC',
-        },
-    });
-
-    const posts = data.map((item) => ({
-        title: item.title,
-        description: item.brief,
-        author: item.displayName,
-        link: item.guid.replace('http://', 'https://'),
-        pubDate: parseDate(item.publishTime, 'x'),
-    }));
-
-    const items = await Promise.all(posts.map((item) => parseBlogArticle(item, ctx.cache.tryGet)));
-
-    ctx.state.data = {
-        title: `财新博客 - ${authorName}`,
-        link,
-        description: introduce,
-        image: avatar,
-        item: items,
-    };
 };
diff --git a/lib/v2/caixin/router.js b/lib/v2/caixin/router.js
index 1ec1c676c4..962a36a4f7 100644
--- a/lib/v2/caixin/router.js
+++ b/lib/v2/caixin/router.js
@@ -1,6 +1,6 @@
 module.exports = (router) => {
     router.get('/article', require('./article'));
-    router.get('/blog/:column', require('./blog'));
+    router.get('/blog/:column?', require('./blog'));
     router.get('/database', require('./database'));
     router.get('/k', require('./k'));
     router.get('/latest', require('./latest'));
diff --git a/website/docs/routes/blog.md b/website/docs/routes/blog.md
index 1fa2692325..5265ee5f1a 100644
--- a/website/docs/routes/blog.md
+++ b/website/docs/routes/blog.md
@@ -373,7 +373,7 @@ Limit the number of entries to be retrieved by adding `?limit=x` to the end of t
 
 ### 用户博客 {#cai-xin-bo-ke-yong-hu-bo-ke}
 
-<Route author="Maecenas" example="/caixin/blog/zhangwuchang" path="/caixin/blog/:column" paramsDesc={['博客名称，可在博客主页的 URL 找到']}>
+<Route author="Maecenas" example="/caixin/blog/zhangwuchang" path="/caixin/blog/:column?" paramsDesc={['博客名称，可在博客主页的 URL 找到']}>
 
 通过提取文章全文，以提供比官方源更佳的阅读体验.