From 876c7be15f67a6bca6020adcbfbbc75fdea3165a Mon Sep 17 00:00:00 2001
From: cerebrater <49020242+cerebrater@users.noreply.github.com>
Date: Fri, 30 Oct 2020 00:36:38 +0800
Subject: [PATCH] feat: add gwern (#6021)

---
 docs/blog.md                 |  4 +++
 lib/router.js                |  3 ++
 lib/routes/gwern/category.js | 58 ++++++++++++++++++++++++++++++++++++
 3 files changed, 65 insertions(+)
 create mode 100644 lib/routes/gwern/category.js

diff --git a/docs/blog.md b/docs/blog.md
index 831f8b83d4..ca5ba72b73 100644
--- a/docs/blog.md
+++ b/docs/blog.md
@@ -16,6 +16,10 @@ pageClass: routes
 
 <Route author="hoilc" example="/google/sites/outlierseconomics" path="/google/sites/:id" :paramsDesc="['Site ID, 可在 URL 中找到']" radar="1" rssbud="1"/>
 
+## Gwern Branwen
+
+<Route author="cerebrater" example="/gwern/newest" path="/gwern/:category" :paramsDesc="['網誌主頁的分類訊息']"/>
+
 ## Hedwig.pub
 
 <Route author="zwithz" example="/blogs/hedwig/zmd" path="/blogs/hedwig/:type" :paramsDesc="['分类, 见下表']"/>
diff --git a/lib/router.js b/lib/router.js
index b7aa6d7156..b890fd1786 100644
--- a/lib/router.js
+++ b/lib/router.js
@@ -468,6 +468,9 @@ router.get('/xueqiu/hots', require('./routes/xueqiu/hots'));
 // Greasy Fork
 router.get('/greasyfork/:language/:domain?', require('./routes/greasyfork/scripts'));
 
+// Gwern Bran­wen
+router.get('/gwern/:category', require('./routes/gwern/category'));
+
 // LinkedKeeper
 router.get('/linkedkeeper/:type/:id?', require('./routes/linkedkeeper/index'));
 
diff --git a/lib/routes/gwern/category.js b/lib/routes/gwern/category.js
new file mode 100644
index 0000000000..afb86063a3
--- /dev/null
+++ b/lib/routes/gwern/category.js
@@ -0,0 +1,58 @@
+const cheerio = require('cheerio');
+const got = require('@/utils/got');
+
+module.exports = async (ctx) => {
+    const category = ctx.params.category;
+    const name = category.charAt(0).toUpperCase() + category.slice(1);
+    const baseURL = `https://www.gwern.net`;
+    const url = baseURL + `/index`;
+    const res = await got.get(url);
+    const $ = cheerio.load(res.data);
+    const list = $(`section#${category} ul > li`).get();
+
+    const out = await Promise.all(
+        list.map(async (item) => {
+            const grasp = cheerio.load(item);
+            const link = baseURL + grasp('a').attr('href');
+            const cache = await ctx.cache.get(link);
+            if (cache) {
+                return Promise.resolve(JSON.parse(cache));
+            }
+            const res = await got.get(link);
+            const $ = cheerio.load(res.data);
+
+            const title = $('header > h1').text();
+            const time = $('span#page-creation').text();
+            // const modified = $('span#page-modified').text();
+
+            $('a').each((index, item) => {
+                item = $(item);
+                item.removeAttr('data-popup-title');
+                item.removeAttr('data-popup-title-html');
+                item.removeAttr('data-popup-author');
+                item.removeAttr('data-popup-abstract');
+                item.removeAttr('data-popup-doi');
+                item.removeAttr('data-popup-date');
+            });
+            $;
+
+            const contents = $('div#markdownBody').html().replace('&shy;', '');
+
+            const single = {
+                title: title,
+                author: 'Gwern Bran­wen',
+                description: contents,
+                link: link,
+                guid: link,
+                pubDate: new Date(time).toUTCString(),
+            };
+            ctx.cache.set(link, JSON.stringify(single));
+            return Promise.resolve(single);
+        })
+    );
+    ctx.state.data = {
+        title: `Gwern - ${name}`,
+        link: url,
+        item: out,
+    };
+};