feat(route): 公众号(Telegram 频道来源)增加筛选 (#7648)

This commit is contained in:
Rongrong
2022-01-22 19:03:18 +00:00
committed by GitHub
parent 75b9a49aab
commit ce9743012a
3 changed files with 50 additions and 13 deletions
+1 -1
View File
@@ -2980,7 +2980,7 @@ column 为 third 时可选的 category:
### 公众号(Telegram 频道来源)
<Route author="LogicJake" example="/wechat/tgchannel/lifeweek" path="/wechat/tgchannel/:id" :paramsDesc="['公众号绑定频道 id']">
<Route author="LogicJake" example="/wechat/tgchannel/lifeweek" path="/wechat/tgchannel/:id/:mpName?" :paramsDesc="['公众号绑定频道 id', '欲筛选的公众号全名(精确匹配),在频道订阅了多个公众号时可选用']">
::: warning 注意
+1 -1
View File
@@ -512,7 +512,7 @@ router.get('/wechat/csm/:id', lazyloadRouteHandler('./routes/tencent/wechat/csm'
router.get('/wechat/ce/:id', lazyloadRouteHandler('./routes/tencent/wechat/ce'));
router.get('/wechat/announce', lazyloadRouteHandler('./routes/tencent/wechat/announce'));
router.get('/wechat/miniprogram/plugins', lazyloadRouteHandler('./routes/tencent/wechat/miniprogram/plugins'));
router.get('/wechat/tgchannel/:id', lazyloadRouteHandler('./routes/tencent/wechat/tgchannel'));
router.get('/wechat/tgchannel/:id/:mpName?', lazyloadRouteHandler('./routes/tencent/wechat/tgchannel'));
router.get('/wechat/uread/:userid', lazyloadRouteHandler('./routes/tencent/wechat/uread'));
router.get('/wechat/ershicimi/:id', lazyloadRouteHandler('./routes/tencent/wechat/ershcimi'));
router.get('/wechat/wjdn/:id', lazyloadRouteHandler('./routes/tencent/wechat/wjdn'));
+48 -11
View File
@@ -3,26 +3,62 @@ const cheerio = require('cheerio');
module.exports = async (ctx) => {
const id = ctx.params.id;
const mpName = ctx.params.mpName ?? '';
const { data } = await got.get(`https://t.me/s/${id}`);
const $ = cheerio.load(data);
const list = $('.tgme_widget_message_wrap').slice(-10);
const list = $('.tgme_widget_message_wrap').slice(-20);
const out = await Promise.all(
list
.map(async (index, item) => {
item = $(item);
let author;
let title = item.find('.tgme_widget_message_text > a:nth-child(5)').text() || item.find('.tgme_widget_message_text > a:nth-child(2)').text();
// [ div.tgme_widget_message_text 格式简略说明 ]
// 若频道只订阅一个公众号:
// 第 1 个元素: <a href="${用于 link priview 的预览图 url}"><i><b>${emoji(链接)}</b></i></a>
// 第 2 个元素: <a href="${文章 url}">${文章标题}</a>
// (余下是文章简介,一般是裸文本,这里用不到)
//
// 若频道订阅多于一个公众号:
// 第 1 个元素: <i><b>${emoji(标注消息来源于什么 slave,这里是表示微信的对话气泡)}</b></i>
// 第 2 个元素: <i><b>${emoji(标注对话类型,这里是表示私聊的半身人像)</b></i>
// 裸文本: (半角空格)${公众号名}(半角冒号)
// 第 3 个元素: <br />
// 第 4 个元素: <a href="${用于 link priview 的预览图 url}"><i><b>${emoji(链接)}</b></i></a>
// 第 5 个元素: <a href="${文章 url}">${文章标题}</a>
// (余下是文章简介,一般是裸文本,这里用不到)
const all_text = item.find('.tgme_widget_message_text').text();
if (all_text.indexOf(':') !== -1) {
author = all_text.split(':')[0].split(' ')[1];
title = author + ': ' + title;
const title_elem = item.find('.tgme_widget_message_text > a:nth-of-type(2)'); // 第二个 a 元素会是文章链接
if (title_elem.length === 0) {
// 获取不到第二个 a 元素,这可能是公众号发的服务消息,丢弃它
return;
}
let author;
let title = title_elem.text();
const link = title_elem.attr('href');
const br_node = item.find('.tgme_widget_message_text > br:nth-of-type(1)').get(0); // 获取第一个换行
const author_node = br_node && br_node.prev; // br_node 不为 undefined 时获取它的前一个节点
if (author_node && author_node.type === 'text') {
// 只有这个节点是一个裸文本时它才可能是公众号名
const spaceIndex = author_node.data.indexOf(' ');
const colonIndex = author_node.data.indexOf(':');
if (spaceIndex !== -1 && colonIndex !== -1) {
// 找到了公众号名,说明这个频道订阅了多个公众号
author = author_node.data.slice(spaceIndex + 1, colonIndex); // 提取作者
if (mpName && author !== mpName) {
// 指定了要筛选的公众号名,且该文章不是该公众号发的
return; // 丢弃
} else if (!mpName) {
// 没有指定要筛选的公众号名
title = author + ': ' + title; // 给标题里加上获取到的作者
}
}
}
const link = item.find('.tgme_widget_message_text > a:nth-child(5)').attr('href') || item.find('.tgme_widget_message_text > a:nth-child(2)').attr('href');
const pubDate = new Date(item.find('.tgme_widget_message_date time').attr('datetime')).toUTCString();
const single = {
@@ -51,15 +87,16 @@ module.exports = async (ctx) => {
}
}
return Promise.resolve(single);
return single;
})
.get()
);
out.reverse();
ctx.state.data = {
title: $('.tgme_channel_info_header_title').text(),
title: mpName ?? $('.tgme_channel_info_header_title').text(),
link: `https://t.me/s/${id}`,
item: out,
item: out.filter((item) => item),
allowEmpty: !!mpName,
};
};