不要问我为啥要写这样的代码，嘿嘿嘿~~~~~
1.文章自动上传到steemit上
本来想分析网站的api进行上传的，没想到steemit做了众多限制，感觉有些麻烦所以采用直接控制浏览器的方法上传文章。 node.js使用selenium-webdriver模块可以轻松控制浏览器。
/**
* 登录
*/
async login() {
let url = `https://steemit.com/login.html`;
let _this = this;
let returnJson = {
result: false,
value: ""
};
return new Promise(async (resolve, reject) => {
await _this.web.get(url);
await _this.web.findElements(By.css("input")).then(async res => {
await res[1].sendKeys(_this.__username);//设置登录用户名
await res[2].sendKeys(_this.__password);//设置登录密码
await _this.web
.findElement(By.css(".login-modal-buttons>button"))
.click();
await _this.web.wait(until.urlIs("https://steemit.com/welcome"), 4000);
await _this.web
.findElement(By.css(".show-for-medium.submit-story"))
.click();
returnJson.result = true;
returnJson.value = "login success";
console.log("login success");
resolve(returnJson);
return returnJson;
});
});
}
/**
* 提交文章
* @param {*} title 文章标题
* @param {*} post 文章体
* @param {*} tags 文章标签
*/
async submit(title, post, tags) {
let url = `https://steemit.com/submit.html`;
let _this = this;
let returnJson = {
result: false,
value: ""
};
return new Promise(async (resolve, reject) => {
await _this.web.get(url);
//设定延时,确保进入提交页面
await _this.web.wait(
until.urlIs("https://steemit.com/submit.html"),
4000
);
//标签定位，打开网页分析一下就oK了
await _this.web.findElements(By.css("input")).then(async res => {
await res[1].sendKeys(title);
await res[3].sendKeys(tags);
await _this.web.wait(until.elementLocated(By.css("textarea")), 3000);
await _this.web.findElement(By.css("textarea")).sendKeys(post);
await _this.web.findElement(By.css(".button")).click();
returnJson.result = true;
returnJson.value = "submit success";
resolve(returnJson);
return returnJson;
});
});
}
标签定位大家只用打开浏览器看一下就知道了 先调用登录接口，登录成功后调用上传文章接口
2.文章来源-从medium批量下载文章到redis队列
根据一位道友提供的api接口可以根据作者名获取到该作者的文章列表。
/**
* 获取Medium文章列表
* @param {*} user
* @param {*} limit
*/
async getPostByUsername(username, limit) {
let returnJson = {
result: false,
value: ""
};
let _this = this;
return new Promise((resolve, reject) => {
if (
typeof limit != "number" ||
limit <= 0 ||
limit > 100 ||
typeof username != "string" ||
username === ""
) {
returnJson.result = false;
returnJson.value = "参数不正确";
resolve(returnJson);
return returnJson;
} else {
let JSONDate = {
query: `query PostQuery($username: String!, $limit: Int!){
posts(username: $username, limit: $limit) {
title
firstPublishedAt
url
content {
subtitle
}
}
user(username: $username) {
username
name
bio
}
}`,
variables: `{
"username": "${username}",
"limit": ${limit}
}`,
operationName: "PostQuery"
};
superagent
.post("https://micro-medium-api.now.sh/graphql")
.send(JSONDate) // sends a JSON post body
.set("accept", "json")
.end((err, res) => {
if (err) {
console.log("获取文章列表出错： ", err);
returnJson.result = false;
returnJson.value = "获取文章列表出错： " + err;
resolve(returnJson);
return returnJson;
} else {
returnJson.result = true;
returnJson.value = res.text;
resolve(returnJson);
return returnJson;
}
});
}
});
}
又是一个神器h2m,(根据url将html转化成md)
npm install h2m -g
h2m https://baidu.com
于是我有封装了一个接口
/**
* 根据url下载Medium文章到本地
* @param {*} url
* @param {*} savePath
* @param {*} saveName
*/
async downloadPostByUrl(url, saveName,author,category) {
let returnJson = {
result: false,
value: ""
};
let _this = this;
url = encodeURI(url);
return new Promise((resolve, reject) => {
if (url === "" || saveName === "" ) {
returnJson.result = false;
returnJson.value = "参数不正确";
resolve(returnJson);
return returnJson;
} else {
let command = `h2m ${url}`;
console.log(command);
exec(command, {timeout: 1000*60*3,maxBuffer: 20*1024*1024},(error, stdout, stderr) => {
console.log(`stderr: ${stderr}`);
if (error) {
console.log(`exec error: ${error}`);
returnJson.result = false;
returnJson.value = `exec error: ${error}`;
resolve(returnJson);
} else {
returnJson.result = true;
returnJson.value = `根据url下载Medium文章到本地成功`;
console.log('根据url下载Medium文章到本地成功');
resolve(returnJson);
let saveStr = stdout.slice(stdout.indexOf('\n\n---\n'),stdout.indexOf('One clap, two clap, three clap, forty?'))
let title = saveName;
let postJson = {
title: title,
author: author,
category: category,
content: saveStr
}
//存储到redis队列中
_this.queueJson[category].publish(postJson)
}
return returnJson;
});
}
});
}
小坑：exec使用时stdout, stderr默认大小是200K,要把maxBuffer设置大一点才行。
3.如何保证上传的文章不重复呢？
真心感谢无所不能的npm 使用redis-message-queue可以轻松创建出值唯一的redis队列。
this.client = new rmq.UniqueQueue(this.name, port, host);
4.运行
代码位置：https://gitee.com/null_639_7345/steemit 1.git代码到本地 2.下载firefox驱动文件到本地，安装firefox浏览器 3.npm i h2m -g 4.修改配置文件config/default.js
module.exports = {
cwd : 'F:\\test1\\steemit\\',//项目根目录
redisHost: '192.168.10.6',//redis服务器ip
redisPost: 6379,
promulgatorName: '***',//steemit用户名
promulgatorPassword: "***",//steemit密码
categories:[
{
name: 'popular',//steemit上传文章的分类（tag）
//Medium的作者列表
origin: ['joshrose','JessicaLexicus','ThunderPuff','usemuzli','black_metallic']
}
]
};
5.node index.js
最后可以告诉大家一个激动的好消息。
