问题
今天项目需求要求采集几个网站的信息,包括一些区块链统计图表之类的信息。
笔者使用的是node.js+axios库发送get请求来获取在图片,下载到本地。测试代码如下:
import fs from 'fs';
import path from 'path';
import http from 'http';
import https from 'https';
const __dirname = path.resolve();
let filePath = path.join(__dirname,'/imgtmp/');
async function downloadfile(url,filename,callback){
try {
let ext = path.extname(url);
console.log('下载的文件名:',filename)
let mod = null;//http、https 别名
if(url.indexOf('https://')!==-1){
mod = https;
}else{
mod = http;
}
const req = mod.get(url, {
headers:{
"Content-Type": "application/x-www-form-urlencoded"
}
},(res)=>{
let writePath = '';
writePath = filePath + '/' + filename;
const file = fs.createWriteStream(writePath)
res.pipe (file)
file.on ("error", (error) => {
console.log (`There was an error writing the file. Details: `,error)
return false;
})
file.on ("close", () => {
callback (filename)
})
file.on ('finish', () => {
file.close ()
console.log ("Completely downloaded.")
})
})
req.on ("error", (error) => {
console.log (`Error downloading file. Details: $ {error}`)
})
} catch (error) {
console.log('图片下载失败!',error);
}
}
let url = 'https://xx.xxxx.com/d/file/zxgg/a2cffb8166f07c0232eca49f8c9cc242.jpg';//图片url
let filename = path.basename(url);
await downloadfile(url,filename,()=>{
console.log(filename,"文件已下载成功");
})
运行代码,图示文件下载成功!
然而当笔者打开图片一看,就傻眼了~图片显示损坏,再看大小,只有304字节~
目测应该是图片保存了一些错误信息,于是用editplus以文本形式打开该图片,果然看到了错误信息~
解决方法
百度了一下,确定是图片nginx服务器Referer防盗链设置,于是继续百度,找到了问题的关键~
谷歌浏览器打开网址,在控制台上看到了这段Referer信息:
对方的网站在Referer设置的就是他的网址,于是改进代码,在headers中加入Referer参数"referer":'https://www.xxxx.com/'
:
import fs from 'fs';
import path from 'path';
import http from 'http';
import https from 'https';
const __dirname = path.resolve();
let filePath = path.join(__dirname,'/imgtmp/');
async function downloadfile(url,filename,callback){
try {
let ext = path.extname(url);
console.log('下载的文件名:',filename)
let mod = null;//http、https 别名
if(url.indexOf('https://')!==-1){
mod = https;
}else{
mod = http;
}
const req = mod.get(url, {
headers:{
"Content-Type": "application/x-www-form-urlencoded",
"referer":'https://www.xxxx.com/'
}
},(res)=>{
let writePath = '';
writePath = filePath + '/' + filename;
const file = fs.createWriteStream(writePath)
res.pipe (file)
file.on ("error", (error) => {
console.log (`There was an error writing the file. Details: `,error)
return false;
})
file.on ("close", () => {
callback (filename)
})
file.on ('finish', () => {
file.close ()
console.log ("Completely downloaded.")
})
})
req.on ("error", (error) => {
console.log (`Error downloading file. Details: $ {error}`)
})
} catch (error) {
console.log('图片下载失败!',error);
}
}
let url = 'https://xx.xxxx.com/d/file/zxgg/a2cffb8166f07c0232eca49f8c9cc242.jpg';//图片url
let filename = path.basename(url);
await downloadfile(url,filename,()=>{
console.log(filename,"文件已下载成功");
})
再次运行代码,图片文件下载成功,打开显示一切正常!
后记
笔者又测试了另一种实现方法,即使用playwright调用浏览器打开页面,再使用await page.locator('selector路径').screenshot({ path: 'image图片保存路径'});
将图片网页截图保存下载。
对比了一番,发现使用playwright截图的方法需要在遍历图片元素的时候根据当前元素逆向获取parentNode节点以及遍历childNodes节点,算法相对比较复杂!而且screenshot函数截图的效果也会比原图略显模糊,因此推荐使用axios传递Referer参数的方法获取原图。
PS:方法二的调试过程中写了一段逆向遍历selector的函数,提供给大家参考,如有不足之处,欢迎指正~
function getSelectorPath(element) {
if (!!element.id !== false) {
return '#' + element.id;
}
if (element === document.body && !!element) {
return element.tagName.toLowerCase();
}
let ix = 0;
const siblings = element.parentNode?.childNodes;
for (let i = 0; i < siblings?.length; i++) {
const sibling = siblings[i];
if (sibling.innerHTML === element.innerHTML && !!element.parentNode) {
return `${getSelectorPath(element.parentNode)} > ${element.tagName.toLowerCase()}:nth-child(${ix + 1})`;
}
if (sibling.nodeType === 1) {
ix++;
}
}
}