如何在不使用第三方库的情况下使用Node.js下载文件?

我不需要什么特别的东西。我只想从给定的URL下载文件,然后将其保存到给定的目录。


当前回答

也许node.js已经改变了,但似乎其他解决方案(使用node v8.1.2)有一些问题:

You don't need to call file.close() in the finish event. Per default the fs.createWriteStream is set to autoClose: https://nodejs.org/api/fs.html#fs_fs_createwritestream_path_options file.close() should be called on error. Maybe this is not needed when the file is deleted (unlink()), but normally it is: https://nodejs.org/api/stream.html#stream_readable_pipe_destination_options Temp file is not deleted on statusCode !== 200 fs.unlink() without a callback is deprecated (outputs warning) If dest file exists; it is overridden

下面是一个修改后的解决方案(使用ES6和promises),它可以处理这些问题。

const http = require("http");
const fs = require("fs");

function download(url, dest) {
    return new Promise((resolve, reject) => {
        const file = fs.createWriteStream(dest, { flags: "wx" });

        const request = http.get(url, response => {
            if (response.statusCode === 200) {
                response.pipe(file);
            } else {
                file.close();
                fs.unlink(dest, () => {}); // Delete temp file
                reject(`Server responded with ${response.statusCode}: ${response.statusMessage}`);
            }
        });

        request.on("error", err => {
            file.close();
            fs.unlink(dest, () => {}); // Delete temp file
            reject(err.message);
        });

        file.on("finish", () => {
            resolve();
        });

        file.on("error", err => {
            file.close();

            if (err.code === "EEXIST") {
                reject("File already exists");
            } else {
                fs.unlink(dest, () => {}); // Delete temp file
                reject(err.message);
            }
        });
    });
}

其他回答

✅So if you use pipeline, it would close all other streams and make sure that there are no memory leaks. Working example: const http = require('http'); const { pipeline } = require('stream'); const fs = require('fs'); const file = fs.createWriteStream('./file.jpg'); http.get('http://via.placeholder.com/150/92c952', response => { pipeline( response, file, err => { if (err) console.error('Pipeline failed.', err); else console.log('Pipeline succeeded.'); } ); });

从我的回答“在流上。pipe和。pipeline之间有什么区别”。

超时解决方案,防止内存泄漏:

下面的代码是基于Brandon Tilley的回答:

var http = require('http'),
    fs = require('fs');

var request = http.get("http://example12345.com/yourfile.html", function(response) {
    if (response.statusCode === 200) {
        var file = fs.createWriteStream("copy.html");
        response.pipe(file);
    }
    // Add timeout.
    request.setTimeout(12000, function () {
        request.abort();
    });
});

当您得到一个错误时,不要创建文件,并倾向于使用超时在X秒后关闭您的请求。

Gfxmonk的答案在回调和file.close()完成之间有一个非常紧张的数据竞赛。File.close()实际上接受一个回调函数,该函数在close完成时被调用。否则,立即使用文件可能会失败(非常罕见!)。

一个完整的解决方案是:

var http = require('http');
var fs = require('fs');

var download = function(url, dest, cb) {
  var file = fs.createWriteStream(dest);
  var request = http.get(url, function(response) {
    response.pipe(file);
    file.on('finish', function() {
      file.close(cb);  // close() is async, call cb after close completes.
    });
  });
}

如果不等待finish事件,幼稚的脚本可能最终得到一个不完整的文件。如果不通过close调度cb回调,您可能会在访问文件和文件实际准备就绪之间出现竞争。

Vince Yuan的代码很棒,但似乎有问题。

function download(url, dest, callback) {
    var file = fs.createWriteStream(dest);
    var request = http.get(url, function (response) {
        response.pipe(file);
        file.on('finish', function () {
            file.close(callback); // close() is async, call callback after close completes.
        });
        file.on('error', function (err) {
            fs.unlink(dest); // Delete the file async. (But we don't check the result)
            if (callback)
                callback(err.message);
        });
    });
}

根据上面的其他答案和一些微妙的问题,下面是我的尝试。

Check the file does not exist before hitting the network by using fs.access. Only create the fs.createWriteStream if you get a 200 OK status code. This reduces the amount of fs.unlink commands required to tidy up temporary file handles. Even on a 200 OK we can still possibly reject due to an EEXIST file already exists (imagine another process created the file whilst we were doing network calls). Recursively call download if you get a 301 Moved Permanently or 302 Found (Moved Temporarily) redirect following the link location provided in the header. The issue with some of the other answers recursively calling download was that they called resolve(download) instead of download(...).then(() => resolve()) so the Promise would return before the download actually finished. This way the nested chain of promises resolve in the correct order. It might seem cool to clean up the temp file asynchronously, but I chose to reject only after that completed too so I know that everything start to finish is done when this promise resolves or rejects.

const https = require('https');
const fs = require('fs');

/**
 * Download a resource from `url` to `dest`.
 * @param {string} url - Valid URL to attempt download of resource
 * @param {string} dest - Valid path to save the file.
 * @returns {Promise<void>} - Returns asynchronously when successfully completed download
 */
function download(url, dest) {
  return new Promise((resolve, reject) => {
    // Check file does not exist yet before hitting network
    fs.access(dest, fs.constants.F_OK, (err) => {

        if (err === null) reject('File already exists');

        const request = https.get(url, response => {
            if (response.statusCode === 200) {
       
              const file = fs.createWriteStream(dest, { flags: 'wx' });
              file.on('finish', () => resolve());
              file.on('error', err => {
                file.close();
                if (err.code === 'EEXIST') reject('File already exists');
                else fs.unlink(dest, () => reject(err.message)); // Delete temp file
              });
              response.pipe(file);
            } else if (response.statusCode === 302 || response.statusCode === 301) {
              //Recursively follow redirects, only a 200 will resolve.
              download(response.headers.location, dest).then(() => resolve());
            } else {
              reject(`Server responded with ${response.statusCode}: ${response.statusMessage}`);
            }
          });
      
          request.on('error', err => {
            reject(err.message);
          });
    });
  });
}