javascript - 上传 Node 请求-响应响应到 MongoDB

标签 javascript node.js mongodb response cheerio

我正在 Node.js 中使用 Cheerio。我有一个抓取文章列表的抓取器,抓取所有文章的 URLS,然后转到每篇文章并抓取标题和 url。一切正常,除了当我尝试将结果更新到我的 Mongodb 时,我得到了 undefined。

我假设它在定义值之前尝试更新插入...但即使使用请求-响应我也无法让它工作。任何帮助将不胜感激!由于代码不是太长,我只是粘贴整个代码,以便更容易看到我正在尝试做什么。同样,主要问题是让 upsertArticle 实际更新插入变量。

const request = require('request');
const cheerio = require('cheerio');
const rp = require('request-promise');
const mongoose = require('mongoose');
const Article = require('./models/article');

var urls = [];
//get the list of articles to scrape
rp('https://www.somesite.com/', function(error, response, html) {
    if (!error && response.statusCode == 200) {
        var $ = cheerio.load(html);
        $('.c-entry-box--compact__title').each(function(i, element) {
            var a = $(this);
            urls.push(a.children().attr('href'));
        });     }  })
    //scrape over each article individually
    .then(function(getStuff) {
        var arrayLength = urls.length;
        //get the list of articles to scrape and upsert each one
        for (var i = 0; i < arrayLength; i++) {
            const result = rp(urls[i], function(error, response, html) {
                if (!error && response.statusCode == 200) {
                    var $ = cheerio.load(html);
                    var parsedResults = [];
                    $('.l-main-content').each(function(n, element) {
                        var a = $(this);
                        var title = a.find('.c-page-title').text();
                        var url = response.request.uri.href;
                        //I also tried upserting the variables right here, that didn't work
                        return { title, url };
                    });
                } else {console.log(error);}
            }).then(function(upsertStuff) {
                    //also tried returning and upserting stuff here... but nothing gets upserted
                    upsertArticle({
                        title: result.title,
                        source: result.url,
                        dateCrawled: new Date()
                    });
                    console.log('Upserted ' + result.title);
                }).catch(function(err) {console.log(err);   });     }
    })  .catch(function(err) {console.log(err); });

function upsertArticle(userObj) {
    const DB_URL = 'mongodb://localhost/articles';
    if (mongoose.connection.readyState == 0) {
        mongoose.connect(DB_URL, {
            useMongoClient: true
        });
    }
    let conditions = {
        title: userObj.title
    };
    let options = {
        upsert: true,
        new: true,
        setDefaultsOnInsert: true
    };
    Article.findOneAndUpdate(conditions, userObj, options, (err, result) => {
        if (err) throw err;
    });
}

最佳答案

我对提供的代码做了一些修改。也就是说,我使用 promises 而不是回调来让您的逻辑保持一致并确保一切都在应有的时间运行。

对于 for 循环,我将 upsertArticle({...}) 移回到 each 函数内部,所以titleurl 是在运行时定义的。

最后,我使用 Bluebird 的 Promise.all(request-promise 已经依赖于 Bluebird)在所有链接都已更新时发出信号。此更改是可选的,但我认为在完成所有操作后获得反馈会很有用:

试一试:

const request = require('request');
const cheerio = require('cheerio');
const rp = require('request-promise');
const mongoose = require('mongoose');
const Article = require('./models/article');
const Promise = require("bluebird");

var urls = [];

rp({uri: 'https://www.somesite.com',  resolveWithFullResponse: true}).then(function(response) {

    if(response.statusCode != 200) throw "Response: " + response.statusCode;

    var html = response.body;

    var $ = cheerio.load(html);

    $('.c-entry-box--compact__title').each(function(i, element) {
        var a = $(this);
        urls.push(a.children().attr('href'));
    });

}).then(function(getStuff) {

    var arrayLength = urls.length;
    var promiseArray = [];

    for(var i = 0; i < arrayLength; i++) {

        const p = rp({uri: urls[i],  resolveWithFullResponse: true}).then(function(response) {

            if(response.statusCode != 200) throw "Response: " + response.statusCode;

            var html = response.body;

            var $ = cheerio.load(html);
            var parsedResults = [];

            $('.l-main-content').each(function(n, element) {

                var a = $(this);
                var title = a.find('.c-page-title').text();
                var url = response.request.uri.href;

                upsertArticle({
                    title: title,
                    source: url,
                    dateCrawled: new Date()
                });

                console.log('Upserted ' + title);
            });

        });

        promiseArray.push(p);
    }

    return Promise.all(promiseArray);

}).then(function() {
    console.log("Done upserting!");
})
.catch(function(err) {
    console.log(err); 
});

function upsertArticle(userObj) {
    const DB_URL = 'mongodb://localhost/articles';
    if (mongoose.connection.readyState == 0) {
        mongoose.connect(DB_URL, {
            useMongoClient: true
        });
    }
    let conditions = {
        title: userObj.title
    };
    let options = {
        upsert: true,
        new: true,
        setDefaultsOnInsert: true
    };
    Article.findOneAndUpdate(conditions, userObj, options, (err, result) => {
        if (err) throw err;
    });
}

我无法在不知道 https://www.somesite.com 的真实值的情况下测试代码,所以如果代码给您带来任何新错误,请告诉我。

关于javascript - 上传 Node 请求-响应响应到 MongoDB,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/47536596/

相关文章:

javascript - 如果所有浏览器都使用同源策略,CSRF 攻击如何发挥作用?

javascript - 为视频添加静音按钮

javascript - 关于 this/@ 在 Javascript/Coffeescript 中的一个难题

javascript - 子域名无法使用 express-vhost

mongodb - 在 Go 中将任何编码转换为 UTF 8

javascript - 图像上关闭按钮的 anchor 标记在 safari 中不起作用

javascript - 如何克隆 -> 反向 -> 和 prependTo?

node.js - gmail API 快速入门在 google.gmail ('v1' 处失败)

java mongodb sort()和limit()函数

java - MongoDB Java 驱动程序 : Multiple Date query