我最近潜入nodejs(并使用nightmare.js)来解析网站,并且遇到回调函数问题并显示返回的结果。我试图在另一个函数中调用一个单独的函数,但似乎无法返回任何结果。他们都返回undefined。非常感谢任何帮助。
function getDetails(loadURL, callback){
nightmare.goto(loadURL)
.wait(2000)
.evaluate(function(){
var gigs = [];
$('.hidden-xs .used-vehicle').each(function(){
item = {}
item["year"] = $(this).attr('data-year')
item["make"] = $(this).attr('data-make')
item["model"] = $(this).attr('data-model')
item["body"] = $(this).attr('data-body')
item["color"] = $(this).attr('data-ext-color')
item["trim"] = $(this).attr('data-trim')
item["mileage"] = $(this).attr('data-mileage')
item["transmission"] = $(this).attr('data-transmission')
item["vin"] = $(this).find(".vehicle-overview").attr('id')
item["title"] = $(this).find(".vehicle-overview h2 a").text()
item["link"] = $(this).find(".vehicle-overview h2 a").attr('href')
item["price"] = $(this).find(".vehicle-content .price").text()
gigs.push(item)
})
return gigs
})
.end()
.then(function(result){
var returnString = '';
for(gig in result){
returnString = returnString + result[gig].title + " " + result[gig].link + " " + result[gig].year + " " + result[gig].make + " " + result[gig].model + " " + result[gig].body + " " + result[gig].color + " " + result[gig].trim + " " + result[gig].transmission + " " + result[gig].vin + " " + result[gig].price + "\n"
}
callback(returnString)
})
}
// We will need to get the total amount of pages that we need to parse
function getInventory(sURL, callback){
nightmare.goto(sURL)
.wait(2000)
.evaluate(function(){
totals = [];
items = {}
totalCars = $('.total-found .count').text()
carsOnPage = $('.hidden-xs .used-vehicle').size()
items['carTotal'] = totalCars
items['onPage'] = carsOnPage
var pageCalc = (totalCars / carsOnPage)
items['tPages'] = Math.ceil(pageCalc)
totals.push(items)
return totals
})
.end()
.then(function(result){
var totalCars = '';
var totalPages = '';
for (item in result){
totalPages = result[item].tPages
totalCars = result[item].carTotal
}
counter = 0;
newURL = '';
returnDetails = '';
for (i =0; i < totalPages; i++){
if (i == 0){
newURL = sURL;
} else {
counter = i + 1;
newURL = sURL + "#action=im_ajax_call&perform=get_results&_post_id=5&page=" + counter + "&show_all_filters=false";
}
//console.log(newURL)
getINV = getDetails(newURL, function(returnString){
callback(returnString)
})
returnDetails = returnDetails + getINV
}
callback(returnDetails)
})
}
getInventory(startURL, function(result){
console.log(result)
})
答案 0 :(得分:1)
我不会打扰你告诉你,你不应该将回调与这样的承诺混在一起。但是现在让我们看看这个问题。
<案例1你也检查错误怎么样?也许你的脚本会抛出错误。我可以看到你在.then
上调用回调,但在.catch
上没有。也许那时永远不会得到任何数据。
让我们检查一下你的功能。您每次都在致电.end
。您是否每次都在创建新的Nightmare实例?
在getInventory
功能上,您不应拨打.end
。在getDetails
功能上,您不应该致电.end
。它结束了噩梦般的实例,你正在丢失你的数据。
完成所有功能后,请致电nightmare.end()
。为了做到这一点,你需要了解更多关于Promises check case 3的信息。
了解承诺如何运作。在下面的行中,您永远不会等待该功能完成。
getINV = getDetails(newURL, function(returnString){
callback(returnString)
})
你应该等待承诺完成。另外,请确保梦魇不会同时浏览两个链接。
所以继续学习Promises和async等待东西。
我如何解决您的代码?
我会使用Promise.all,.map和其他一些新东西。以下是为您完成的一些示例代码,不要直接复制粘贴或运行代码,尝试理解为什么它与您的代码不同以及它的结果可能是什么。
const pLimit = require("promise-limit")(2);
function getDetails(loadURL) {
return nightmare
.goto(loadURL)
.wait(2000)
.evaluate(() => {
const gigs = [];
$(".hidden-xs .used-vehicle").each(function() {
item = {};
item["year"] = $(this).attr("data-year");
item["make"] = $(this).attr("data-make");
item["model"] = $(this).attr("data-model");
item["body"] = $(this).attr("data-body");
item["color"] = $(this).attr("data-ext-color");
item["trim"] = $(this).attr("data-trim");
item["mileage"] = $(this).attr("data-mileage");
item["transmission"] = $(this).attr("data-transmission");
item["vin"] = $(this)
.find(".vehicle-overview")
.attr("id");
item["title"] = $(this)
.find(".vehicle-overview h2 a")
.text();
item["link"] = $(this)
.find(".vehicle-overview h2 a")
.attr("href");
item["price"] = $(this)
.find(".vehicle-content .price")
.text();
gigs.push(item);
});
return gigs;
})
.then(result => {
let returnString = "";
for (gig in result) {
returnString =
`${returnString +
result[gig].title} ${result[gig].link} ${result[gig].year} ${result[gig].make} ${result[gig].model} ${result[gig].body} ${result[gig].color} ${result[gig].trim} ${result[gig].transmission} ${result[gig].vin} ${result[gig].price}\n`;
}
return returnString;
})
.catch(error => {
throw new Error(error);
});
}
// We will need to get the total amount of pages that we need to parse
function getInventory(sURL) {
return nightmare
.goto(sURL)
.wait(2000)
.evaluate(() => {
totals = [];
items = {};
totalCars = $(".total-found .count").text();
carsOnPage = $(".hidden-xs .used-vehicle").size();
items["carTotal"] = totalCars;
items["onPage"] = carsOnPage;
const pageCalc = totalCars / carsOnPage;
items["tPages"] = Math.ceil(pageCalc);
totals.push(items);
return totals;
})
.then(result => {
let totalCars = "";
let totalPages = "";
for (item in result) {
totalPages = result[item].tPages;
totalCars = result[item].carTotal;
}
counter = 0;
newURL = "";
urls = [];
returnDetails = [];
for (i = 0; i < totalPages; i++) {
if (i == 0) {
newURL = sURL;
} else {
counter = i + 1;
newURL =
`${sURL}#action=im_ajax_call&perform=get_results&_post_id=5&page=${counter}&show_all_filters=false`;
}
// push to the url array
// use .map for cleaner code
urls.push(newURL);
}
// return a new promise with concurrency limit
return Promise.all(
urls.map(url => {
return limit(() => getDetails(newURL));
})
);
})
.catch(error => {
throw new Error(error);
});
}
getInventory(startURL)
.then(result => {
console.log(result);
})
.catch(error => {
console.err(error);
});
资源:
答案 1 :(得分:0)
尝试使用回调而不是返回
function getDetails(loadURL, callback){
nightmare.goto(loadURL)
.wait(2000)
.evaluate(function(callback){
var gigs = [];
$('.hidden-xs .used-vehicle').each(function(){
item = {}
item["year"] = $(this).attr('data-year')
item["make"] = $(this).attr('data-make')
item["model"] = $(this).attr('data-model')
item["body"] = $(this).attr('data-body')
item["color"] = $(this).attr('data-ext-color')
item["trim"] = $(this).attr('data-trim')
item["mileage"] = $(this).attr('data-mileage')
item["transmission"] = $(this).attr('data-transmission')
item["vin"] = $(this).find(".vehicle-overview").attr('id')
item["title"] = $(this).find(".vehicle-overview h2 a").text()
item["link"] = $(this).find(".vehicle-overview h2 a").attr('href')
item["price"] = $(this).find(".vehicle-content .price").text()
gigs.push(item)
})
callback(gigs)
})
.end()
.then(function(result){
var returnString = '';
for(gig in result){
returnString = returnString + result[gig].title + " " + result[gig].link + " " + result[gig].year + " " + result[gig].make + " " + result[gig].model + " " + result[gig].body + " " + result[gig].color + " " + result[gig].trim + " " + result[gig].transmission + " " + result[gig].vin + " " + result[gig].price + "\n"
}
callback(returnString)
})
}
// We will need to get the total amount of pages that we need to parse
function getInventory(sURL, callback){
nightmare.goto(sURL)
.wait(2000)
.evaluate(function(){
totals = [];
items = {}
totalCars = $('.total-found .count').text()
carsOnPage = $('.hidden-xs .used-vehicle').size()
items['carTotal'] = totalCars
items['onPage'] = carsOnPage
var pageCalc = (totalCars / carsOnPage)
items['tPages'] = Math.ceil(pageCalc)
totals.push(items)
return totals
})
.end()
.then(function(result){
var totalCars = '';
var totalPages = '';
for (item in result){
totalPages = result[item].tPages
totalCars = result[item].carTotal
}
counter = 0;
newURL = '';
returnDetails = '';
for (i =0; i < totalPages; i++){
if (i == 0){
newURL = sURL;
} else {
counter = i + 1;
newURL = sURL + "#action=im_ajax_call&perform=get_results&_post_id=5&page=" + counter + "&show_all_filters=false";
}
//console.log(newURL)
getINV = getDetails(newURL, function(returnString){
callback(returnString)
})
returnDetails = returnDetails + getINV
}
callback(returnDetails)
})
}
getInventory(startURL, function(result){
console.log(result)
})
答案 2 :(得分:0)
也许以下内容可行,更改循环以减少和映射,取出jQuery并进行一些小的更改。
最重要的是:
以下是代码:
const getDetails = loadURL =>
nightmare.goto(loadURL)//return promise here
.wait(2000)
.evaluate(
()=>
Array.from(document.querySelectorAll('.hidden-xs .used-vehicle'))
.reduce(
(all,item)=>
all.concat(
[
element.getAttribute('data-year'),
element.getAttribute('data-make'),
element.getAttribute('data-model'),
element.getAttribute('data-body'),
element.getAttribute('data-ext-color'),
element.getAttribute('data-trim'),
element.getAttribute('data-mileage'),
element.getAttribute('data-transmission'),
element.querySelector(".vehicle-overview").getAttribute('id'),
element.querySelector(".vehicle-overview h2 a").innerText,
element.querySelector(".vehicle-overview h2 a").getAttribute('href'),
element.querySelector(".vehicle-content .price").innerText
].join(" ")
),
[]//the all array
)
);
// We will need to get the total amount of pages that we need to parse
const getInventory = sURL =>
nightmare.goto(sURL)
.wait(2000)
.evaluate(
()=> {
//there is only one item here, not sure why you push it into totals
// and call it items
const item = {}
//getAttribute returns a string, parse it to number
totalCars = parseInt(document.querySelector('.total-found .count').innerText,10);
carsOnPage = document.querySelectorAll('.hidden-xs .used-vehicle').length;
item['carTotal'] = totalCars
item['onPage'] = carsOnPage
var pageCalc = (totalCars / carsOnPage)
item['tPages'] = Math.ceil(pageCalc)
return item;
}
)
.then(
totalItem =>{
var totalCars = '';
var totalPages = '';
totalPages = totalItem.tPages
totalCars = totalItem.carTotal
newURL = '';
returnDetails = '';
return Array.from(new Array(totalPages),(_,index)=>index+1)
.reduce(
(p,counter)=>
p.then(
results=>{
if (counter === 1) {
newURL = sURL;
} else {
newURL = sURL + "#action=im_ajax_call&perform=get_results&_post_id=5&page=" + counter + "&show_all_filters=false";
}
return getDetails(newURL)
.then(
result=>results.concat(result)
);
}
),
Promise.resolve([])
);
}
);
getInventory(startURL)
.then(
result=>
console.log(result)
).catch(
err=>
console.warn("Something went wrong:",err)
);
答案 3 :(得分:0)
学习具有异步功能的承诺和回调是一项艰巨的任务。我能够通过以下方式实现此目的。谢谢大家的帮助和指导。每个答案都把我送到了另一个兔子洞,最终找到了解决办法。
function getInventory(sURL){
nightmare.goto(sURL)
.wait(2000)
.evaluate(function(){
totals = [];
items = {}
totalCars = $('.total-found .count').text()
carsOnPage = $('.hidden-xs .used-vehicle').size()
items['carTotal'] = totalCars
items['onPage'] = carsOnPage
var pageCalc = (totalCars / carsOnPage)
items['tPages'] = Math.ceil(pageCalc)
totals.push(items)
return totals
})
.then(result => {
var totalCars = '';
var totalPages = '';
for (item in result){
totalPages = result[item].tPages
totalCars = result[item].carTotal
}
counter = 0;
let links = [];
let returnLinks = '';
newURL = '';
for (i = 0; i < totalPages; i++){
if (i == 0){
newURL = sURL;
} else {
counter = i + 1;
newURL = sURL + "#action=im_ajax_call&perform=get_results&_post_id=5&page=" + counter + "&show_all_filters=false";
}
links.push(newURL);
}
return links;
})
.then(results => {
var arrayLinks = results;
arrayLinks.reduce(function(accumulator, url){
return accumulator.then(function(newResults){
return nightmare.goto(url)
.wait(5000)
.evaluate(() => {
const gigs = [];
$(".hidden-xs .used-vehicle").each(function() {
item = {};
item["year"] = $(this).attr("data-year");
item["make"] = $(this).attr("data-make");
item["model"] = $(this).attr("data-model");
item["body"] = $(this).attr("data-body");
item["color"] = $(this).attr("data-ext-color");
item["trim"] = $(this).attr("data-trim");
item["mileage"] = $(this).attr("data-mileage");
item["transmission"] = $(this).attr("data-transmission");
item["vin"] = $(this).find(".vehicle-overview").attr("id");
item["title"] = $(this).find(".vehicle-overview h2 a").text();
item["link"] = $(this).find(".vehicle-overview h2 a").attr("href");
item["price"] = $(this).find(".vehicle-content .price").text();
gigs.push(item);
});
return gigs;
})
.then(detail => {
for (gig in detail) {
try {
var carVin = detail[gig].vin;
var carTitle = detail[gig].title;
var carDescrip = detail[gig].year + " " + detail[gig].make + " " + detail[gig].model;
var carURL = detail[gig].link;
var carMake = detail[gig].make;
var carModel = detail[gig].model;
var carYear = detail[gig].year;
var carMileageFull = detail[gig].mileage;
var carMileage = carMileageFull.replace(',', '');
var carTransmission = detail[gig].transmission;
var carBody = detail[gig].body;
var carPriceFull = detail[gig].price;
var carPriceFull = carPriceFull.replace('$', '');
var carPriceFull = carPriceFull.replace('*', '');
var carPriceFull = carPriceFull.replace(',', '');
var carPrice = carPriceFull.trim();
var dealerAddress = "{addr1: '"+ addressFull.addr1 +"', city: '"+ addressFull.city +"', region: '"+ addressFull.region +"', postal_code: '"+ addressFull.postal_code +"', country: '"+ addressFull.country +"'}";
var dealerLat = latLongFull.latitude;
var dealerLong = latLongFull.longitude;
var carColor = detail[gig].color;
arrSetup = [carVin, carTitle, carDescrip, carURL, carMake, carModel, carYear, carMileage, 'MI', '', '', 'AUTOMATIC', 'GASOLINE', 'OTHER', 'Other', carVin, 'OTHER', carPrice + " USD", dealerAddress, carColor, carPrice + " USD", 'AVAILABLE', 'USED', dealerLat, dealerLong];
newResults.push(arrSetup);
}
catch(error){
returnString += error;
}
}
return newResults;
})
.catch(error => {
throw new Error(error);
});
});
}, Promise.resolve([]))
.then(function(finalCall){
/*
We need to get the 3rd image on every vdp in the array. We will need to create a loop, go to the page, get the image and properly insert it into the proper array index
*/
finalCall.reduce(function(accumulator, resultArray){
return accumulator.then(function(finalResults){
var vdp = resultArray[3];
return nightmare.goto(vdp)
.wait(500)
.evaluate(() => {
var thirdIMG = $('.gallery-thumbs .owl-item:nth-of-type(3) img').attr('src');
return thirdIMG;
})
.then(imgResult => {
// 9
resultArray.splice(9, 1, imgResult);
console.log(resultArray);
finalResults.push(resultArray);
return finalResults;
})
.catch(error => {
throw new Error(error);
});
});
}, Promise.resolve([]))
.then(finalInsert => {
const csvWriter = createCsvWriter({
header: ["vehicle_id", "title", "description", "url", "make", "model", "year", "mileage.value", "mileage.unit", "image[0].url", "image[0].tag[0]", "transmission", "fuel_type", "body_style", "drivetrain", "vin", "condition", "price", "address", "exterior_color", "sale_price", "availability", "state_of_vehicle", "latitude", "longitude"],
path: 'test.csv'
});
var records = finalInsert;
console.log(records)
csvWriter.writeRecords(records)
.then(() => {
nightmare.end();
console.log('...Done');
});
})
});
})
.catch(function(error){
return error;
})
}
getInventory(startURL, function(response){
try {
console.log("This is the response" + response);
}
catch(error){
console.log(error)
}
});