-
-
Save tridungle/33f6439d20501f6a9c455cda2b6a77d6 to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
const crawler = require("web-crawljs"); | |
/**// | |
* @description removes duplicate hotels for the array | |
* @param arr | |
* @returns {Array.<T>|*} | |
*/ | |
function makeUnique(arr) { | |
"use strict"; | |
const key = {}; | |
return arr.filter(hotel => { | |
if (key[`${hotel.hotel_name}_${hotel.city}_${hotel.state}`]) { | |
return false | |
} | |
key[`${hotel.hotel_name}_${hotel.city}_${hotel.state}`] = true; | |
return true | |
}) | |
} | |
const mongoose = require("mongoose"); | |
require("dotenv").config(); | |
const env = process.env; | |
mongoose.Promise = Promise; | |
mongoose.connect(env.MONGO_URI); | |
const hotelModel = require("./hotelModel"); | |
let allData = []; | |
const config = { | |
fetchSelector: { | |
hotel_name: "div.about-item > h2 > a", | |
location: "div p.item-location span", | |
city: "div p.item-location a:first-child", | |
state: "div p.item-location a:nth-child(2)", | |
present_price: "div.present-price span", | |
features: ".item-features", | |
hotelsNg_link: "div.about-item > h2 > a" | |
}, | |
fetchSelectBy: { | |
hotel_name: "text", | |
location: "text", | |
city: "text", | |
state: "text", | |
present_price: "text", | |
features: "text", | |
hotelsNg_link: ['attr', 'href'] | |
}, | |
nextSelector: {links: 'ul.pagination a'}, | |
nextSelectBy: {links: ['attr', 'href']}, | |
fetchFn: (err, data, url) => { | |
if (err) console.error(err.message); | |
const val = []; | |
data.hotel_name.forEach((item, i) => { | |
"use strict"; | |
let f = data.features[i].trim().replace(/(\s{2,}|\n)/g, ":"); //removing unneeded spaces and tabs | |
f = (f.length > 0) ? f.split(":") : []; | |
let location = data.location[i] ? data.location[i] : " "; | |
val.push({ | |
hotel_name: item.trim(), | |
location: location.trim().replace(/(\n|\s\s)/g, ""), | |
city: data.city[i].replace(/,$/, ""), | |
state: data.state[i], | |
present_price: data.present_price[i], | |
features: f, | |
hotelsNg_link: data.hotelsNg_link[i] | |
}) | |
}); | |
allData = allData.concat(val) | |
// console.log(val, val.length) | |
}, | |
finalFn: function () { | |
hotelModel.insertMany(makeUnique(allData)).then(model => { | |
allData = []; | |
console.log("model inserted", model.length) | |
}) | |
.catch(err => { | |
allData = []; | |
console.error(err) | |
}) | |
}, | |
depth: process.env.DEPTH, | |
urls: [ | |
"https://hotels.ng/hotels-in-abia", | |
"https://hotels.ng/hotels-in-abuja", | |
"https://hotels.ng/hotels-in-lagos", | |
] | |
}; | |
const Crawler = crawler(config); | |
// hotelModel.remove({}).then(p => console.log(p)) | |
// .catch(err => console.error(err)) | |
//crawl all the link | |
Crawler.CrawlAllUrl(); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment