Bringing back immoscout (#21)

Bringing back immoscout support 🎉
This commit is contained in:
Christian Kellner
2021-05-11 11:25:14 +02:00
committed by GitHub
parent 9726079f62
commit 70cab66651
13 changed files with 225 additions and 23 deletions

View File

@@ -3,6 +3,7 @@ const { setKnownListings, getKnownListings } = require('./services/storage/listi
const notify = require('./notification/notify');
const xray = require('./services/scraper');
const scrapingAnt = require('./services/scrapingAnt');
class FredyRuntime {
/**
@@ -41,15 +42,24 @@ class FredyRuntime {
_getListings(url) {
return new Promise((resolve, reject) => {
let x = xray(url, this._providerConfig.crawlContainer, [this._providerConfig.crawlFields]);
x((err, listings) => {
if (err) {
const id = this._providerId;
if (scrapingAnt.isImmoscout(id) && !scrapingAnt.isScrapingAntApiKeySet()) {
const error = 'Immoscout can only be used with if you have set an apikey for scrapingAnt.';
/* eslint-disable no-console */
console.log(error);
/* eslint-enable no-console */
reject(error);
return;
}
const u = scrapingAnt.isImmoscout(id) ? scrapingAnt.transformUrlForScrapingAnt(url, id) : url;
xray(u, this._providerConfig.crawlContainer, [this._providerConfig.crawlFields])
.then((listings) => {
resolve(listings == null ? [] : listings);
})
.catch((err) => {
reject(err);
} else {
resolve(listings);
}
});
console.error(err);
});
});
}

44
lib/provider/immoscout.js Normal file
View File

@@ -0,0 +1,44 @@
const utils = require('../utils');
let appliedBlackList = [];
function normalize(o) {
const title = o.title.replace('NEU', '');
const address = (o.address || '').replace(/\(.*\),.*$/, '').trim();
const link = `https://www.immobilienscout24.de${o.link.substring(o.link.indexOf('/expose'))}`;
return Object.assign(o, { title, address, link });
}
function applyBlacklist(o) {
return !utils.isOneOf(o.title, appliedBlackList);
}
const config = {
url: null,
crawlContainer: '#resultListItems li.result-list__listing',
crawlFields: {
id: '.result-list-entry@data-obid | int',
price: '.result-list-entry .result-list-entry__criteria .grid-item:first-child dd | removeNewline | trim',
size: '.result-list-entry .result-list-entry__criteria .grid-item:nth-child(2) dd | removeNewline | trim',
title: '.result-list-entry .result-list-entry__brand-title-container h5 | removeNewline | trim',
link: '.result-list-entry .result-list-entry__brand-title-container@href',
address: '.result-list-entry .result-list-entry__map-link',
},
paginate: '#pager .align-right a@href',
normalize: normalize,
filter: applyBlacklist,
};
exports.init = (sourceConfig, blacklist) => {
config.enabled = sourceConfig.enabled;
config.url = sourceConfig.url;
appliedBlackList = blacklist || [];
};
exports.metaInformation = {
name: 'Immoscout',
baseUrl: 'https://www.immobilienscout24.de/',
id: __filename.slice(__dirname.length + 1, -3),
};
exports.config = config;

View File

@@ -0,0 +1,33 @@
const axios = require('axios');
function makeDriver(headers = {}) {
let cookies = '';
return async function driver(context, callback) {
const url = context.url;
let result;
try {
result = await axios({
url,
headers: {
...headers,
Cookie: cookies,
},
});
} catch (exception) {
callback(exception, null);
}
if (typeof result.data === 'object' && url.toLowerCase().indexOf('scrapingant') !== -1) {
//assume we have gotten a response from scrapingAnt
if (cookies.length === 0) {
cookies = result.data.cookies;
}
callback(null, result.data.content);
} else {
callback(null, result.data);
}
};
}
module.exports = makeDriver;

View File

@@ -1,4 +1,5 @@
const makeDriver = require('request-x-ray');
const config = require('../../conf/config.json');
const makeDriver = require('./requestDriver');
const Xray = require('x-ray');
class Scraper {
@@ -9,14 +10,16 @@ class Scraper {
int: this._int,
};
const driver = makeDriver({
headers: {
'User-Agent':
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.85 Safari/537.36',
cookie:
'longUnreliableState="dWlkcg==:YS1kZDViMzVhZWRhMTk0MDdmYWRjNDNkY2VmYTcxZmVkOQ=="; eveD=eyJldnRfZ2FfYWN0aW9uIjpbInNlYXJjaCJdLCJldnRfZ2FfY2F0ZWdvcnkiOlsicmVzdWx0bGlzdCJdLCJnZW9fYmxuIjpbIm5vcmRyaGVpbl93ZXN0ZmFsZW4iXSwiZXZ0X2dhX2xhYmVsIjpbImRpc3RyaWN0Il0sIm9ial9pdHlwIjpbIndvaG51bmdfa2F1ZiJdLCJnZW9fa3JzIjpbImTDvHNzZWxkb3JmIl0sImdlb19sYW5kIjpbImRldXRzY2hsYW5kIl0sIm9ial9yZXN1bHRsaXN0X2NvdW50IjpbIjI4NCJdLCJvYmpfY3Jvc3N0eXBlIjpbImxpdl9hcGFydG1lbnRfYnV5Il19; ABNTEST=9526230109; is24_experiment_visitor_id=d568590b-951b-45c3-b890-13feef6ee472; reese84=3:Xf3JwcTIC3yeubDXqWBTfg==:oqnDVs58wBxZRMfpzPnlzLzscVQhboRBffkM4caxNe+vLBdozdtdrCwpcTKyvIuhB9MOMCAinb2qnSTL4D9kLpqL72gl+jtl7QdiNAEn2erDKLqX4b9/K5wFU7j6qzxFWdfcMUm295qU3o3s7O8CM8HdghKYOVtoif+qTkeztphyYMfmAePYkfYRhZXZaFwHwxUfkRVUEX2VKoepkTf9TudCHsTYXWqvnpUt/CT+yrFHlUdTgdTWfD5tQJvn3inPqKERAB8TTKoHIvM4duBJV/5fZDax07CHNqHcKhrws0pq4y2ssKfdxLxCE0OIpnMSOtmn7O0koDoV6RzRjNUC+UZ7mhPFH+YSPHTb+6VJsZQDnRufEIz4B1WWIORV+jvHzfIli9OHsmOPnskA6mnCpFwEvQAfJu9R+jI9dccjFno=:Oc7c2wwYiNMBJnvZeDCIKLP0LuVVPWJ4kzd5MPlsoTg=',
},
});
const headers = {
'User-Agent':
'Mozilla/5.0 (Windows NT 6.1; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/40.0.2214.85 Safari/537.36',
useQueryString: true,
};
if (config.scrapingAnt != null && config.scrapingAnt.apiKey != null) {
headers['x-api-key'] = config.scrapingAnt.apiKey;
}
const driver = makeDriver(headers);
const xray = Xray({ filters });
xray.driver(driver);

View File

@@ -0,0 +1,24 @@
const { metaInformation } = require('../provider/immoscout');
//to better confure re-capture chose a random proxy each time we do a call
const proxies = ['ae', 'br', 'cn', 'de', 'es', 'fr', 'gb', 'hk', 'in', 'it', 'il', 'jp', 'nl', 'ru', 'sa', 'us', 'cz'];
const config = require('../../conf/config.json');
const isImmoscout = (id) => {
return id.toLowerCase() === metaInformation.id;
};
exports.transformUrlForScrapingAnt = (url, id) => {
const randomProxy = proxies[Math.floor(Math.random() * proxies.length)];
if (isImmoscout(id)) {
//only do calls to scrapingAnt when dealing with Immoscout
url = `https://api.scrapingant.com/v1/general?url=${encodeURIComponent(url)}&proxy_country=${randomProxy}`;
}
return url;
};
exports.isScrapingAntApiKeySet = () => {
return config.scrapingAnt != null && config.scrapingAnt.apiKey != null && config.scrapingAnt.apiKey.length > 0;
};
exports.isImmoscout = isImmoscout;