correct version

* preparing scraping ant proxies

* adding general settings for scraping ant proxy

* retrying with new ui settings

CHANGELOG.md

@@ -2,6 +2,12 @@ Newer release changelog see https://github.com/orangecoding/fredy/releases
 
 ------------
 
+###### [V5.5.0]
+- Upgrading dependencies
+- fixing provider
+- allow multiple instances of 1 provider  
+- __BREAKING__: Minimum node version is now 16
+
 ###### [V5.4.6]
 - Adding Instana node.js monitoring
 - 

conf/config.json

@@ -1 +1 @@
-{"interval":"60","port":9998,"scrapingAnt":{"apiKey":""},"workingHours":{"from":"","to":""}}
+{"interval":"60","port":9998,"scrapingAnt":{"apiKey":"","proxy":"datacenter"},"workingHours":{"from":"","to":""}}

index.js

@@ -31,33 +31,20 @@ setInterval(
 
     if (isDuringWorkingHoursOrNotSet) {
       config.lastRun = Date.now();
+      const fetchedProvider = provider
+        .filter((provider) => provider.endsWith('.js'))
+        .map((pro) => require(`${path}/${pro}`));
+
       jobStorage
         .getJobs()
         .filter((job) => job.enabled)
         .forEach((job) => {
-          const providerIds = job.provider.map((provider) => provider.id);
-
-          provider
-            .filter((provider) => provider.endsWith('.js'))
-            .map((pro) => require(`${path}/${pro}`))
-            .filter((provider) => providerIds.indexOf(provider.metaInformation.id) !== -1)
-            .forEach(async (pro) => {
-              const providerId = pro.metaInformation.id;
-              if (providerId == null || providerId.length === 0) {
-                throw new Error('Provider id must not be empty. => ' + pro);
-              }
-              const providerConfig = job.provider.find((jobProvider) => jobProvider.id === providerId);
-              if (providerConfig == null) {
-                throw new Error(`Provider Config for provider with id ${providerId} not found.`);
-              }
-              pro.init(providerConfig, job.blacklist);
-              await new FredyRuntime(
-                pro.config,
-                job.notificationAdapter,
-                providerId,
-                job.id,
-                similarityCache
-              ).execute();
+          job.provider
+            .filter((p) => fetchedProvider.find((fp) => fp.metaInformation.id === p.id) != null)
+            .forEach(async (prov) => {
+              const pro = fetchedProvider.find((fp) => fp.metaInformation.id === prov.id);
+              pro.init(prov, job.blacklist);
+              await new FredyRuntime(pro.config, job.notificationAdapter, prov.id, job.id, similarityCache).execute();
               setLastJobExecution(job.id);
             });
         });

lib/notification/adapter/telegram.js

@@ -2,6 +2,8 @@ const { markdown2Html } = require('../../services/markdown');
 const { getJob } = require('../../services/storage/jobStorage');
 const axios = require('axios');
 
+const MAX_ENTITIES_PER_CHUNK = 8;
+const RATE_LIMIT_INTERVAL = 1010;
 /**
  * splitting an array into chunks because Telegram only allows for messages up to
  * 4096 chars, thus we have to split messages into chunks
@@ -29,7 +31,7 @@ exports.send = ({ serviceName, newListings, notificationConfig, jobKey }) => {
   const jobName = job == null ? jobKey : job.name;
 
   //we have to split messages into chunk, because otherwise messages are going to become too big and will fail
-  const chunks = arrayChunks(newListings, 3);
+  const chunks = arrayChunks(newListings, MAX_ENTITIES_PER_CHUNK);
 
   const promises = chunks.map((chunk) => {
     let message = `<i>${jobName}</i> (${serviceName}) found <b>${newListings.length}</b> new listings:\n\n`;
@@ -40,11 +42,21 @@ exports.send = ({ serviceName, newListings, notificationConfig, jobKey }) => {
         '\n\n'
     );
 
-    return axios.post(`https://api.telegram.org/bot${token}/sendMessage`, {
-      chat_id: chatId,
-      text: message,
-      parse_mode: 'HTML',
-      disable_web_page_preview: true,
+    /**
+     * This is to not break the rate limit. It is to only send 1 message per second
+     */
+    return new Promise((resolve, reject) => {
+      setTimeout(() => {
+        axios
+          .post(`https://api.telegram.org/bot${token}/sendMessage`, {
+            chat_id: chatId,
+            text: message,
+            parse_mode: 'HTML',
+            disable_web_page_preview: true,
+          })
+          .then(() => resolve())
+          .catch(() => reject());
+      }, RATE_LIMIT_INTERVAL);
     });
   });
 

lib/services/requestDriver.js

@@ -1,14 +1,21 @@
 const axios = require('axios');
-const axiosRetry = require('axios-retry');
+const config = require('../../conf/config.json');
 
-axiosRetry(axios, { retryDelay: axiosRetry.exponentialDelay, retries: 3 });
+const { makeUrlResidential } = require('./scrapingAnt');
+//if ScrapingAnt got blocked, this http status is returned
+const BLOCKED_HTTP_STATUS = 423;
+const NOT_FOUND_HTTP_STATUS = 404;
+const MAX_RETRIES_SCRAPING_ANT = 10;
+const EXPECTED_STATUS_CODES = [BLOCKED_HTTP_STATUS, NOT_FOUND_HTTP_STATUS];
 
 function makeDriver(headers = {}) {
   let cookies = '';
 
-  return async function driver(context, callback) {
+  async function scrapingAntDriver(context, callback, retryCounter = 0) {
+    const proxyType = config.scrapingAnt?.proxy || 'datacenter';
+
     try {
-      const url = context.url;
+      const url = proxyType === 'residential' ? makeUrlResidential(context.url) : context.url;
       const result = await axios({
         url,
         headers: {
@@ -17,15 +24,50 @@ function makeDriver(headers = {}) {
         },
       });
 
-      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);
+      if (cookies.length === 0) {
+        cookies = result.data.cookies;
       }
+
+      callback(null, result.data.content);
+    } catch (exception) {
+      /* eslint-disable no-console */
+      if (!EXPECTED_STATUS_CODES.includes(exception.response?.status)) {
+        console.error(`Error while trying to scrape data from scraping ant. Received error: ${exception.message}`);
+        callback(null, []);
+        return;
+      }
+
+      if (retryCounter <= MAX_RETRIES_SCRAPING_ANT) {
+        retryCounter++;
+        console.debug(`ScrapingAnt got blocked. Retrying ${retryCounter} / ${MAX_RETRIES_SCRAPING_ANT}`);
+        await scrapingAntDriver(context, callback, retryCounter);
+      } else {
+        console.error(`Error while trying to scrape data from scraping ant. Received error: ${exception.message}`);
+        callback(null, []);
+      }
+      /* eslint-enable no-console */
+    }
+  }
+
+  /**
+   * The regular request driver is taking care of everyting, that doesn't need to be scraped by ScrapingAnt (which is
+   * everything != Immoscout as of writing this)
+   */
+  return async function driver(context, callback) {
+    if (context.url.toLowerCase().indexOf('scrapingant') !== -1) {
+      return scrapingAntDriver(context, callback);
+    }
+
+    try {
+      const result = await axios({
+        url: context.url,
+        headers: {
+          ...headers,
+          Cookie: cookies,
+        },
+      });
+
+      callback(null, result.data);
     } catch (exception) {
       console.error(`Error while trying to scrape data. Received error: ${exception.message}`);
       callback(null, []);

lib/services/scrapingAnt.js

@@ -1,6 +1,5 @@
 const { metaInformation } = require('../provider/immoscout');
 //to better configure 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) => {
@@ -8,13 +7,9 @@ const isImmoscout = (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}&proxy_type=residential`;
+    url = `https://api.scrapingant.com/v1/general?url=${encodeURIComponent(url)}&proxy_type=datacenter`;
   }
   return url;
 };
@@ -24,3 +19,7 @@ exports.isScrapingAntApiKeySet = () => {
 };
 
 exports.isImmoscout = isImmoscout;
+
+exports.makeUrlResidential = (url) => {
+  return url.replace('datacenter', 'residential');
+};

package.json

@@ -1,13 +1,17 @@
 {
   "name": "fredy",
-  "version": "5.5.0",
+  "version": "5.8.0",
   "description": "[F]ind [R]eal [E]states [d]amn eas[y].",
   "scripts": {
     "start": "node index.js",
-    "dev": "yarn && export BUILD_DEV='true' && export NODE_ENV='development' && webpack serve --progress --color --config ./webpack.dev.js",
-    "prod": "export BUILD_DEV='false' && webpack --node-env=production --config ./webpack.prod.js",
+    "dev": "run-script-os",
+    "dev:win32": "yarn && set BUILD_DEV='true' && set NODE_ENV='development' && webpack serve --progress --color --config ./webpack.dev.js",
+    "dev:default": "yarn && export BUILD_DEV='true' && export NODE_ENV='development' && webpack serve --progress --color --config ./webpack.dev.js",
+    "prod": "run-script-os",
+    "prod:win32": "set BUILD_DEV='false' && webpack --node-env=production --config ./webpack.prod.js",
+    "prod:default": "export BUILD_DEV='false' && webpack --node-env=production --config ./webpack.prod.js",
     "format": "prettier --write lib/**/*.js ui/src/**/*.js test/**/*.js *.js --single-quote --print-width 120",
-    "test": "mocha --timeout 20000 test/**/*.test.js",
+    "test": "mocha --timeout 3000000 test/**/*.test.js",
     "lint": "eslint ./index.js ./lib/**/*.js ./test/**/*.js"
   },
   "husky": {
@@ -43,7 +47,7 @@
   },
   "license": "MIT",
   "engines": {
-    "node": ">=14.0.0",
+    "node": ">=16.0.0",
     "npm": ">=7.0.0"
   },
   "browserslist": [
@@ -55,61 +59,61 @@
   "dependencies": {
     "@rematch/core": "2.2.0",
     "@rematch/loading": "2.1.2",
-    "@sendgrid/mail": "7.6.2",
-    "axios": "0.26.1",
-    "axios-retry": "^3.2.4",
-    "better-sqlite3": "^7.5.0",
-    "body-parser": "1.19.2",
+    "@sendgrid/mail": "7.7.0",
+    "axios": "0.27.2",
+    "better-sqlite3": "8.0.1",
+    "body-parser": "1.20.1",
     "cookie-session": "2.0.0",
     "handlebars": "4.7.7",
-    "highcharts": "10.0.0",
+    "highcharts": "10.3.2",
     "highcharts-react-official": "3.1.0",
     "lowdb": "1.0.0",
     "markdown": "^0.5.0",
-    "nanoid": "3.3.1",
-    "node-mailjet": "3.3.7",
-    "query-string": "7.1.1",
+    "nanoid": "3.3.3",
+    "node-mailjet": "3.3.13",
+    "query-string": "7.1.3",
     "react": "17.0.2",
     "react-dom": "17.0.2",
-    "react-redux": "7.2.6",
+    "react-redux": "8.0.5",
     "react-router": "5.2.1",
     "react-router-dom": "5.3.0",
-    "react-switch": "^6.0.0",
-    "redux": "4.1.2",
-    "redux-thunk": "2.4.1",
-    "restana": "4.9.3",
-    "semantic-ui-react": "2.1.2",
+    "react-switch": "6.0.0",
+    "redux": "4.2.0",
+    "redux-thunk": "2.4.2",
+    "restana": "4.9.7",
+    "semantic-ui-react": "2.1.4",
     "serve-static": "1.15.0",
     "slack": "11.0.2",
     "string-similarity": "^4.0.4",
     "x-ray": "2.3.4"
   },
   "devDependencies": {
-    "@babel/core": "7.17.8",
-    "@babel/preset-env": "7.16.11",
-    "@babel/preset-react": "7.16.7",
+    "@babel/core": "7.20.5",
+    "@babel/preset-env": "7.20.2",
+    "@babel/preset-react": "7.18.6",
     "babel-eslint": "10.1.0",
-    "babel-loader": "8.2.4",
-    "chai": "4.3.6",
+    "babel-loader": "8.2.5",
+    "chai": "4.3.7",
     "clean-webpack-plugin": "4.0.0",
     "copy-webpack-plugin": "10.2.4",
-    "css-loader": "6.7.1",
+    "css-loader": "6.7.2",
     "eslint": "7.32.0",
     "eslint-config-prettier": "8.5.0",
-    "eslint-plugin-react": "7.29.4",
+    "eslint-plugin-react": "7.31.11",
     "file-loader": "6.2.0",
     "history": "5.3.0",
     "husky": "4.3.8",
-    "less": "4.1.2",
+    "less": "4.1.3",
     "less-loader": "10.2.0",
-    "lint-staged": "12.3.7",
+    "lint-staged": "12.4.1",
     "mocha": "9.2.2",
-    "prettier": "2.6.1",
+    "prettier": "2.8.1",
     "proxyquire": "2.1.3",
     "redux-logger": "3.0.6",
+    "run-script-os": "^1.1.6",
     "style-loader": "3.3.1",
     "url-loader": "4.1.1",
-    "webpack": "5.70.0",
+    "webpack": "5.75.0",
     "webpack-cli": "4.9.2",
     "webpack-dev-server": "3.11.2",
     "webpack-merge": "5.8.0"

test/provider/einsAImmobilien.test.js

@@ -22,7 +22,7 @@ describe('#einsAImmobilien testsuite()', () => {
 
   it('should test einsAImmobilien provider', async () => {
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'einsAImmobilien', similarityCache);
       fredy.execute().then((listings) => {
         expect(listings).to.be.a('array');
 
@@ -39,7 +39,6 @@ describe('#einsAImmobilien testsuite()', () => {
           expect(notify.link).to.be.a('string');
 
           /** check the values if possible **/
-          expect(notify.price).that.does.include('EUR');
           expect(notify.size).to.be.not.empty;
           expect(notify.title).to.be.not.empty;
           expect(notify.link).that.does.include('https://www.1a-immobilienmarkt.de');

test/provider/immonet.test.js

@@ -21,7 +21,7 @@ describe('#immonet testsuite()', () => {
 
   it('should test immonet provider', async () => {
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'immonet', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
 

test/provider/immoscout.test.js

@@ -29,7 +29,7 @@ describe('#immoscout testsuite()', () => {
         return;
       }
 
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'immoscout', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
 

test/provider/immoswp.test.js

@@ -21,7 +21,7 @@ describe('#immoswp testsuite()', () => {
 
   it('should test immoswp provider', async () => {
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'immoswp', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
 

test/provider/immowelt.test.js

@@ -20,7 +20,7 @@ describe('#immowelt testsuite()', () => {
     });
 
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'immowelt', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
 

test/provider/kleinanzeigen.test.js

@@ -20,7 +20,7 @@ describe('#kleinanzeigen testsuite()', () => {
     });
 
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'kleinanzeigen', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
 
@@ -31,10 +31,8 @@ describe('#kleinanzeigen testsuite()', () => {
         notificationObj.payload.forEach((notify) => {
           /** check the actual structure **/
           expect(notify.id).to.be.a('number');
-          expect(notify.size).to.be.a('string');
           expect(notify.title).to.be.a('string');
           expect(notify.link).to.be.a('string');
-          expect(notify.price).to.be.a('string');
           expect(notify.address).to.be.a('string');
 
           /** check the values if possible **/

test/provider/neubauKompass.test.js

@@ -20,7 +20,7 @@ describe('#neubauKompass testsuite()', () => {
 
   it('should test neubauKompass provider', async () => {
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'neubauKompass', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
 

test/provider/wgGesucht.test.js

@@ -20,7 +20,7 @@ describe('#wgGesucht testsuite()', () => {
 
   it('should test wgGesucht provider', async () => {
     return await new Promise((resolve) => {
-      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'test1', similarityCache);
+      const fredy = new Fredy(provider.config, null, provider.metaInformation.id, 'wgGesucht', similarityCache);
       fredy.execute().then((listing) => {
         expect(listing).to.be.a('array');
         const notificationObj = mockNotification.get();

ui/src/views/generalSettings/GeneralSettings.js

@@ -2,7 +2,7 @@ import React from 'react';
 
 import { useDispatch, useSelector } from 'react-redux';
 
-import { Button, Form, Icon, Message, Segment } from 'semantic-ui-react';
+import { Button, Form, Icon, Message, Segment, Radio } from 'semantic-ui-react';
 import ToastContext from '../../components/toasts/ToastContext';
 import Headline from '../../components/headline/Headline';
 import { xhrPost } from '../../services/xhr';
@@ -18,6 +18,7 @@ const GeneralSettings = function Users() {
   const [interval, setInterval] = React.useState('');
   const [port, setPort] = React.useState('');
   const [scrapingAntApiKey, setScrapingAntApiKey] = React.useState('');
+  const [scrapingAntProxy, setScrapingAntProxy] = React.useState('');
   const [workingHourFrom, setWorkingHourFrom] = React.useState(null);
   const [workingHourTo, setWorkingHourTo] = React.useState(null);
   const ctx = React.useContext(ToastContext);
@@ -33,6 +34,7 @@ const GeneralSettings = function Users() {
     setScrapingAntApiKey(settings?.scrapingAnt?.apiKey);
     setWorkingHourFrom(settings?.workingHours?.from);
     setWorkingHourTo(settings?.workingHours?.to);
+    setScrapingAntProxy(settings?.scrapingAnt?.proxy || 'datacenter');
   }, [settings]);
 
   const nullOrEmpty = (val) => val == null || val.length === 0;
@@ -69,6 +71,7 @@ const GeneralSettings = function Users() {
         port,
         scrapingAnt: {
           apiKey: scrapingAntApiKey,
+          proxy: scrapingAntProxy,
         },
         workingHours: {
           from: workingHourFrom,
@@ -144,6 +147,48 @@ const GeneralSettings = function Users() {
               />
             </SegmentPart>
 
+            <SegmentPart
+              name="ScrapingAnt proxy settings"
+              helpText="Scraping ant provides different proxies."
+              icon="key"
+            >
+              <Message info>
+                ScrapingAnt is needed to scrape Immoscout. ScrapingAnt itself is using 2 different types of proxies.{' '}
+                <br />
+                <h4>Datacenter-Proxy</h4>
+                Proxy server located in one of the datacenters across the world. Datacenter proxies are slower and more
+                likely to fail, but they are cheaper. A call with a datacenter proxy cost 10 credits.
+                <h4>Residential-Proxy</h4>
+                High-quality proxy server located in one of the real people houses across the world. Datacenter proxies
+                are faster and more likely to success, but they are more expensive. A call with a datacenter proxy cost
+                250 credits.
+                <br />
+                <br />
+                <b>
+                  On the free tier, you have 10.000 credits, so chose your option wisely. Keep in mind, only successful
+                  calls will be charged.
+                </b>
+              </Message>
+              <Form.Field>
+                <Radio
+                  label="Datacenter proxy"
+                  name="scrapingAntProxy"
+                  value="datacenter"
+                  checked={scrapingAntProxy === 'datacenter'}
+                  onChange={(e, { value }) => setScrapingAntProxy(value)}
+                />
+              </Form.Field>
+              <Form.Field>
+                <Radio
+                  label="Residential proxy"
+                  name="scrapingAntProxy"
+                  value="residential"
+                  checked={scrapingAntProxy === 'residential'}
+                  onChange={(e, { value }) => setScrapingAntProxy(value)}
+                />
+              </Form.Field>
+            </SegmentPart>
+
             <SegmentPart
               name="Working hours"
               helpText="During this hours, Fredy will search for new apartments. If nothing is configured, Fredy will search around the clock."
@@ -153,7 +198,7 @@ const GeneralSettings = function Users() {
                 <Form.Input
                   className="generalSettings__time"
                   type="time"
-                  placeholder="ScrapingAnt Api Key"
+                  placeholder="Working hours from"
                   inverted
                   size="mini"
                   width={2}
@@ -163,7 +208,7 @@ const GeneralSettings = function Users() {
                 <div className="generalSettings__until">until</div>
                 <Form.Input
                   type="time"
-                  placeholder="ScrapingAnt Api Key"
+                  placeholder="Working hours to"
                   inverted
                   size="mini"
                   width={2}

ui/src/views/jobs/ProcessingTimes.js

@@ -43,7 +43,7 @@ export default function ProcessingTimes({ processingTimes }) {
             ScrapingAnt
           </a>
           . You can use the code <b>FREDY10</b> to get 10% off. (No affiliation, we are <b>not</b> getting paid to
-          recommend ScrapingAnt.
+          recommend ScrapingAnt.)
         </Segment>
       )}
     </React.Fragment>

ui/src/views/jobs/mutation/JobMutation.js

@@ -81,7 +81,6 @@ export default function JobMutator() {
       <ProviderMutator
         visible={providerCreationVisible}
         onVisibilityChanged={(visible) => setProviderCreationVisibility(visible)}
-        selected={providerData}
         onData={(data) => {
           setProviderData([...providerData, data]);
         }}

ui/src/views/jobs/mutation/components/provider/ProviderMutator.js

@@ -16,7 +16,7 @@ const sortProvider = (a, b) => {
   return 0;
 };
 
-export default function ProviderMutator({ onVisibilityChanged, visible = false, selected = [], onData } = {}) {
+export default function ProviderMutator({ onVisibilityChanged, visible = false, onData } = {}) {
   const provider = useSelector((state) => state.provider);
   const [selectedProvider, setSelectedProvider] = useState(null);
   const [providerUrl, setProviderUrl] = useState(null);
@@ -107,8 +107,6 @@ export default function ProviderMutator({ onVisibilityChanged, visible = false,
                   text: pro.name,
                 };
               })
-              //filter out those, that have already been selected
-              .filter((option) => selected.find((selectedOption) => selectedOption.id === option.key) == null)
               .sort(sortProvider)}
             onChange={(e, { value }) => {
               const selectedProvider = provider.find((pro) => pro.id === value);