mirror of
https://github.com/orangecoding/fredy.git
synced 2026-06-16 12:31:07 +00:00
better logging, fixing code smells
This commit is contained in:
@@ -1,5 +1,5 @@
|
||||
import {NoNewListingsWarning} from './errors.js';
|
||||
import {setKnownListings, getKnownListings} from './services/storage/listingsStorage.js';
|
||||
import { NoNewListingsWarning } from './errors.js';
|
||||
import { setKnownListings, getKnownListings } from './services/storage/listingsStorage.js';
|
||||
import * as notify from './notification/notify.js';
|
||||
import Extractor from './services/extractor/extractor.js';
|
||||
import urlModifier from './services/queryStringMutator.js';
|
||||
@@ -47,11 +47,17 @@ class FredyRuntime {
|
||||
_getListings(url) {
|
||||
const extractor = new Extractor();
|
||||
return new Promise((resolve, reject) => {
|
||||
extractor.execute(url,this._providerConfig.waitForSelector)
|
||||
extractor
|
||||
.execute(url, this._providerConfig.waitForSelector)
|
||||
.then(() => {
|
||||
const listings = extractor.parseResponseText(this._providerConfig.crawlContainer, this._providerConfig.crawlFields);
|
||||
const listings = extractor.parseResponseText(
|
||||
this._providerConfig.crawlContainer,
|
||||
this._providerConfig.crawlFields,
|
||||
url,
|
||||
);
|
||||
resolve(listings == null ? [] : listings);
|
||||
}).catch(err => {
|
||||
})
|
||||
.catch((err) => {
|
||||
reject(err);
|
||||
/* eslint-disable no-console */
|
||||
console.error(err);
|
||||
|
||||
@@ -1,11 +1,9 @@
|
||||
import restana from 'restana';
|
||||
import fetch from 'node-fetch';
|
||||
import * as jobStorage from '../../services/storage/jobStorage.js';
|
||||
import * as userStorage from '../../services/storage/userStorage.js';
|
||||
import * as immoscoutProvider from '../../provider/immoscout.js';
|
||||
import { config } from '../../utils.js';
|
||||
import { isAdmin } from '../security.js';
|
||||
import {trackDemoJobCreated} from '../../services/tracking/Tracker.js';
|
||||
import { trackDemoJobCreated } from '../../services/tracking/Tracker.js';
|
||||
const service = restana();
|
||||
const jobRouter = service.newRouter();
|
||||
function doesJobBelongsToUser(job, req) {
|
||||
@@ -28,7 +26,7 @@ jobRouter.get('/', async (req, res) => {
|
||||
jobRouter.get('/processingTimes', async (req, res) => {
|
||||
res.body = {
|
||||
interval: config.interval,
|
||||
lastRun: config.lastRun || null
|
||||
lastRun: config.lastRun || null,
|
||||
};
|
||||
res.send();
|
||||
});
|
||||
@@ -51,7 +49,7 @@ jobRouter.post('/', async (req, res) => {
|
||||
trackDemoJobCreated({
|
||||
name,
|
||||
provider,
|
||||
adapter: notificationAdapter
|
||||
adapter: notificationAdapter,
|
||||
});
|
||||
res.send();
|
||||
});
|
||||
|
||||
@@ -1,19 +1,18 @@
|
||||
import {setDebug} from './utils.js';
|
||||
import { setDebug } from './utils.js';
|
||||
import puppeteerExtractor from './puppeteerExtractor.js';
|
||||
import {loadParser, parse} from './parser/parser.js';
|
||||
import { loadParser, parse } from './parser/parser.js';
|
||||
|
||||
const DEFAULT_OPTIONS = {
|
||||
debug: false,
|
||||
puppeteerTimeout: 20_000,
|
||||
puppeteerHeadless: true
|
||||
|
||||
puppeteerTimeout: 60_000,
|
||||
puppeteerHeadless: true,
|
||||
};
|
||||
|
||||
export default class Extractor {
|
||||
constructor(options) {
|
||||
this.options = {
|
||||
...DEFAULT_OPTIONS,
|
||||
...options
|
||||
...options,
|
||||
};
|
||||
this.responseText = null;
|
||||
setDebug(this.options);
|
||||
@@ -29,7 +28,7 @@ export default class Extractor {
|
||||
this.responseText = null;
|
||||
try {
|
||||
this.responseText = await puppeteerExtractor(url, waitForSelector, this.options);
|
||||
if(this.responseText != null) {
|
||||
if (this.responseText != null) {
|
||||
loadParser(this.responseText);
|
||||
}
|
||||
} catch (error) {
|
||||
@@ -38,8 +37,7 @@ export default class Extractor {
|
||||
return this;
|
||||
};
|
||||
|
||||
|
||||
parseResponseText = (crawlContainer, crawlFields) => {
|
||||
return parse(crawlContainer, crawlFields, this.responseText);
|
||||
parseResponseText = (crawlContainer, crawlFields, url) => {
|
||||
return parse(crawlContainer, crawlFields, this.responseText, url);
|
||||
};
|
||||
}
|
||||
|
||||
@@ -6,21 +6,21 @@ export function loadParser(text) {
|
||||
$ = cheerio.load(text);
|
||||
}
|
||||
|
||||
export function parse(crawlContainer, crawlFields, text) {
|
||||
export function parse(crawlContainer, crawlFields, text, url) {
|
||||
if (!text) {
|
||||
console.warn('Cannot parse, text was empty.');
|
||||
console.warn('Cannot parse, text was empty for url ', url);
|
||||
return null;
|
||||
}
|
||||
|
||||
if (!crawlContainer || !crawlFields) {
|
||||
console.warn('Cannot parse, selector was empty.');
|
||||
console.warn('Cannot parse, selector was empty for url ', url);
|
||||
return null;
|
||||
}
|
||||
|
||||
const result = [];
|
||||
|
||||
if ($(crawlContainer).length === 0) {
|
||||
console.error('No elements in crawl container found!');
|
||||
console.error('No elements in crawl container found for url ', url);
|
||||
}
|
||||
|
||||
$(crawlContainer).each((_, element) => {
|
||||
@@ -32,8 +32,9 @@ export function parse(crawlContainer, crawlFields, text) {
|
||||
let value;
|
||||
|
||||
try {
|
||||
|
||||
const selector = fieldSelector.includes('|') ? fieldSelector.substring(0, fieldSelector.indexOf('|')).trim() : fieldSelector;
|
||||
const selector = fieldSelector.includes('|')
|
||||
? fieldSelector.substring(0, fieldSelector.indexOf('|')).trim()
|
||||
: fieldSelector;
|
||||
|
||||
if (selector.includes('@')) {
|
||||
const [sel, attr] = selector.split('@');
|
||||
@@ -48,7 +49,9 @@ export function parse(crawlContainer, crawlFields, text) {
|
||||
|
||||
// Apply modifiers if specified
|
||||
if (fieldSelector.includes('|')) {
|
||||
const [_, ...modifiers] = fieldSelector.split('|').map(s => s.trim());
|
||||
/* eslint-disable no-unused-vars */
|
||||
const [_, ...modifiers] = fieldSelector.split('|').map((s) => s.trim());
|
||||
/* eslint-disable no-unused-vars */
|
||||
value = applyModifiers(value, modifiers);
|
||||
}
|
||||
|
||||
@@ -73,7 +76,7 @@ export function parse(crawlContainer, crawlFields, text) {
|
||||
function applyModifiers(value, modifiers) {
|
||||
if (!value) return value;
|
||||
|
||||
modifiers.forEach(modifier => {
|
||||
modifiers.forEach((modifier) => {
|
||||
switch (modifier) {
|
||||
case 'int':
|
||||
value = parseInt(value, 10);
|
||||
@@ -91,4 +94,3 @@ function applyModifiers(value, modifiers) {
|
||||
|
||||
return value;
|
||||
}
|
||||
|
||||
|
||||
@@ -1,34 +1,31 @@
|
||||
let debuggingOn = false;
|
||||
|
||||
export const DEFAULT_HEADER = {
|
||||
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
|
||||
Accept: 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8',
|
||||
'Accept-Language': 'en-US,en;q=0.5',
|
||||
'Connection': 'keep-alive',
|
||||
Connection: 'keep-alive',
|
||||
'Upgrade-Insecure-Requests': '1',
|
||||
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/114.0.0.0 Safari/537.36'
|
||||
'User-Agent':
|
||||
'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/114.0.0.0 Safari/537.36',
|
||||
};
|
||||
|
||||
export const setDebug = options => {
|
||||
export const setDebug = (options) => {
|
||||
debuggingOn = !!options?.debug;
|
||||
};
|
||||
|
||||
export const debug = (message) => {
|
||||
if(debuggingOn) {
|
||||
if (debuggingOn) {
|
||||
/* eslint-disable no-console */
|
||||
console.debug(message);
|
||||
/* eslint-enable no-console */
|
||||
}
|
||||
};
|
||||
|
||||
export const botDetected = (pageSource, statusCode) => {
|
||||
const suspiciousStatusCodes = [
|
||||
403, 429
|
||||
];
|
||||
const botDetectionPatterns = [
|
||||
/verify you are human/i,
|
||||
/access denied/i,
|
||||
/x-amz-cf-id/i,
|
||||
];
|
||||
const suspiciousStatusCodes = [403, 429];
|
||||
const botDetectionPatterns = [/verify you are human/i, /access denied/i, /x-amz-cf-id/i];
|
||||
|
||||
const detectedInSource = botDetectionPatterns.some(pattern => pattern.test(pageSource));
|
||||
const detectedInSource = botDetectionPatterns.some((pattern) => pattern.test(pageSource));
|
||||
const detectedByStatus = suspiciousStatusCodes.includes(statusCode);
|
||||
|
||||
return detectedInSource || detectedByStatus;
|
||||
|
||||
Reference in New Issue
Block a user