2025-09-22 09:57:50 +02:00
|
|
|
import { buildHash, isOneOf } from '../utils.js';
|
|
|
|
|
import checkIfListingIsActive from '../services/listings/listingActiveTester.js';
|
2024-09-05 13:34:14 +02:00
|
|
|
|
2020-02-26 09:05:20 +01:00
|
|
|
let appliedBlackList = [];
|
2024-09-05 13:34:14 +02:00
|
|
|
|
2018-01-20 20:23:27 +01:00
|
|
|
function normalize(o) {
|
2024-10-03 16:03:47 +02:00
|
|
|
const id = buildHash(o.id, o.price);
|
|
|
|
|
return Object.assign(o, { id });
|
2018-01-20 20:23:27 +01:00
|
|
|
}
|
2024-09-05 13:34:14 +02:00
|
|
|
|
2018-01-20 20:23:27 +01:00
|
|
|
function applyBlacklist(o) {
|
2025-09-22 09:57:50 +02:00
|
|
|
const titleNotBlacklisted = !isOneOf(o.title, appliedBlackList);
|
|
|
|
|
const descNotBlacklisted = !isOneOf(o.description, appliedBlackList);
|
2024-10-03 16:03:47 +02:00
|
|
|
return titleNotBlacklisted && descNotBlacklisted;
|
2018-01-20 20:23:27 +01:00
|
|
|
}
|
2024-09-05 13:34:14 +02:00
|
|
|
|
2020-02-26 09:05:20 +01:00
|
|
|
const config = {
|
2024-10-03 16:03:47 +02:00
|
|
|
url: null,
|
|
|
|
|
crawlContainer:
|
2024-12-17 12:38:28 +01:00
|
|
|
'div[data-testid="serp-core-scrollablelistview-testid"]:not(div[data-testid="serp-enlargementlist-testid"] div[data-testid="serp-card-testid"]) div[data-testid="serp-core-classified-card-testid"]',
|
2024-10-03 16:03:47 +02:00
|
|
|
sortByDateParam: 'order=DateDesc',
|
2025-01-07 12:37:50 +01:00
|
|
|
waitForSelector: 'div[data-testid="serp-gridcontainer-testid"]',
|
2024-10-03 16:03:47 +02:00
|
|
|
crawlFields: {
|
2024-12-17 12:38:28 +01:00
|
|
|
id: 'a@href',
|
2024-10-03 16:03:47 +02:00
|
|
|
price: 'div[data-testid="cardmfe-price-testid"] | removeNewline | trim',
|
|
|
|
|
size: 'div[data-testid="cardmfe-keyfacts-testid"] | removeNewline | trim',
|
2025-05-26 10:43:13 +02:00
|
|
|
title: 'div[data-testid="cardmfe-description-box-text-test-id"] > div:nth-of-type(2)',
|
2024-10-03 16:03:47 +02:00
|
|
|
link: 'a@href',
|
2025-09-27 09:42:08 +02:00
|
|
|
description: 'div[data-testid="cardmfe-description-text-test-id"] > div:nth-of-type(2) | removeNewline | trim',
|
2024-10-03 16:03:47 +02:00
|
|
|
address: 'div[data-testid="cardmfe-description-box-address"] | removeNewline | trim',
|
2025-09-27 09:42:08 +02:00
|
|
|
image: 'div[data-testid="cardmfe-picture-box-opacity-layer-test-id"] img@src',
|
2024-10-03 16:03:47 +02:00
|
|
|
},
|
|
|
|
|
normalize: normalize,
|
|
|
|
|
filter: applyBlacklist,
|
2025-09-22 09:57:50 +02:00
|
|
|
activeTester: checkIfListingIsActive,
|
2018-01-20 20:23:27 +01:00
|
|
|
};
|
2023-03-13 13:42:43 +01:00
|
|
|
export const init = (sourceConfig, blacklist) => {
|
2024-10-03 16:03:47 +02:00
|
|
|
config.enabled = sourceConfig.enabled;
|
|
|
|
|
config.url = sourceConfig.url;
|
|
|
|
|
appliedBlackList = blacklist || [];
|
2021-01-21 16:09:23 +01:00
|
|
|
};
|
2023-03-13 13:42:43 +01:00
|
|
|
export const metaInformation = {
|
2024-10-03 16:03:47 +02:00
|
|
|
name: 'Immowelt',
|
|
|
|
|
baseUrl: 'https://www.immowelt.de/',
|
|
|
|
|
id: 'immowelt',
|
2020-02-26 09:05:20 +01:00
|
|
|
};
|
2024-10-03 16:03:47 +02:00
|
|
|
export { config };
|