mirror of
https://github.com/orangecoding/fredy.git
synced 2026-06-16 12:31:07 +00:00
Adding images to scraping data (#157)
* Fredy now supports pulling the main Image from the listing and send it together with the usual information
This commit is contained in:
committed by
GitHub
parent
da743c8279
commit
f0b146fd7f
@@ -2,10 +2,12 @@ import utils, { buildHash } from '../utils.js';
|
||||
let appliedBlackList = [];
|
||||
|
||||
function normalize(o) {
|
||||
const link = `https://www.1a-immobilienmarkt.de/expose/${o.id}.html`;
|
||||
const baseUrl = 'https://www.1a-immobilienmarkt.de';
|
||||
const link = `${baseUrl}/expose/${o.id}.html`;
|
||||
const price = normalizePrice(o.price);
|
||||
const id = buildHash(o.id, price);
|
||||
return Object.assign(o, { id, price, link });
|
||||
const image = baseUrl + o.image;
|
||||
return Object.assign(o, { id, price, link, image });
|
||||
}
|
||||
|
||||
/**
|
||||
@@ -41,6 +43,7 @@ const config = {
|
||||
price: '.inner_object_data .single_data_price | removeNewline | trim',
|
||||
size: '.tabelle .tabelle_inhalt_infos .single_data_box | removeNewline | trim',
|
||||
title: '.inner_object_data .tabelle_inhalt_titel_black | removeNewline | trim',
|
||||
image: '.inner_object_pic img@src',
|
||||
},
|
||||
normalize: normalize,
|
||||
filter: applyBlacklist,
|
||||
|
||||
Reference in New Issue
Block a user