Nulo/preciazo
1
0
Fork 0
mirror of https://github.com/catdevnull/preciazo.git synced 2024-11-26 03:26:19 +00:00
Code Issues Projects Releases Packages Wiki Activity

carrefour: solo intentar parsear jsonld en error

Browse source
This commit is contained in:
Cat /dev/Nulo 2024-01-01 16:49:43 -03:00
parent eb271fb496
commit 518a7cb2f5
1 changed files with 14 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

20
scraper/parsers/carrefour.ts
View file

@ -1,6 +1,6 @@
import { parseHTML } from "linkedom"; import { parseHTML } from "linkedom";
import { Precioish, type Precio } from "../scrap.js"; import { Precioish, type Precio } from "../scrap.js";
import { getProductJsonLd, priceFromMeta } from "../common.js"; import { getProductJsonLd, priceFromMeta, stockFromMeta } from "../common.js";
function getEanByTable(dom: Window): string { function getEanByTable(dom: Window): string {
const eanLabelEl = dom.window.document.querySelector( const eanLabelEl = dom.window.document.querySelector(
@ -54,14 +54,22 @@ export function getCarrefourProduct(html: string | Buffer): Precioish {
const dom = parseHTML(html); const dom = parseHTML(html);
const precioCentavos = priceFromMeta(dom); const precioCentavos = priceFromMeta(dom);
const inStock = stockFromMeta(dom);
const ean = eanFromSeedState(dom); const ean = eanFromSeedState(dom);
const ld = getProductJsonLd(dom); let name, imageUrl;
const name = ld.name; try {
const imageUrl = ld.image; const ld = getProductJsonLd(dom);
const inStock = name = ld.name;
ld.offers.offers[0].availability === "http://schema.org/InStock"; imageUrl = ld.image;
} catch (error) {
if (inStock) {
throw error;
} else {
// algunas paginas sin stock no tienen json ld
}
}
return { return {
name, name,