scraper: reordenar codigo

- borrar código viejo
- centralizar scrapers de links

carrefour-link-scraper/package.json

@@ -1,17 +0,0 @@
-{
-  "name": "carrefour-link-scraper",
-  "type": "module",
-  "version": "1.0.0",
-  "description": "",
-  "main": "index.js",
-  "scripts": {
-    "test": "echo \"Error: no test specified\" && exit 1"
-  },
-  "keywords": [],
-  "author": "",
-  "license": "ISC",
-  "dependencies": {
-    "linkedom": "^0.16.5",
-    "p-map": "^7.0.1"
-  }
-}

dia-link-scraper/package.json

@@ -1,17 +0,0 @@
-{
-  "name": "dia-link-scraper",
-  "type": "module",
-  "version": "1.0.0",
-  "description": "",
-  "main": "index.js",
-  "scripts": {
-    "test": "echo \"Error: no test specified\" && exit 1"
-  },
-  "keywords": [],
-  "author": "",
-  "license": "ISC",
-  "dependencies": {
-    "linkedom": "^0.16.5",
-    "p-map": "^7.0.0"
-  }
-}

link-scrapers/coto.ts

@@ -1,4 +1,3 @@
-import { getHtml } from "../scraper/fetch.js";
 import { parseHTML } from "linkedom";
 import PQueue from "p-queue";
 import { saveUrls } from "db-datos/urlHelpers.js";
@@ -28,12 +27,13 @@ function getPage(url: string) {
   return async () => {
     let html;
     try {
-      html = await getHtml(url);
+      const res = await fetch(url);
+      html = await res.text();
     } catch (error) {
       await getPage(url)();
       return;
     }
-    const { document } = parseHTML(html.toString("utf-8"));
+    const { document } = parseHTML(html);
 
     const hrefs = Array.from(
       document.querySelectorAll<HTMLAnchorElement>(".product_info_container a"),

link-scrapers/dia.ts

@@ -1,7 +1,6 @@
 import pMap from "p-map";
 import { decodeXML } from "entities";
 import { parseHTML } from "linkedom";
-import { getHtml } from "../scraper/fetch.js";
 import { saveUrls } from "db-datos/urlHelpers.js";
 
 const categorias = [
@@ -111,8 +110,9 @@ async function scrapBySite() {
   await pMap(
     links,
     async (url) => {
-      const html = await getHtml(url);
-      const { document } = parseHTML(html.toString("utf-8"));
+      const res = await fetch(url);
+      const html = await res.text();
+      const { document } = parseHTML(html);
 
       const hrefs = Array.from(
         document.querySelectorAll<HTMLAnchorElement>(

link-scrapers/package.json

@@ -1,5 +1,5 @@
 {
-  "name": "coto-link-scraper",
+  "name": "link-scrapers",
   "type": "module",
   "version": "1.0.0",
   "description": "",

package.json

@@ -2,9 +2,7 @@
   "name": "preciazo",
   "private": true,
   "workspaces": [
-    "dia-link-scraper",
-    "coto-link-scraper",
-    "carrefour-link-scraper",
+    "link-scrapers",
     "scraper",
     "sitio",
     "db-datos"

readme.md

@@ -4,7 +4,7 @@ scrapeo "masivo" de precios y datos en supermercados argentinos
 
 ## componentes (en orden de proceso)
 
-- los link scrapers ([coto-link-scraper](./coto-link-scraper/), [dia-link-scraper](./dia-link-scraper/) y [carrefour-link-scraper](./carrefour-link-scraper)) crean listas de links a productos para scrapear
+- los link scrapers ([link-scrapers/](./link-scrapers/)) crean listas de links a productos para scrapear
 
   (no hace falta correrlos porque ya hay listas armadas en [data/](./data/))
 

scraper/auto.ts

@@ -8,9 +8,9 @@ import { downloadList } from "./scrap.js";
 import { db } from "db-datos/db.js";
 import { like } from "drizzle-orm";
 import { productoUrls } from "db-datos/schema.js";
-import { scrapDiaProducts } from "../dia-link-scraper/index.js";
-import { scrapCotoProducts } from "../coto-link-scraper/index.js";
-import { scrapCarrefourProducts } from "../carrefour-link-scraper/index.js";
+import { scrapDiaProducts } from "../link-scrapers/dia.js";
+import { scrapCotoProducts } from "../link-scrapers/coto.js";
+import { scrapCarrefourProducts } from "../link-scrapers/carrefour.js";
 
 const supermercados: Supermercado[] = [
   Supermercado.Carrefour,

scraper/cli.ts

@@ -1,6 +1,6 @@
-import { scrapCarrefourProducts } from "../carrefour-link-scraper/index.js";
-import { scrapCotoProducts } from "../coto-link-scraper/index.js";
-import { scrapDiaProducts } from "../dia-link-scraper/index.js";
+import { scrapCarrefourProducts } from "../link-scrapers/carrefour.js";
+import { scrapCotoProducts } from "../link-scrapers/coto.js";
+import { scrapDiaProducts } from "../link-scrapers/dia.js";
 import { auto } from "./auto.js";
 import { downloadList, getProduct } from "./scrap.js";
 

scraper/fetch.ts

@@ -1,13 +0,0 @@
-export async function getHtml(url: string) {
-  const res = await fetch(url);
-  return readableToBuffer(res.body!);
-}
-
-async function readableToBuffer(source: AsyncIterable<any>) {
-  // https://stackoverflow.com/a/72891118
-  const buffers = [];
-  for await (const data of source) {
-    buffers.push(data);
-  }
-  return Buffer.concat(buffers);
-}

scraper/parsers/carrefour.ts

@@ -1,6 +1,6 @@
 import { parseHTML } from "linkedom";
 import { Precioish } from "../scrap.js";
-import { getProductJsonLd, priceFromMeta, stockFromMeta } from "../common.js";
+import { getProductJsonLd, priceFromMeta, stockFromMeta } from "./common.js";
 
 function parseScriptJson<T>(dom: Window, varname: string): T {
   const script = dom.window.document.querySelector<HTMLTemplateElement>(

scraper/parsers/dia.ts

@@ -1,6 +1,6 @@
 import { parseHTML } from "linkedom";
 import { type Precioish } from "../scrap.js";
-import { getMetaProp, getProductJsonLd, priceFromMeta } from "../common.js";
+import { getMetaProp, getProductJsonLd, priceFromMeta } from "./common.js";
 
 export function getDiaProduct(html: string | Buffer): Precioish {
   const dom = parseHTML(html);