import { Buffer } from "node:buffer";
import jsdom from "npm:jsdom";
const JSDOM = jsdom.JSDOM;
export default async function getMetadata(url) {
if (url == null || url.trim() == "") {
return { "status": 400, "error_msg": "Valid URL is required as input. URL was empty." };
}
let resp = await fetch(url);
let text = await resp.text();
let frag = new JSDOM(text).window.document;
let imgUrl = getImageUrl(frag, url);
let imgData = await getImageDataFromUrl(imgUrl);
let title = getTitle(frag);
let description = getDescription(frag);
return { "status": 200, "url": url, title, description, imgUrl, imgData };
}
function getImageUrl(frag, url) {
let imgUrl = "";
let selectors = [
"meta[property=\"og:image:secure_url\"]",
"meta[property=\"og:image:url\"]",
"meta[property=\"og:image\"]",
"meta[name=\"twitter:image:src\"]",
"meta[property=\"twitter:image:src\"]",
"meta[name=\"twitter:image\"]",
"meta[property=\"twitter:image\"]",
"meta[itemprop=\"image\"]",
];
let element;
for (let i = 0; i < selectors.length; i++) {
element = frag.querySelector(selectors[i]);
if (!imgUrl && element && element.content) {
imgUrl = element.content;
}
}
element = frag.querySelector("img[alt*=\"author\" i]");
if (!imgUrl && element && element.src) {
imgUrl = element.getAttribute("src");
}
element = frag.querySelector("img[src]:not([aria-hidden=\"true\"])");
if (!imgUrl && element && element.src) {
imgUrl = element.getAttribute("src");
}
if (imgUrl !== "") {
imgUrl = new URL(imgUrl, url).href;
}
return imgUrl;
}
async function getImageDataFromUrl(url) {
const response = await fetch(url);
const base64data = Buffer.from(await response.arrayBuffer()).toString("base64");
return "data:image/png;base64," + base64data;
}
function getTitle(frag) {
let element;
let title = "";
let selectors = [
"meta[property=\"og:title\"]",
"meta[name=\"twitter:title\"]",
"meta[property=\"twitter:title\"]",
];
for (let i = 0; i < selectors.length; i++) {
element = frag.querySelector(selectors[i]);
if (element && element.content) {
title = element.content;
return title;
}
}
element = frag.querySelector("title");
if (element && (element.innerText || element.text || element.textContent)) {
title = element.innerText || element.text || element.textContent;
return title;
}
return title;
}