Вариант:
'use strict';
const puppeteer = require('puppeteer');
(async function main() {
try {
const browser = await puppeteer.launch();
const [page] = await browser.pages();
const html = `
<!doctype html>
<html>
<head><meta charset='UTF-8'><title>Test</title></head>
<body>
<table><tbody>
<tr>
<td data-name="Name"><a href="https://test.ca">test</a></td>
<td data-name="Active Ingredients"><a href="http://test1.ca">test13</a></td>
<td data-name="Strength"><a href="http://test3.ca">20mg</a></td>
<td data-name="Picture"><a href=""><img src=""></a>
</tbody></table>
</body>
</html>`;
await page.goto(`data:text/html,${html}`);
const data = await page.evaluate(() => Array.from(
document.querySelectorAll('td'),
td => ({ [td.dataset.name]: td.querySelector('a[href]').href }),
));
console.log(data);
await browser.close();
} catch (err) {
console.error(err);
}
})();
Вывод:
[
{ Name: 'https://test.ca/' },
{ 'Active Ingredients': 'http://test1.ca/' },
{ Strength: 'http://test3.ca/' },
{ Picture: '' }
]