[Question] Download with headless: true works fine , but with headless: false not; plw 1.8.0

[osmenia]

Hi Team,

plw 1.8.0
with headless: false code works fine, but with headless: true only first element is downloaded.

plw 1.3.0
works fine with both params (headless: false & headless: true)

What could be a problem with 1.8.0?

tnx a lot

const playwright = require("playwright");

(async () => {
  const browser = await playwright.chromium.launch({ headless: false });
  const context = await browser.newContext({
    acceptDownloads: true,
  });
  const page = await context.newPage();
  await page.goto(
    "https://file-examples.com/index.php/sample-documents-download/sample-xls-download/"
  );

  const handles = await page.$$("#table-files .file-link");

  for (let num of handles) {
    const [download] = await Promise.all([
      page.waitForEvent("download"), // wait for download to start
      (await num.$("a.btn")).click(),
    ]);
    const path = await download.path();
    console.log(path);
  }
  await browser.close();
})();

[pavelfeldman]

It works for me, downloads a bunch of files:

/tmp/playwright_downloads-FjfvTw/4ea77094-fdc6-4c34-84a1-fbb03511b9e6
/tmp/playwright_downloads-FjfvTw/2ad79520-533d-4e0c-bb20-24713b007e47
/tmp/playwright_downloads-FjfvTw/530c8eaa-205b-4562-bc76-d264664a3656
/tmp/playwright_downloads-FjfvTw/f0789fc3-2a20-4f28-8a78-54558f7eff61
/tmp/playwright_downloads-FjfvTw/e820f9e7-369a-4dc5-bdad-95e38fee549a
/tmp/playwright_downloads-FjfvTw/72384378-45bc-4f68-9da3-3656793f3cfa
/tmp/playwright_downloads-FjfvTw/d4b138c4-ac42-4e70-8634-013a76f76035
/tmp/playwright_downloads-FjfvTw/a616a379-bdee-42c3-ac9b-56a8972c74bd
/tmp/playwright_downloads-FjfvTw/1f24ac29-e774-4b5c-bf8e-7cfcb0f809c1
/tmp/playwright_downloads-FjfvTw/9b4585c6-abca-48f5-9676-708e99b1d0c7

What is you OS?

[osmenia]

Windows 10 and Ubuntu

[pavelfeldman]

Oh, you said

with headless: false code works fine, but with headless: true only first element is downloaded.
but it looks the opposite.

Either way, what happens is that once you click on the first link, the page navigates to a grammarly ad. So you can't really click on any other links - they no longer exist, you are on a different page, there is nothing to click.

It looks like the site is using some heuristic to either show or not show the ad for you. Or maybe headless is just much faster, so it gets a chance to click all the links before the page navigates.

[pavelfeldman]

Closing this issue as it isn't a bug, you should probably re-navigate to the original page after each download and click i-th link in the loop.