RuntimeError: Event loop is closed

[JacobLee23]

After running pip install playwright and playwright install, I went to the documentation to try some of the features out. Here is the code which I ran:

from playwright.sync_api import sync_playwright
with sync_playwright() as p:
    browser = p.chromium.launch()
page = browser.new_page()

However, I received an error: RuntimeError: Event loop is closed.

I proceeded to try the same process by with browser=p.firefox.launch(), but to no avail.

Specifications:

• Python 3.9.2
• Playwright v1.9.1

[kumaraditya303]

You are creating new page outside of the playwright context manager, so it erroring out

from playwright.sync_api import sync_playwright
with sync_playwright() as p:
  browser = p.chromium.launch()
  page = browser.new_page()

[JacobLee23]

So is there a way to use page and browser outside a with statement? I tried this:

p = sync_playwright()
browser = p.chromium.launch()

but it does not work either. I have a feeling there is something discussing this later in the documentation, but it would be much appreciated if it were to be pointed out now.

[kumaraditya303]

Yes, if you want to use without with then you need to manually call start and stop @JLpython-py

from playwright.sync_api import sync_playwright

p = sync_playwright().start()
browser = p.chromium.launch(
    executable_path=r"C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe",
    headless=False,
)
page = browser.new_page()
page.goto("http://www.google.com/")
browser.close()
p.stop()

[JacobLee23]

Okay, that clarifies quit a bit. Thank you so much!

[kumaraditya303]

@pavelfeldman May be discussions for this repo should be enabled for these type of question

[JacobLee23]

Yes, if you want to use without with then you need to manually call start and stop @JLpython-py

from playwright.sync_api import sync_playwright

p = sync_playwright().start()
browser = p.chromium.launch(
    executable_path=r"C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe",
    headless=False,
)
page = browser.new_page()
page.goto("http://www.google.com/")
browser.close()
p.stop()

It took me a while, but I was finally able to implement this into my scripts and I tested it out. However, not I get a warning message:

(node:4464) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a pr
omise which was not handled with .catch(). To terminate the node process on unhandled promise rejection, use the CLI flag `--unhandled-rejections=strict` (see https://nodejs.org/api/cli.h
tml#cli_unhandled_rejections_mode). (rejection id: 2)

Do you know what this means and how I can fix it?

[pavelfeldman]

What is your edge browser version?

[JacobLee23]

What is your edge browser version?

Version 89.0.774.57

[pavelfeldman]

Sorry for the late reply. Your snippet looks correct, but we need more information to address this. Could you run your snippet with the DEBUG=pw:browser* environment variable set? That would help us identifying the issue.

[JacobLee23]

I think I was able to solve it, assuming my fix is relevant to the warning. I spent a while debugging and one of my tests didn't call browser.close() before calling p.stop() (from the above comment). I haven't seen the warning message much after fixing this, so I think this may be the solution to my issue.