AIOHTTP: Event loop closed

[LordFckHelmchen]

Hey @layday I recently updated to 3.0 and it appears the error mentioned in issues #31 is there again:

My Setup:	Name	Version
Windows	Version 10.0.19045.3155
Python	3.9.5 (tags/v3.9.5:0a7dcbd, May 3 2021, 17:27:52) [MSC v.1928 64 bit (AMD64)] on win32
Installed packages (pip freeze)	aiohttp==3.8.4 aiohttp-client-cache==0.8.1 aiohttp-rpc==0.7.1 aiosignal==1.2.0 aiosqlite==0.17.0 alembic==1.11.1 appdirs==1.4.4 async-timeout==4.0.2 attrs==23.1.0 cattrs==23.1.2 cffi==1.15.1 chardet==4.0.0 charset-normalizer==2.1.1 click==8.1.4 clr-loader==0.2.4 colorama==0.4.4 colosseum==0.2.0 exceptiongroup==1.0.4 frozenlist==1.3.1 greenlet==1.1.0 idna==3.1 instawow==3.0.0 iso8601==1.0.2 itsdangerous==2.1.2 jarowinkler==1.2.1 jedi==0.18.2 jellyfish==0.8.2 Jinja2==3.0.3 loguru==0.7.0 Mako==1.2.4 MarkupSafe==2.0.1 multidict==5.1.0 packaging==23.1 parso==0.8.3 pluggy==0.13.1 prompt-toolkit==3.0.31 ptpython==3.0.21 pycparser==2.21 pydantic==1.9.0 Pygments==2.13.0 python-dateutil==2.8.1 python-editor==1.0.4 python-forge==18.6.0 pythonnet==3.0.0.post1 questionary==1.10.0 rapidfuzz==2.13.6 six==1.16.0 SQLAlchemy==2.0.18 toga-core==0.2.15 typing_extensions==4.3.0 url-normalize==1.4.3 wcwidth==0.2.5 win32-setctime==1.0.3 yarl==1.8.1
Git Bash	Git for Windows 2.41.0.windows.1 (64-bit)

I updated aiohttp to the most recent version (for py3.9), but calling instawow update still causes the error (from cmd & Git Bash):

❯ instawow --debug update
2023-07-15 11:26:35.030 | INFO     | instawow.wa_updater:extract_installed_auras:171
  loading C:\Program Files (x86)\World of Warcraft\_classic_\WTF\Account\<MY_ACC>\SavedVariables\WeakAuras.lua from cache at C:\Users\<MY_USER>\AppData\Local\Temp\instawow\cache\103d1d1b8368cf55e16a8d6204a1b921
2023-07-15 11:26:35.038 | INFO     | instawow.wa_updater:extract_installed_auras:171
  loading C:\Program Files (x86)\World of Warcraft\_classic_\WTF\Account\<MY_ACC>\SavedVariables\Plater.lua from cache at C:\Users\<MY_USER>\AppData\Local\Temp\instawow\cache\a1027651d1452fba2a15095ab867c210
✗ tukui:13-benikui-wotlk
  package does not exist
Exception ignored in: <function _ProactorBasePipeTransport.__del__ at 0x00000185D88434C0>
Traceback (most recent call last):
  File "C:\Program Files\Python39\lib\asyncio\proactor_events.py", line 116, in __del__
    self.close()
  File "C:\Program Files\Python39\lib\asyncio\proactor_events.py", line 108, in close
    self._loop.call_soon(self._call_connection_lost, None)
  File "C:\Program Files\Python39\lib\asyncio\base_events.py", line 746, in call_soon
    self._check_closed()
  File "C:\Program Files\Python39\lib\asyncio\base_events.py", line 510, in _check_closed
    raise RuntimeError('Event loop is closed')
RuntimeError: Event loop is closed

Any thoughts?

[layday]

Thanks for the detailed report! I think I might've been a little too eager to remove the workaround; I could swear the issue was fixed in Python 3.9, but it looks like it was only backported as far as 3.10 (https://github.com/python/cpython/issues/92841).

[layday]

Interestingly this error does not appear in CI: https://github.com/layday/instawow/actions/runs/5562045095/jobs/10160107275.

[layday]

Should be fixed in v3.0.1.

[LordFckHelmchen]

Hey @layday thanks for the quick action! Regarding the missing CLI - could that be because Pytest swallows your cli output and only shows it if your test fails? But since you handle it gracefully, your test didn't fail, right?

[layday]

On closer inspection I think it's cuz we're reusing the event loop for CLI tests (we are not spawning a new Python process) and this error is only emitted when the loop is shutting down.