J. Nick Koston

I'm going to try to work on a fully async aiodns+zeroconf based resolver this weekend if I don't run out of time.

I've create an initial version of https://github.com/aio-libs/aiohttp-asyncmdnsresolver I still have some more docs to write for it but the basic usage is https://github.com/aio-libs/aiohttp-asyncmdnsresolver/blob/main/examples/run.py

https://github.com/aio-libs/aiohttp-asyncmdnsresolver 0.0.1 is available with docs and example

@PeteRager would you please loop me into the email chain? I’ll take a look when I get back from holiday next week. [email protected] thanks

Merged in dev since this PR is a bit behind to make sure merging it won't cause any lint issues

Maybe register it here as well https://www.iana.org/form/ports-services

Currently we only normalize the path if the host (netloc) is set If you add a host, then the assertion will pass ``` assert yarl.URL("web+demo://host/path/..//not-a-host/") == yarl.URL("web+demo://host/.//not-a-host/") ```

It seems this behavior is quite intentional as if I remove the check these tests fail: ``` FAILED tests/test_url.py::test_join_from_rfc_3986_normal[../..-http://a/] - AssertionError: assert URL('http://a/b/c/d;p?q') == URL('http://a/') FAILED tests/test_url.py::test_join_from_rfc_3986_normal[.-http://a/b/c/] - AssertionError: assert...

> Possibly, we should be normalising the paths of relative URLs on equality check though? I _think_ that would be ok but not sure if thats a breaking change or...

There doesn't seem to be any interesting changes that might cause this https://github.com/aio-libs/yarl/compare/v1.12.0...v1.13.1 I wonder if a build dep has changed