Greedy router

[yourball]

Fixes issue #3056. Fixes hanging greedy router on some circuit instances by partially reverting changes from 6d81ddb.

[googlebot]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here with @googlebot I signed it! and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

ℹ️ Googlers: Go here for more info.

[yourball]

@googlebot I signed it!

[googlebot]

CLAs look good, thanks!

ℹ️ Googlers: Go here for more info.

[balopat]

Thank you for taking the time digging this out!! The fix looks good to me, as it is covered by tests. However, it would be great to put a test around it so that we can prevent regression (like what we have now).

The test would run a minimal instance where we can prove that the algorithm would run on infinite loops without this fix, but fast (<1s) with the fix. The algorithm would run in a separate process that we should measure if it finishes within the expected time. Can you find a graph instance like that?

[balopat]

LGTM - One more nit
Thanks for adding the test, I just tried it out and it works nicely! :)
We should make the timeout 5 seconds to avoid flakes, on my Mac it runs just above 2 seconds.

[balopat]

I would do something like this instead - this takes care of the coverage issue as well

    try:
        assert not process.is_alive(), "Greedy router timeout"
    finally:
        process.terminate()

[yourball]

Cool, thanks! Was happy to contribute. It looks like all tests passed now.:)