Thanks for finding this. There does seem to be a regression here checking against M62 stable. I'm guessing it might be because https://chromium.googlesource.com/chromium/src/+/2cd55a1304342c5d9a1de8961b24fa8b1817c0bd stopped using a pair of (origin, origin) when the embedder was null (which was the original behaviour; it was changed to do this to fix a NullPointerException crash)

Given how many other things are crazy and weird in Android All Sites, I'm probably going to let this sit until we get the go ahead to properly redo the entire thing.

Hmm, it's a bit unfortunate to regress here because this list is already bad and I'm a bit sad to make things worse. Is there anything we can do as a short term fix? 

I'd prefer not to make too many short term hacks here since the general agreement is that Android All Sites needs a complete revamping. Plus, if the theory in c#1 is correct, this is the actual intended original behaviour (!).

+emilyschechter for thoughts on whether this is worth addressing now or later

q888888888888@gmail.com

This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue.

Sorry for the inconvenience if the bug really should have been left as Available.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot

Assigning to Emily for thoughts, this seems like the opposite of crbug.com/909619 so it might no longer be a thing?