Hi, 

thanks for reporting. I've tested and everything works fine.

Basically, you have a typo there.

You should enter 111.252.39.131 instead of 111.252.039.131.

Well http://111.252.039.131/ with the zero works just fine on chromium for Android 7. So are you saying that one must remove the 0 for desktop Linux chromium but the zero is fine for Android chromium?

I can confirm that works on Android Chrome stable, but not on Windows.  Chrome doesn't recognize "111.252.039.131" as a valid IP address, since it isn't, so it sends it to the system's DNS resolver.  Android's DNS resolver returns the IP address 111.252.39.131, while Window's DNS resolver indicates there's no such DNS entry.  Seems like an Android bug to me, it should fail on both platforms.

I think this is probably just a WontFix.

So should I send bug reports to
Firefox
w3m
midori
lynx
etc. all other browsers I tested here today on Debian/Linux,
requesting that they should "break this too"?

So all the world, and half of chromium, is doing it wrong.

Wouldn't it just be better to make all of chromium just go with the flow?

All I know is the user can
$ ping
such addresses, so would it be so bad to go along with 99% of the other browsers,,,
How do the big browsers on the big OSs deal with it?

And... if you are printing helpful messages, just tell the user to remove the zeros. The current messages send the user on a wild goose chase...

Just tried it on FireFox and IE on Windows, and it doesn't work.  It's clearly a platform things - looks like on many POSIX systems, the platform host resolver maps "111.252.039.131" to "111.252.39.131", and other browsers are just sending things that aren't IP addresses to it.  On Linux, we don't use the system resolver, so don't inherit its quirks.  One of the cases where it's actually Microsoft that's doing the standards-compliant thing.

Anyway even if it is supposed to be octal...
I think you should file bugs against all those other browsers if you think they are really incorrect.

http://superuser.com/questions/857603/are-ip-addresses-with-and-without-leading-zeroes-the-same

OK. Then apparently it is really bad. So it would be better if chromium (just like with removing Flash) stopped supporting it on all platforms. That way I wouldn't pass around URLs that would fail elsewhere.

Ah, thanks - indeed you're right, there is no standard here.

The cross-browser attempt at the spec is https://url.spec.whatwg.org/ , which makes it clear that this would be prohibited.

Going to mark this as WontFix/WorkingAsIntended (as it's a known interaction with OS layers and resolver libraries)