Python 3 made huge changes, especially to string handling and input/output which is what most applications do all the time.
And perl 6 is practically a completely different language. I wouldn't count that.
What is interesting about php7 is how much effort and how little complaining goes into porting the C extensions: that API has changed a lot and the documentation is still spread all over wikis and blog posts. And yet there is a php7 port of many third party extensions already
But I guess the huge performance improvements are a strong motivator even there
too bad for py3's fundamental problem: that it broke the interface so badly, people still don't use it. PHP7 has no such problem, and gets to add boatloads of new features (which will be immediately used/adopted) while deprecating old versions of the language.
Php7 did nothing like that, that's why the "porting" process was so light. But it also means it didn't fix any of the big things.
This is nonsense, PHP 7 contains a long list of backwards-compatibility breaks:
http://php.net/manual/en/migration70.php
If PHP 7.0 didn't break anything, it would have been called 5.7.
The upgrade is easy simply because these breaks do not have as much of an impact as Python's did.
http://php.net/manual/en/language.types.type-juggling.php
How would you fix this without breaking most PHP code out there?
How would you fix ==, > and < behaviour without breaking most PHP code out there?
How would you fix Unicode without breaking most PHP code out there?
If PHP fixed things like this, it would be a decent programming language, but it would also break a lot of things. Kinda of like what python3 did, only that it didn't have to break quite so much, since python2 was already pretty decent.