gh-146268: Fix case where assigning an infinite generator to an extend slice hangs instead of raising ValueError

[csm10495]

gh-146268: Fix case where assigning an infinite generator to an extend slice hangs instead of raising ValueError

Also add in some unit tests and a blurb entry.

step != 1 is special since we will only perform that assignment if the size of the right matches that of the left. An infinitely sized right hand side can never fill in a (non infinite) left hand side. So we calculate out the left hand side's slice length, then use that to determine if we went over (+1 the left hand's length).

[python-cla-bot]

All commit authors signed the Contributor License Agreement.

[johnslavik]

This isn't ready yet.

❯ py t.py
Traceback (most recent call last):
  File "/Users/bartosz.slawecki/Python/cpython/t.py", line 10, in <module>
    l[::2] = MyGen()
    ~^^^^^
ValueError: attempt to assign sequence of size 3 to extended slice of size 2

The error message from this approach leaks internal implementation details. MyGen() (from the PR) is not a sequence of size 3, it's a generator.

[csm10495]

@johnslavik updated based off your feedback. Please check again.

[picnixz]

How does bytearray and array.array handle those cases? likewise what about assigning with itself? does it change the current behavior? a[::2] = a should not be behave differently or if so we should test it

[picnixz]

Don't we have an FastSequence check for that? (fast sequences are lists or tuples). Also we should not hit this path if we have a known length.

[csm10495]

I don't think we have a way to check if it is a fast seq vs trying to create a new one. Agreed on the second part.

[bedevere-app]

A Python core developer has requested some changes be made to your pull request before we can consider merging it. If you could please address their requests along with any other requests in other reviews from core developers that would be appreciated.

Once you have made the requested changes, please leave a comment on this pull request containing the phrase I have made the requested changes; please review again. I will then notify any core developers who have left a review that you're ready for them to take another look at this pull request.

[csm10495]

How does bytearray and array.array handle those cases? likewise what about assigning with itself? does it change the current behavior? a[::2] = a should not be behave differently or if so we should test it

For me on 3.14.3 and on the latest head of this branch this scenario works the same:

In [1]: a = list(range(10))

In [2]: a
Out[2]: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

In [3]: a[::2] = a
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[3], line 1
----> 1 a[::2] = a

ValueError: attempt to assign sequence of size 10 to extended slice of size 5

In [4]:

tbd on bytearray and arrray.array

[csm10495]

@picnixz I think i addressed your points. Thanks!