Delaying 3.12 beta 2 by a week

[thomas (Thomas Wouters)]

Because there are a number of important bugs being worked on that impede the testing of 3.12, it would be a shame to delay all that by a month, and it’s only been a week since beta 1, I’m delaying beta 2 until next week. (In retrospect I should have realised a week is not enough for third parties to find bugs and us to fix them, so lesson learned).

For reference, some of the bugs I’m talking about:

• Type object's ob_type does not get set when tp_bases is set before PyType_Ready · Issue #104614 · python/cpython · GitHub
• `tp_bases` of `object` is `NULL`: undocumented or unintentional behavior change in 3.12? · Issue #105020 · python/cpython · GitHub
• 3.12.0b1 includes backwards incompatible change to operation of `super()` · Issue #105035 · python/cpython · GitHub
• object.h includes C++ incompatible code · Issue #105059 · python/cpython · GitHub

[guido (Guido van Rossum)]

(Could you turn those references into a bulleted list? It looks like spaghetti. :-))

[thomas (Thomas Wouters)]

(Sure, done.)

[eric.snow]

@thomas

For reference, some of the bugs I’m talking about:

• Type object’s ob_type does not get set when tp_bases is set before PyType_Ready · Issue #104614 · python/cpython · GitHub

I’m looking into this.

@thomas

• tp_bases of object is NULL: undocumented or unintentional behavior change in 3.12? · Issue #105020 · python/cpython · GitHub

I have a fix up: https://github.com/python/cpython/pull/105115.

@thomas

• object.h includes C++ incompatible code · Issue #105059 · python/cpython · GitHub

While this is a result of the immortal objects work, I wasn’t very involved in the implementation, so I’ll defer to @eelizondo.

---

FYI, I also finally have a good fix up for gh-104341 (sporadic buildbot/CI failures): https://github.com/python/cpython/pull/105109.

[carljm (Carl Meyer)]

@thomas

3.12.0b1 includes backwards incompatible change to operation of super() · Issue #105035 · python/cpython · GitHub

This one is now fixed.