[issue19692] Rename Py_SAFE_DOWNCAST

hai shi added the comment: Looks we have a fast version rhythm. IMHO, If user would be affected and reported, reversing the PR is ok. -- ___ Python tracker ___

[issue19692] Rename Py_SAFE_DOWNCAST

Change by Sanyam Khurana : -- nosy: +CuriousLearner versions: +Python 3.7, Python 3.8, Python 3.9 -Python 3.4 ___ Python tracker ___

[issue19692] Rename Py_SAFE_DOWNCAST

hai shi added the comment: It's a big problem, no ability to answer this question :( -- ___ Python tracker ___ ___

[issue19692] Rename Py_SAFE_DOWNCAST

Inada Naoki added the comment: While Py_SAFE_DOWNCAST is not documented, it doesn't start with underscore. How many 3rd party code are broken by changing the name? -- nosy: +inada.naoki ___ Python tracker

[issue19692] Rename Py_SAFE_DOWNCAST

STINNER Victor added the comment: I dislike this macro. A cast is either safe or unsafe. If it is safe, (type)var would be better. If it is unsafe, well, it would be better to add a runtime check. No? (I mean better error reporting than abort() pnly in debug mode.) --

[issue19692] Rename Py_SAFE_DOWNCAST

Stefan Krah added the comment: Why would one not abort() in release mode? If the cast is inexact, the results will usually be so wrong that even on a web server a hard exit is preferable. I don't think the check costs much time with branch prediction. -- nosy: +skrah

[issue19692] Rename Py_SAFE_DOWNCAST

hai shi added the comment: Rename Py_SAFE_DOWNCAST in PR_15090. In the C API: Py_TRASHCAN_SAFE_BEGIN and Py_TRASHCAN_SAFE_END should be removed or keep it due to compatibility? In the stdlib: Looks that it's not changed is ok. -- nosy: +shihai1991

[issue19692] Rename Py_SAFE_DOWNCAST

Change by hai shi : -- keywords: +patch pull_requests: +14835 stage: -> patch review pull_request: https://github.com/python/cpython/pull/15090 ___ Python tracker ___

[issue19692] Rename Py_SAFE_DOWNCAST

New submission from Antoine Pitrou: Py_SAFE_DOWNCAST's name is a bit misleading: it isn't safe except in debug mode. I propose to rename it to Py_DOWNCAST, so that developers are reminded that the burden of the sanity checks is on them. -- components: Interpreter Core messages: 203764

[issue19692] Rename Py_SAFE_DOWNCAST

Christian Heimes added the comment: +1 -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue19692 ___ ___ Python-bugs-list mailing list Unsubscribe:

[issue19692] Rename Py_SAFE_DOWNCAST

STINNER Victor added the comment: I like Py_DOWNCAST name. -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue19692 ___ ___ Python-bugs-list mailing

[issue19692] Rename Py_SAFE_DOWNCAST

Antoine Pitrou added the comment: Actually, _Py_DOWNCAST may be better (not a public API). -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue19692 ___

[issue19692] Rename Py_SAFE_DOWNCAST

Martin v. Löwis added the comment: -1. The macro name doesn't claim the cast to be safe, i.e. it's not Py_SAFELY_DOWNCAST, but safe downcast, i.e. it's an assertion that the cast actually has been verified as being safe. -- nosy: +loewis ___ Python

[issue19692] Rename Py_SAFE_DOWNCAST

Antoine Pitrou added the comment: -1. The macro name doesn't claim the cast to be safe, i.e. it's not Py_SAFELY_DOWNCAST, but safe downcast, i.e. it's an assertion that the cast actually has been verified as being safe. It's not an assertion, it's a cast. Otherwise it should be named

[issue19692] Rename Py_SAFE_DOWNCAST

Tim Peters added the comment: Goodness. Name it _Py_DOWNCAST_AND_IN_DEBUG_MODE_ASSERT_UPCASTING_THE_RESULT_COMPARES_EQUAL_TO_THE_ORIGINAL_ARGUMENT ;-) -- ___ Python tracker rep...@bugs.python.org http://bugs.python.org/issue19692

[issue19692] Rename Py_SAFE_DOWNCAST

Antoine Pitrou added the comment: Well, I obviously won't fight very hard for this one. But I would like to point out that APIs with safe (not safely :-)) in their name usually imply that the API is safe, not that the input has been sanitized beforehand. For example in the stdlib: