Add `PyTuple_FromPair`
@sergey-miryanov 已經在處理了。
開始於 2026年2月26日。
- 主要語言
- Python
- 星號
- 77.2k
- 分支
- 35.9k
- PR 合併指標
- PR 指標待擷取
描述
Feature or enhancement
Proposal:
Rationale
We have had quite a few PRs lately suggesting replacing PyTuple_New (which requires manually setting the items) with PyTuple_FromArray.
Many of them have been rejected because of (maybe excessive) requirements for the authors to do benchmarking, even though the code is IMO a clear improvement. Almost all of these have been for pairs.
I propose adding PyTuple_FromPair to provide an idiomatic, fast and safe way to create 2-tuples.
We should also add a "steal" variant as well to simplify replacing PyTuple_New() ... PyTuple_SET_ITEM.
Note: I originally wanted to call this
PyTuple_MakePair, but I'm takingPyTuple_FromPairfrom https://github.com/capi-workgroup/decisions/issues/84#issuecomment-3529843707
Here's all the relevant PRs I could easily find:
https://github.com/python/cpython/pull/144529
https://github.com/python/cpython/pull/144531
https://github.com/python/cpython/pull/144532
https://github.com/python/cpython/pull/144760
https://github.com/python/cpython/pull/144771
https://github.com/python/cpython/pull/144772
https://github.com/python/cpython/pull/144773
https://github.com/python/cpython/pull/144829
Prior abandoned issue:
https://github.com/python/cpython/issues/140052 which also proposes adding PyTuple_FromSingle and which links to https://github.com/capi-workgroup/decisions/issues/84.
https://github.com/capi-workgroup/decisions/issues/84 rejects the idea without any clear rationale.
Clarity
t = PyTuple_New(2);
// error handling here
// Care must be taken not to leak the invalid tuple
PyTuple_SET_ITEM(t, a);
PyTuple_SET_ITEM(t, b);
compare with
t = PyTuple_FromPair(a, b);
// error handling here
Performance
Compared to PyTuple_New
PyTuple_FromPair can allocate directly from the free list without additional checks. It does not need to check for weird sizes, or special case size zero. It does not need to NULL out the items before they are set. It can determine whether the tuple needs to be tracked or not at construction without the overhead of tracking it, then later untracking it.
Compared to PyTuple_FromArray
The advantage is less than for PyTuple_New, but still significant, as it reduces memory traffic by not storing the items in an array, then reading them out again.
Safety
This has the same safety advantages as PyTuple_FromArray, as it never produces an incomplete tuple. Since much of the code it will replace is using PyTuple_New, this is a worthwhile improvement
Implementation
We should make this a private API to start with. Once it proves it use, then we can make it public.
PyObject *_PyTuple_MakePair(PyObject *a, PyObject *b);
PyObject *_PyTuple_MakePairSteal(PyObject *a, PyObject *b);
Linked PRs
- gh-145325
- gh-145842
- gh-145884
- gh-145910
- gh-146538
- gh-146537
- gh-148000
- gh-148003
貢獻指南
從這裡開始
- 先讀完整個 Issue,再讀專案的貢獻指南。
- 在 Issue 下留言說明你要接手 —— 這能避免兩個人做同樣的事。
- Fork 儲存庫,在一個分支上完成修改。
- 送出 Pull Request,並在描述裡引用這個 Issue 編號。
評估
這個 Issue 還沒有評估資料。