Message 355141 - Python tracker

➜

This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author	wim.glenn
Recipients	docs@python, wim.glenn
Date	2019-10-22.15:51:00
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1571759460.27.0.105772262724.issue38557@roundup.psfhosted.org>
In-reply-to

Content
https://docs.python.org/3/c-api/tuple.html#c.PyTuple_GetSlice In the c-api it says (emphasis mine): PyObject* PyTuple_GetSlice(PyObject p, Py_ssize_t low, Py_ssize_t high)¶ Return value: New reference. Take a slice of the tuple pointed to by p from low to high and return it as a new tuple*. But when slicing the entire tuple, CPython will return already existing instance, not a new tuple. https://github.com/python/cpython/blob/3.8/Objects/tupleobject.c#L448-L451 I propose the language in the docs be loosened so as not to tie the hands of implementation

https://docs.python.org/3/c-api/tuple.html#c.PyTuple_GetSlice

In the c-api it says (emphasis mine):

    PyObject* PyTuple_GetSlice(PyObject *p, Py_ssize_t low, Py_ssize_t high)¶
    Return value: New reference.

    Take a slice of the tuple pointed to by p from low to high and return it **as a new tuple**.

But when slicing the entire tuple, CPython will return already existing instance, not a new tuple.

https://github.com/python/cpython/blob/3.8/Objects/tupleobject.c#L448-L451

I propose the language in the docs be loosened so as not to tie the hands of implementation

History
Date	User	Action	Args
2019-10-22 15:51:00	wim.glenn	set	recipients: + wim.glenn, docs@python
2019-10-22 15:51:00	wim.glenn	set	messageid: <1571759460.27.0.105772262724.issue38557@roundup.psfhosted.org>
2019-10-22 15:51:00	wim.glenn	link	issue38557 messages
2019-10-22 15:51:00	wim.glenn	create