gh-50333: Deprecate support of non-tuple sequences in PyArg_ParseTuple() #128374
Conversation
…seTuple() Non-tuple sequences are deprecated as argument for the "(items)" format unit in PyArg_ParseTuple() and other argument parsing functions if items contains format units which store borrowed buffer or reference (e.g. "s" and "O"). str and bytearray are no longer accepted as valid sequences.
Misc/NEWS.d/next/C_API/2024-12-31-15-28-14.gh-issue-50333.KxQUXa.rst
Outdated
Show resolved
Hide resolved
Co-authored-by: Bénédikt Tran <[email protected]> Co-authored-by: Stan Ulbrych <[email protected]>
In this case, I think we should consider being explicit, rather than worrying about the repeated word. |
| levels[0] = 0; | ||
| PyOS_snprintf(msgbuf, bufsize, | ||
| "must be %d-item sequence, not %.50s", | ||
| "must be %d-item tuple, not %.50s", |
There was a problem hiding this comment.
snprintf will keep the string within bufsize; we should not need to use the sized specifier.
| "must be %d-item tuple, not %.50s", | |
| "must be %d-item tuple, not %s", |
There was a problem hiding this comment.
This is for consistency with other formats. Also, if we will add more text after the type name, it is easier to not forget to truncate.
There was a problem hiding this comment.
No worry; snprintf will truncate.
| if (PyTuple_Check(arg)) { | ||
| Py_INCREF(arg); | ||
| } |
There was a problem hiding this comment.
We can easily avoid the unneeded incref/decref if we refactor out the convertitem loop. I'll make a suggestion to your fork for your consideration.
There was a problem hiding this comment.
Suggestion opened:
There was a problem hiding this comment.
I thought about this, but I am not sure that there is a significant benefit in this. It would complicate the code, for sure. So I left the simpler code.
There was a problem hiding this comment.
So I left the simpler code.
However, the "simpler" code is more complex with regards to reference counting. IMO, keeping the ref counting simple is worth it; ref count bugs are hard to catch. But it is your call. BTW, I addressed your remarks on your fork.
Co-authored-by: Erlend E. Aasland <[email protected]>
Non-tuple sequences are deprecated as argument for the "(items)" format unit in PyArg_ParseTuple() and other argument parsing functions if items contains format units which store borrowed buffer or reference (e.g. "s" and "O").
str and bytearray are no longer accepted as valid sequences.
📚 Documentation preview 📚: https://cpython-previews--128374.org.readthedocs.build/