On 22. 07. 21 15:03, Ethan Furman wrote:
On 7/22/21 1:01 AM, Petr Viktorin wrote:
> On 21. 07. 21 14:18, Nick Coghlan wrote:
>> ====================
>> typedef enum {
>> PyLocals_UNDEFINED = -1;
>> PyLocals_DIRECT_REFERENCE = 0,
>> PyLocals_SHALLOW_COPY = 1
>> } PyLocals_Kind;
>>
>> PyLocals_Kind PyLocals_GetKind(void);
>> PyLocals_Kind PyFrame_GetLocalsKind(PyFrameObject *);
>> ====================
>
> Please don't put the enum in the stable ABI. If we would add another
value and then
> an older extension would receive it, we'd get undefined behavior.
Probably a stupid question, but wouldn't the same thing happen if we
didn't use an enum, added another option later, and on older extension
received that newer value?
No.
Consider code like:
if (PyLocals_GetKind() == PyLocals_DIRECT_REFERENCE) {
...
}
where PyLocals_GetKind() is defined as above, but returns 4.
Technically it's undefined behavior, so the compiler could decide to
wipe your disk or eat your pets in this case, but that's not very realistic.
More realistically, the compiler is free to only look at the last two
bits of the value, so 4 will be equivalent to 0, and the comparison will
be true. There are definitely architestures/compilers that do tricks
similar to that.
But if PyLocals_GetKind() returns an int, 4 != 0 isn't a problem.
_______________________________________________
Python-Dev mailing list -- python-dev@python.org
To unsubscribe send an email to python-dev-le...@python.org
https://mail.python.org/mailman3/lists/python-dev.python.org/
Message archived at
https://mail.python.org/archives/list/python-dev@python.org/message/HFH5MG3LXYEI4YPO3MTDVXTVO7XCUXEA/
Code of Conduct: http://python.org/psf/codeofconduct/
