bpo-41366: Fix clang warning for sign conversion

[henryiii]

Affects 3.9 beta and master.

Should I make a back port PR too?

https://bugs.python.org/issue41366

[the-knights-who-say-ni]

Hello, and thanks for your contribution!

I'm a bot set up to make sure that the project can legally accept this contribution by verifying everyone involved has signed the PSF contributor agreement (CLA).

Recognized GitHub username

We couldn't find a bugs.python.org (b.p.o) account corresponding to the following GitHub usernames:

@henryiii

This might be simply due to a missing "GitHub Name" entry in one's b.p.o account settings. This is necessary for legal reasons before we can look at this contribution. Please follow the steps outlined in the CPython devguide to rectify this issue.

You can check yourself to see if the CLA has been received.

Thanks again for the contribution, we look forward to reviewing it!

[WildCard65]

Are you using a std-C implementation that doesn't define a "size_t" type?

[henryiii]

I see a few other uses of size_t in the code base, so I assume this is allowed?

The problem is that length is now forced to Py_ssize_t (it used to be a macro argument in Python 3.8), so length * sizeof(Py_UNICODE) is signed * unsigned, and then used as unsigned in the memcpy function. Clang warns about this sign conversion unless it is explicit.

Note the Azure error is a timeout.

[WildCard65]

@henryiii I ask because Python defines Py_ssize_t to size_t IF it exists.

#ifdefHAVE_SSIZE_Ttypedefssize_tPy_ssize_t;
#elifSIZEOF_VOID_P==SIZEOF_SIZE_TtypedefPy_intptr_tPy_ssize_t;
#else# error "Python needs a typedef for Py_ssize_t in pyport.h."
#endif

Edit: NVM, ssize_t is signed.

[henryiii]

The issue here is signed vs. unsigned - ssize_t vs. size_t. size_t is part of the standard, ssize_t is not, which might be why that define is there.

[miss-islington]

Thanks @henryiii for the PR, and @methane for merging it 🌮🎉.. I'm working now to backport this PR to: 3.9.
🐍🍒⛏🤖

[miss-islington]

Sorry @henryiii and @methane, I had trouble checking out the 3.9 backport branch.
Please backport using cherry_picker on command line.
cherry_picker 680254a8dc64e3ada00f88a7c42d41eb02108353 3.9

[miss-islington]

Thanks @henryiii for the PR, and @methane for merging it 🌮🎉.. I'm working now to backport this PR to: 3.9.
🐍🍒⛏🤖

[bedevere-bot]

GH-21600 is a backport of this pull request to the 3.9 branch.