> PyBytes_FromStringAndSize(const char *v, Py_ssize_t len): 
> Return a new bytes object with a copy of the string v as value and length len on success, and NULL on failure. If v is NULL, the contents of the bytes object are uninitialized.

The function actually copies `len` bytes from string v instead of the whole string. (link 1) I suggested we can change to 

> Return a new bytes object with a copy of the first len bytes of string v as value and length len on success...

> readinto(b)
> Read bytes into a pre-allocated, writable bytes-like object b and return the number of bytes read. For example, b might be a bytearray.

The function will call _bufferediobase_readinto_generic(link 2) which may return NULL. Maybe we can change to 

> and return the number of bytes that read succeed (it may be smaller than b or NULL if failed)...

1. https://github.com/python/cpython/blob/master/Objects/bytesobject.c#L126

2. https://github.com/python/cpython/blob/master/Modules/_io/bufferedio.c#L962

Can you prepare a PR?

Sure. Feel free to edit my PR.

> The function actually copies `len` bytes from string v instead of the whole string. 

"and length *len*" means it.  So it is not a bug.

If you want to rewrite it to "the first *len* bytes of", you should remove "and length *len*".

But please note that every small change will break translations.
Unless you are sure it is better than the current version, please don't change it.

> _bufferediobase_readinto_generic(link 2) which may return NULL.

It doesn't return NULL.  NULL in C means it raises the error.
And we don't document it usually.  Otherwise, almost all document has "it may raise an exception when an error happens."