[3.12] gh-110383: Improve accuracy of str.split() and str.rsplit() do…

…cstrings (GH-113355) (#113379)

Co-authored-by: Erlend E. Aasland <erlend@python.org>

Objects/unicodeobject.c

@@ -12410,11 +12410,13 @@ str.split as unicode_split
         character (including \n \r \t \f and spaces) and will discard
         empty strings from the result.
     maxsplit: Py_ssize_t = -1
-        Maximum number of splits (starting from the left).
+        Maximum number of splits.
         -1 (the default value) means no limit.
 
 Return a list of the substrings in the string, using sep as the separator string.
 
+Splitting starts at the front of the string and works to the end.
+
 Note, str.split() is mainly useful for data that has been intentionally
 delimited.  With natural text that includes punctuation, consider using
 the regular expression module.
@@ -12423,7 +12425,7 @@ the regular expression module.
 
 static PyObject *
 unicode_split_impl(PyObject *self, PyObject *sep, Py_ssize_t maxsplit)
-/*[clinic end generated code: output=3a65b1db356948dc input=07b9040d98c5fe8d]*/
+/*[clinic end generated code: output=3a65b1db356948dc input=a29bcc0c7a5af0eb]*/
 {
     if (sep == Py_None)
         return split(self, NULL, maxsplit);