[libcxx] Use integer division

[phosek]

Summary

In Python 3, math.floor returns int when both arguments are ints.
In Python 2, math.floor returns float. This leads to a failure
because the result of math.floor is used as an array index. While
Python 2 is on its way out, it's still used in some places so use an
integer division instead.

[• Quuxplusone]

libcxx/utils/gdb/libcxx/printers.py

442

Even better: word = int(bit / self.bits_per_word)
(Note that math.floor rounds upward when the argument is negative; but we don't ever expect the argument to be negative here, so "cast to int" is exactly the right semantics.)

Even more better: word = bit // self.bits_per_word
to just use integer division in the first place. This is the best option, because // (line 442) and % (line 443) go together nicely as a pair.

[• Quuxplusone]

(Wait for a second libc++ reviewer to give it the green light; then) Ship it!

[curdeius]

LGTM.

libcxx/utils/gdb/libcxx/printers.py

def __init__(self, val):
                            for index in range(self.n_words)]
 
     def to_string(self):
         typename = _prettify_typename(self.val.type)
         return "%s" % typename
 
     def _list_it(self):
         for bit in range(self.bit_count):
-            word = math.floor(bit / self.bits_per_word)
+            word = bit // self.bits_per_word
             word_bit = bit % self.bits_per_word
             if self.values[word] & (1 << word_bit):
                 yield ("[%d]" % bit, 1)
 
     def __iter__(self):
         return self._list_it()
 
     def children(self):