I just applied 10 of your code inspection modernizations to ~40 cpp/h files and the result builds and runs. You guys ROCK!
I found one minor issue with the loop conversion in a cpp file that gets built for both x86 and x64:
static const int count = 10;
size_t m_pathPositions[count];
for (size_t ii = 0; ii < count; ++ii)
m_pathPositions[ii] = 1234;
This gets converted differently for x86 versus x64:
x86: for (unsigned int & pathPosition : m_pathPositions)
x64: for (unsigned long long & pathPosition : m_pathPositions)
Which of course results in a build error in one of the configurations. It seems like the correct conversion is:
for (size_t & pathPosition : m_pathPositions)
which builds fine in both cases.
Is it a reasonable request that size_t is special-cased in the loop conversion?
