On 16/09/2020 5:25 am, gah4 wrote:

> I think I remember this being discussed many years ago.

> One thought was that someone codes bubblesort, and the compiler
> generates quicksort. Small complication that bubblesort is stable, and
> quicksort isn't. (Add an array with the original position to break
> ties.)

Here are even more dragons.

What if you're sorting exactly three elements? In that situation, I'll
be very surprised if bubblesort doesn't outperform all other algorithms.

How does the /compiler/ decide that an algorithm is better, based
on data it never sees? Even if we account for profile guided optimi-
zations, we shouldn't allow the compiler to adjust for /what if this is
run with more data in production?/ kind of questions.

If you want to go down this rabbit hole, I'd much prefer warnings than
outright algorithm changes.

--
Johann

I'm not from the internet, I just work there.