this is sort of fascinating
ohhh, I haven't published the version of helix that includes the key/ref fix for factories
after fixing the :key
issue w/ factories, it matches more what I expected
interesting that not using keys improved the performance in some cases (as seen above this one)
but I didn't bother running the check, so it could be that the swap rows test above was actually incorrect
that would make more sense
i'm not sure why select row is so slow
probably a ref is cleared?
dunno. I also saw that same behavior with helix and reagent before, could be a fluke
also, I released v0.0.13 with the fix for factory components