I would rewrite it in a bit different way: Fixing a bug in one place and being sure only one place was affected and being sure that one place was really fixed

I would rewrite it in a bit different way: Fixing a bug in one place and being sure only one place was affected and being sure that one place was really fixed is cheaper - than fixing a bug in one place affecting 20 where in 15 places it was a proper fix, in 5 places it will break in unforeseen way when user does something different and somehow additional 2 places totally broken because no one ever knew these were affected.