Steven D'Aprano wrote: > On Sat, 09 Feb 2008 14:28:15 -0800, Jeff Schwab wrote: > >> Steven D'Aprano wrote: >>> On Sat, 09 Feb 2008 13:37:23 -0800, Jeff Schwab wrote: >>> >>>> Carl Banks wrote: >>>>> On Feb 8, 10:09 pm, Jeff Schwab <[EMAIL PROTECTED]> wrote: >>>>>> If you expect your data to be pretty nearly sorted already, but you >>>>>> just want to make sure (e.g. because a small number of elements may >>>>>> have been inserted or removed since the last sort), bubble-sort is a >>>>>> good choice. >>>>> But if you're at that stage you probably were doing something wrong >>>>> in the first place. >>>> How do you figure? You don't always have control over the >>>> insertions/replacements, or where they happen. As a silly example, >>>> assume you have a sorted list of children, by height. Maybe you check >>>> your list once per school semester. The new sort order for any given >>>> semester will likely be very close to the previous order; however, a >>>> few >>>> swaps may be in order, according to the different speeds at which >>>> children have grown. >>> You check their heights once per semester, but you're worried about an >>> extra ten or twenty microseconds to sort the data? >> Are you serious? >> >> The fact that you wouldn't use a Python script for this is what makes it >> a "silly" example in the first place. > > > Okay, now you've broken my brain. Why on earth do you think Python isn't > suitable for processing a list of a few hundred, or even thousand, items?
Either you're joking, or we're definitely not getting anywhere. I think we're through here. See you else-thread. No hard feelings. -- http://mail.python.org/mailman/listinfo/python-list