"Neil Cerutti" <[EMAIL PROTECTED]> wrote in message news:[EMAIL PROTECTED] > On 2006-10-04, Paul McGuire <[EMAIL PROTECTED]> wrote: >> "Neil Cerutti" <[EMAIL PROTECTED]> wrote in message >> news:[EMAIL PROTECTED] >> > > It seems to be timing "testing for membership", not "random > access". Random access is just seq[n]; at least, that's the > assumption I made. > > Perhaps I read the test description out of context and got > confused. >
Oh, poor wording on my part, sorry. You got me wondering what seq[n] would give me in the case of a set: >>> z = set(range(100)) >>> z[30] Traceback (most recent call last): File "<stdin>", line 1, in ? TypeError: unindexable object >>> It really doesn't make any sense since sets are unordered. So we couldn't compare lists and sets in this manner anyway. I *did* learn that dicts and sets use a hash table, not a tree, which also helps explain why the initial construction of the 10000-element set takes so much longer than just simple assembly of the range into a list. -- Paul -- http://mail.python.org/mailman/listinfo/python-list
