Steve Holden wrote: > James Stroud wrote: >> Steve Holden wrote: >>> You'd be worth more if you'd used elif and omitted the continue >>> statements, but for a first solution it's acceptable. >> >> Depends on what you are after. >> >> py> s = """ >> ... for i in xrange(1,101): >> ... if not i % 15: >> ... continue >> ... if not i % 5: >> ... continue >> ... if not i % 3: >> ... continue >> ... else: >> ... pass >> ... """ >> py> t = timeit.Timer(stmt=s) >> py> print "%.2f usec/pass" % (1000000 * t.timeit(number=100000)/100000) >> 40.49 usec/pass >> py> s = """ >> ... for i in xrange(1,101): >> ... if not i % 15: >> ... pass >> ... elif not i % 5: >> ... pass >> ... elif not i % 3: >> ... pass >> ... else: >> ... pass >> ... """ >> py> t = timeit.Timer(stmt=s) >> py> print "%.2f usec/pass" % (1000000 * t.timeit(number=100000)/100000) >> 40.88 usec/pass >> > > To be strictly comparable you should have pass statements before the > continue statements as well. Ignoring that, clearly it's well worth > saving that extra 390 nanoseconds each time round the loop. > > Repeat after me "premature optimization is the root of all evil". > > http://en.wikipedia.org/wiki/Optimization_(computer_science) > > regards > Steve
The point is that both work equally as well unless (1) you are biased by a particular style or (2) you split little bitty hairs to even finer hairs. -- http://mail.python.org/mailman/listinfo/python-list
