On Monday, December 22, 2014 3:57:31 PM UTC-8, Dave Angel wrote: > On 12/22/2014 06:48 PM, Ian Kelly wrote: > > On Dec 22, 2014 2:37 PM, "Dave Angel" <da...@davea.name> wrote: > >> > >> I'll pick on one function first, called instructions(). If the user > > types something invalid, you print "Invalid input." and call the function > > again. In this case, because the call is at the end, no harm is usually > > done, but it would be tricky to explain why. If the user happened to type > > the wrong input 1000 times, you'd hit a recursion limit and crash, but > > that's unlikely unless you have a very stubborn user. > > > > Point of interest, Python (or at least CPython) doesn't optimize tail-call > > recursion, so in fact this sort of unbounded recursion (a "stack leak", if > > you will) is equally bad no matter where it appears in the function. > > > > Which is exactly why I mentioned the recursion limit of about 1000. > > > -- > DaveA
Python 2.7.3 (default, Dec 4, 2012, 17:28:13) [GCC 4.4.5 20110214 (Red Hat 4.4.5-6)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> def test(t): ... print(t) ... test(t+1) ... >>> test(1) 1 2 3 4 <snip> 998 999 Traceback (most recent call last): File "<stdin>", line 1, in <module> File "<stdin>", line 3, in test File "<stdin>", line 3, in test File "<stdin>", line 3, in test <snip> File "<stdin>", line 3, in test File "<stdin>", line 3, in test RuntimeError: maximum recursion depth exceeded Huh...there actually is a limit of about 1,000. I'm assuming this is hard-coded? I did a similar test with Java a while back and was getting different results every time. -- https://mail.python.org/mailman/listinfo/python-list
