[Tutor] 'Common' mistake ... for other newbies

[denis]

Well,

Thank you David for this clear message.
Are they here people who think that such a so-called "mistake" is not a
mistake ? That David programmed  the way he (and we, humans  should ; and
that the error lies in language, not in his brain ? Why are indexes based on
0 instead of 1 ?

denis

[PS : if you're interested in such a way of reflexion, like "ameliorating
python", or "the ideal programming language", I would really welcome your
ideas -- outside the list]

> ----- Original Message -----
> From: David Broadwell <dbroadwell at mindspring.com>
> To: <tutor at python.org>
> Sent: Monday, April 12, 2004 5:00 PM
> Subject: [Tutor] 'Common' mistake ... for other newbies
>
>
> > I started with the text below, but if you don't like detailed reports of
> > newbie fumbles, skip this post. Points to not quitting but continuing to
> > hack at it ... my first real bug.
> >
> > ...
> >
> > I'm having an odd problem that I don't know how to debug.
> > What is going on is as a exercise, I'm playing rock paper scissors in
> code.
> > I'm using; Python 2.2, Windows2kServer, PII450 at 386mb w/80gb
> >
> > ...
> >
> > At full run with a 100 item result list it failed on the average 97th
> roll,
> > but with a 6 entry list only averages 10 rolls instead before dying. It
> > seems that the smaller the resultlist the more often an IndexError
occurs.
> >
> > The code is like this (static data for brevity);
> >
> > >>> from random import randint
> > >>> resultlist = ['rock','paper','scissors','rock','paper','scissors']
> > >>> def roll():
> >     ''' returns a random item from resultlist '''
> >     return resultlist[randint(0,len(resultlist))]
> >
> > >>> for item in range(1000):
> > roll()
> >
> > 'scissors'
> > 'paper'
> > 'paper'
> > 'paper'
> > 'paper'
> > 'rock'
> > 'scissors'
> > Traceback (most recent call last):
> >   File "<pyshell#9>", line 2, in ?
> >     roll()
> >   File "<pyshell#3>", line 3, in roll
> >     return resultlist[randint(0,len(resultlist))]
> > IndexError: list index out of range
> > >>>
> >
> > So to debug, I rewrote roll into debugroll()
> >
> > >>> def debugroll():
> >     temp = len(resultlist)
> >     print "len(resultlist) = %s" % temp
> >     temp = randint(0,temp)
> >     print "randint(0,len(resultlist)) = %s" % temp
> >     temp = resultlist[temp]
> >     print "resultlist[randint(0,len(resultlist))] = %s" % temp
> >     return temp
> >
> > >>> for item in range(1000):
> > debugroll()
> >
> > len(resultlist) = 6
> > randint(0,len(resultlist)) = 5
> > resultlist[randint(0,len(resultlist))] = scissors
> > 'scissors'
> > len(resultlist) = 6
> > randint(0,len(resultlist)) = 6
> > Traceback (most recent call last):
> >   File "<pyshell#4>", line 2, in ?
> >     debugroll()
> >   File "<pyshell#0>", line 6, in debugroll
> >     temp = resultlist[temp]
> > IndexError: list index out of range
> >
> > after a bit more testing (or frantically repeating the error) I noted
that
> > it only died when the randomly generated index was 6 ... and then it hit
> me,
> > a list of len 6 has no index six and the uncrashable roll function
needed
> to
> > reflect that is as follows.
> >
> > The solution;
> > >>> def roll():
> >     ''' returns a random item from resultlist '''
> >     return resultlist[randint(0,len(resultlist)-1)]
> >
> > ...
> >
> > I got bit because I count from 1 and the computer counts from 0, a
> 'common'
> > mistake.
> >
> > As a note this was the last hurdle to having the code play rock paper
> > scissors against itself, I'm sure it's not nearly as elegant as it could
> be,
> > but it's very, very, readable.
> >
> > --
> >
> > Programmer's mantra; Observe, Brainstorm, Prototype, Repeat
> >
> > David Broadwell
> >
> >
> > _______________________________________________
> > Tutor maillist  -  Tutor at python.org
> > http://mail.python.org/mailman/listinfo/tutor
> >
>