strange math?

[joe.hrbek@gmail.com]

Hello everyone, I'm experimenting with python and i'm following this
tutorial:
http://docs.python.org/tut/node6.htm...00000000000000 I'm
in section 4.7.5 Lambda Forms. In this section I was working along and
I noticed something strange. It happened because of a typo. Below is
a copy/paste from my idle session:
[color=blue][color=green][color=darkred]
>>>def make_incremento r(n):[/color][/color][/color]
return lambda x: x+n
[color=blue][color=green][color=darkred]
>>>f=make_incre mentor(42)
>>>f(0)[/color][/color][/color]
42[color=blue][color=green][color=darkred]
>>>f(1)[/color][/color][/color]
43[color=blue][color=green][color=darkred]
>>>f(10)[/color][/color][/color]
52[color=blue][color=green][color=darkred]
>>>f(0)[/color][/color][/color]
42[color=blue][color=green][color=darkred]
>>>f(01)[/color][/color][/color]
43[color=blue][color=green][color=darkred]
>>>f(02)[/color][/color][/color]
44[color=blue][color=green][color=darkred]
>>>f(010)[/color][/color][/color]
50[color=blue][color=green][color=darkred]
>>>42+010[/color][/color][/color]
50

The first f(01) was a mistake. I accidentally forgot to delete the
zero, but to my suprise, it yielded the result I expected. So, I tried
it again, and viola, the right answer. So, I decided to really try and
throw it for a loop, f(010), and it produced 50. I expected 52
(42+10). Why doesn't python ignore the first zero and produce a result
of 52? It ignored the first zero for f(01) and f(02). Hmm. I know, I
know, why am I sending it a 01,02, or a 010 to begin with? Like I
said, it was an accident, but now i'm curious. I'm not a computer
science major so please be kind with any explanations.

[Steven Bethard]

Re: strange math?

joe.hrbek@gmail .com wrote:[color=blue][color=green][color=darkred]
>>>> f(01)[/color][/color]
> 43[color=green][color=darkred]
>>>> f(02)[/color][/color]
> 44[color=green][color=darkred]
>>>> f(010)[/color][/color]
> 50[color=green][color=darkred]
>>>> 42+010[/color][/color]
> 50
>
> The first f(01) was a mistake. I accidentally forgot to delete the
> zero, but to my suprise, it yielded the result I expected. So, I tried
> it again, and viola, the right answer. So, I decided to really try and
> throw it for a loop, f(010), and it produced 50. I expected 52
> (42+10). Why doesn't python ignore the first zero and produce a result
> of 52? It ignored the first zero for f(01) and f(02). Hmm. I know, I
> know, why am I sending it a 01,02, or a 010 to begin with? Like I
> said, it was an accident, but now i'm curious. I'm not a computer
> science major so please be kind with any explanations.[/color]

Python isn't ignoring the initial 0. It reads the initial 0 as
indicating that the following number is expressed in octal. You can see
this if you try any number with a digit higher than 7 in it:
[color=blue][color=green][color=darkred]
>>> 08[/color][/color][/color]
Traceback ( File "<interacti ve input>", line 1
08
^
SyntaxError: invalid token

So you get 8 added to your number when you write 010 because eight is
spelled as 10 in octal.

The leading-zero notation is unfortunate, and there has been some recent
discussion[1][2] on python-dev about trying to change the prefix to
``0o`` or ``0c`` in Python 3.0 but for the moment at least it looks like
we're stuck with the confusing leading zero.

[1]http://mail.python.org/pipermail/python-dev/2006-January/060262.html
[2]http://mail.python.org/pipermail/python-dev/2006-February/060277.html

STeVe

[Christian Stapfer]

Re: strange math?

<joe.hrbek@gmai l.com> wrote in message
news:1142745707 .384085.42050@i 39g2000cwa.goog legroups.com...[color=blue]
> Hello everyone, I'm experimenting with python and i'm following this
> tutorial:
> http://docs.python.org/tut/node6.htm...00000000000000 I'm
> in section 4.7.5 Lambda Forms. In this section I was working along and
> I noticed something strange. It happened because of a typo. Below is
> a copy/paste from my idle session:
>[color=green][color=darkred]
>>>>def make_incremento r(n):[/color][/color]
> return lambda x: x+n
>[color=green][color=darkred]
>>>>f=make_incr ementor(42)
>>>>f(0)[/color][/color]
> 42[color=green][color=darkred]
>>>>f(1)[/color][/color]
> 43[color=green][color=darkred]
>>>>f(10)[/color][/color]
> 52[color=green][color=darkred]
>>>>f(0)[/color][/color]
> 42[color=green][color=darkred]
>>>>f(01)[/color][/color]
> 43[color=green][color=darkred]
>>>>f(02)[/color][/color]
> 44[color=green][color=darkred]
>>>>f(010)[/color][/color]
> 50[color=green][color=darkred]
>>>>42+010[/color][/color]
> 50
>
> The first f(01) was a mistake. I accidentally forgot to delete the
> zero, but to my suprise, it yielded the result I expected. So, I tried
> it again, and viola, the right answer. So, I decided to really try and
> throw it for a loop, f(010), and it produced 50. I expected 52
> (42+10). Why doesn't python ignore the first zero and produce a result
> of 52?[/color]

That's because python interprets 010 as *octal* 10
which is *decimal* 8. Thus

42+010 = 42+8 = 50

which is quite as it should be...

Regards,
Christian

[Terry Reedy]

Re: strange math?

[color=blue][color=green][color=darkred]
>>>>f(01)[/color][/color]
> 43[color=green][color=darkred]
>>>>f(02)[/color][/color]
> 44[color=green][color=darkred]
>>>>f(010)[/color][/color]
> 50[color=green][color=darkred]
>>>>42+010[/color][/color]
> 50[/color]

Literal ints with lead 0 are interpreted as octal numbers (base 8) instead
of decimal numbers (base 10). 01==1 either way, but 010 == 8.[color=blue][color=green][color=darkred]
>>> 010[/color][/color][/color]
8

0x (zero eks) prefix indicate hexadecimal numbers:[color=blue][color=green][color=darkred]
>>> 0xff[/color][/color][/color]
255

Terry Jan Reedy

[joe.hrbek@gmail.com]

Re: strange math?

Thanks for the great reply, Steve, et al.

-j

[Ron Adam]

Re: strange math?

joe.hrbek@gmail .com wrote:
[color=blue][color=green][color=darkred]
>>>> f(01)[/color][/color]
> 43[color=green][color=darkred]
>>>> f(02)[/color][/color]
> 44[color=green][color=darkred]
>>>> f(010)[/color][/color]
> 50[color=green][color=darkred]
>>>> 42+010[/color][/color]
> 50
>
> The first f(01) was a mistake. I accidentally forgot to delete the
> zero, but to my suprise, it yielded the result I expected. So, I tried
> it again, and viola, the right answer. So, I decided to really try and
> throw it for a loop, f(010), and it produced 50. I expected 52
> (42+10). Why doesn't python ignore the first zero and produce a result
> of 52? It ignored the first zero for f(01) and f(02). Hmm. I know, I
> know, why am I sending it a 01,02, or a 010 to begin with? Like I
> said, it was an accident, but now i'm curious. I'm not a computer
> science major so please be kind with any explanations.[/color]

Number beginning with the digit '0' are octal (base 8), so 010 == 8.
42 + 8 = 50.

Numbers beginning with '0x' are base 16.

Cheers,
Ron