Home Original page

declaration of variables?

Jp Calderone exarkun at intarweb.us
Sat Feb 15 12:46:48 EST 2003 
On Sat, Feb 15, 2003 at 06:05:09PM +0100, Andr? Jonsson wrote:
> Being a Python-n00b, in the midst of this ?: discussion, it occurred to me 
> that Python does not require variables be delcared before use.
> 
> Why is this so?  Isn't it very helpful if the compiler complains rather 
> than having strange run-time results?
> 
> Small example:
> 
>     if doStuff(42):
>         value=getNextValue()
>     else:
>         valeu=initialValue      # <-- Note the spelling error.
> 
> Surely, in an elightened language like Python it shouldn't be possible to 
> do this kind of error without the compiler complaining.
> 
> The compiler.compileFile() function doesn't complain either, nor does 
> pychecker seem to. It's almost as if it was by design... is it?
> 

  "enlightened" language?  Hrm.  Dynamic, sure... OO, okay... functional,
sometimes!  But "enlightened"?

  Chances are, if you have this kind of error in your program, there will be
an error, a NameError later on when you try to use "value" and it isn't
bound.  Compile time is almost always essentially the same as runtime, so
you get the information at about the same time.  Even better, if you have
unit tests that cover this code path, you find out before you even run your
actual program.

  Could the problem be avoided with variable declarations?  Yep.  Would I
trade the flexibility of the current system for a small improvement in
compile-time checking of my code?  Nah, I don't think so.

  Jp

> 
> /Andr?
> 
> -- 
> http://mail.python.org/mailman/listinfo/python-list

-- 
http://catandgirl.com/view.cgi?44
-- 
 up 6 days, 22:28, 4 users, load average: 0.40, 0.28, 0.13
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 196 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/python-list/attachments/20030215/57de3332/attachment.sig>
More information about the Python-list mailing list