Macro like functionality for shorthand variable names

Larry Bates larry.bates at websafe.com`
Sat Jun 7 21:44:33 CEST 2008


Tilman Kispersky wrote:
> I have python code in a class method translated from C++ that looks
> sort of like this:
> 
>>>>  self.dydt[1] = self.a * (self.b * self.y[0] - self.y[1])
> 
> 
> To make this more readable in C++ I had made macros to achieve this:
> #define du (dydt[1])
> #define u (y[1])
> #define V (y[0])
> 
> du = a * (b * V - u);
> 
> 
> I realize the value of not having macros in Python.  They've tripped
> me up more than once in C++.  My question is:
> Is there any way to write a shorterhand more readable version of the
> python code above?  I'm doing several calculations one after the other
> and some of the lines are quite long.

Names in Python point to objects so you can bind several names to the same location.

 >>>>  self.dydt[1] = self.a * (self.b * self.y[0] - self.y[1])

The following will work.

a = self.a
b = self.b
u = self.y[0]
V = self.y[1]

self.dytd[1] = a * (b * V - u)

In addition to making the lines shorter, I'm pretty sure they will be slightly 
faster because lookups on the local variables is slightly faster than on the 
instance variables (self.?).

Note: I think your parenthesis are incorrect in the equation above.  If they 
aren't then no parenthesis are necessary since multiplication takes precedence 
over subtraction.

Hope this helps.

-Larry



More information about the Python-list mailing list