A couple small things that will help improve memory management
Rather than
avg = sumall / count;
return avg;
Just return sumall/count instead. Then you don't have to waste a register or
assignment operation.
Division is expensive. Avoid it when you can.
Here,
for (a=0; a != count; a++) {
temp = PyFloat_AsDouble(PySequence_Fast_GET_ITEM(seq,a));
sumall += temp;
Again, save variables and operations. Write this as:
for (a=0; a != count; a++) {
sumall +=
PyFloat_AsDouble(PySequence_Fast_GET_ITEM(seq,a));
Similar corrections in var()
It's cheaper when you're using powers of two to just right or left-shift: >> or
<<. Since you want to increase by a power of two, do:
(avg - PyFloat_AsDouble(PySequence_Fast_GET_ITEM(seq,a) << 1; // This means
(...)^(2^1)
Division by powers of two is >>. Note that these only works for powers of two.
Now I haven't worked with pointers in a long time and didn't fully trace this
out so I'm probably wrong, but it doesn't seem like you ever have your pointers
in stat_avg() point to an object. Therefore wouldn't they always be Null?
R.M. ArceJaeger
Author/Publisher, Platypus Press
Contact: [email protected]
Website: http://rmarcejaeger.com
On May 26, 2011, at 8:22 AM, James Reynolds wrote:
> Hello All:
>
> As an intellectual exercise, I wanted to try my hand at writing some
> extensions in C.
>
> I was wondering if you all could look over my code and give some feedback.
>
> Here is the link for the code: http://pastebin.com/jw3ihfsN
>
> I have zero experience coding in C (and not much more coding in Python!).
> Being a kinetic learner, I thought this would be a good exercise to teach me
> some of the underpinnings of Python, how it works, why it works the way it
> does, and as an added bonus, skills to actually write my own extensions if I
> ever wanted to.
>
> I had to learn about pointers to do this, and I'm still not 100% on if I used
> them correctly herein.
>
> I am also very concerned with memory management because I am not sure when I
> should be calling the memory allocation macros to decref or incref when
> needed.
>
> I would also like to get feedback on how I am constructing C algorithms.
>
> As far as the module itself goes, I was able to compile and use it on a
> windows machine compiling with mingw (I use distutils to do the work, so for
> me I do "python setup.py build" in my CMD.
>
> There are three functions, stats.mean, stats.var, stats.stdev (and they do
> what you would expect). One thing though, these are the "population"
> statistics and not "sample" in case you want to test it out.
>
> Also, anything else that you think would be worthwile pointing out, tips and
> tricks, common pitfalls, etc.
>
> Thanks in advance for you feedback.
>
> _______________________________________________
> Tutor maillist - [email protected]
> To unsubscribe or change subscription options:
> http://mail.python.org/mailman/listinfo/tutor
_______________________________________________
Tutor maillist - [email protected]
To unsubscribe or change subscription options:
http://mail.python.org/mailman/listinfo/tutor
