zaheer.ag...@gmail.com wrote:
hi
I am from java background, I have written some code that actually
works :)
Need to some one to look at it and tell me if there are better ways of
doing same things
Will some one help?
Thanks
My crystal ball is a bit cloudy today so forgive me if my suggestions
may seem off-topic in your request.
I find that reviewing code that hasn't conform to 'standards' distracts
me from the original intent. Parsing your code through a lint like
PyLint will suggest 'standard' way of doing things (aim for 10/10) it
will make things easier to read.
Also code that is unnecessary bloated (e.g. conditions that are never
executed) can confuse the issue even more so. A good way would be to
write unit-test against your code out of the perspective of the intended
use (so not just to execute stuff because it is there) then run it
through a coverage program. This will display nicely what code is
executed and what not. The not part is what you look for in this
perspective since if all the unit-test have what you need, why should
the uncalled code be included?
Next stop comments, I am all for comments except if you need them to
translate what the piece of code is doing there, if you can refactor
your code so it looks more English and is understandable without
comments (which should be still in there though) then do so.
For example:
range_8 = range(8)
for number in range_8:
# Loop through 0 to 7 and print it.
print(number)
instead of:
r = range(8)
for i in r:
# print i in range(8) loop
print(i)
* note this is a bad code example, only to clarify what I mean in general
Then look at your code from a functional perspective, if there are
multiple functions that act in a similar way can you replace them with a
single function? Then do so.
Is there functionality that looks so common that you think it should be
in the standard library? It probably is then, make use of it.
Performance; run your code through a execution analyser like profile, do
the numbers make sense? If not try to figure out why and replace the
affected algorithm with one that makes more sense.
Finally memory; Can you scrap duplicate entries of essential the same
data, without affecting all the above? Might be a good idea to do so.
When done all this you might feel it is not necessary to review the code
any more, which is then is a good moment to actually request a review :-)
I'll be happy to have a look at it though you might consider posting it
here, more chance of useful feedback ;-)
--
MPH
http://blog.dcuktec.com
--
http://mail.python.org/mailman/listinfo/python-list
