Simply Python list question

[amiga_os]

Hi, sorry for the dumb question:

In c I can define an array like this:


int mylist[40]

That will give me 40 blank variables to work with. How do I do that in Python? From the docs - and maybe I missed something - I have to do this:


mylist = [range(40)]
for i in range(40):

mylist[i] = 0

Python seems so tidy... there must be a better way of doing this. Can anyone help?

[amiga_os]

And... funnily enough, that code doesn't even work either!

This is a frustratingly simple problem.... but I can't find the right syntax

[amiga_os]

OK...

fiddled more and solved the problem on my own, define the list like this:


mylist = [0] * x

Where x is the size you want the list to be.

[Quikee]

Is there a reason why you want to predefine 40 places for a variable. Lists in Python are dynamic which means that the size doesn't need to be predefined, which is not true for arrays in C.

If you need predefined array/list because you have to operate on 40 values pre setted to 0 you have to create 40 values and set them to 0, which is also true for C (int mylist[40]; also doesn't create an array with all values set to 0 but an array with all values set to "undefined" value, which is mostly 0). This is just what you did - but be cautious with such usage as it works in your case when using numbers, but if you would do [[]] * 40 you will get a list with 40 empty lists which are all the same instance of the list.

try:


a = [ ['a'] ] * 40
a[0].append('b')

and you will se what I mean. ;)

[pmasiar]

Yes, there is reason to predefine size. Look what I mean:



>>> ll = [1] * 40
>>> ll[30]
1
>>> ll[50]

Traceback (most recent call last):
File "<pyshell#5>", line 1, in -toplevel-
ll[50]
IndexError: list index out of range


You can expand list, but it does not happen automagically. Reaching beyond current list limits is fortunately an error.

[dwblas]

For future reference, the code you posted could use either of the following
test_1 = []
test_2 = []
for j in range(10):
test_1 += [0]
test_2.append( 0 )

##--- change some elements to verify that they are unique
test_1[1] = 1
test_1[3] = 3
print test_1
test_2[2] = 2
test_2[4] = 4
print test_2

##--- you can also use
test_3 = [0] * 10
test_3[5] = 5
test_3[6] = 6
print test_3

[Smygis]

Hi, sorry for the dumb question:

In c I can define an array like this:


int mylist[40]

That will give me 40 blank variables to work with. How do I do that in Python? From the docs - and maybe I missed something - I have to do this:


mylist = [range(40)]
for i in range(40):

mylist[i] = 0

Python seems so tidy... there must be a better way of doing this. Can anyone help?


myList = [0 for i in xrange(40)]

:popcorn:

[thumper]

My question would be different.

Why do you think you need a list with 40 zeros in it?

[pmasiar]

To allocate space for 40 elements?

[ankursethi]

Yes, but can't elements be added to the list as the need arises? I think this question could be solved better in context of the problem.

Anyway, I suppose list comprehensions are the way to go.
myList = [0 for i in range(40)]
(Is there a better way to do this?)

[pmasiar]

(Is there a better way to do this?)

I like much more above mentioned my_list = [0] * 40. Less fluff IMHO :-)

[hecato]

Should I ask if there exist a performance difference?? (blame me :P)


[0] * 40
[0,0] * 20
[0,0,0,0] * 10
[0,0,0,0,0]*10
OK the last is my fail I dont remember 5*?=40... o yes

[0,0,0,0,0]*8

[pmasiar]

Why would anyone using Python bothered with possible (and negligible) difference between [0] * 40 and [0,0] * 20? My first reaction was: first option is more obvious, so, preferable.

[nanotube]

To allocate space for 40 elements?

to allocate space for 40 elements, why not just

mylist = range(40)
allocating space doesn't necessitate setting elements to 0...

[pmasiar]

IMHO OP wanted also to initialize values in list to 0. Maybe not :-/