Generating new object instances with different properties in Python

Question

OK. I'm setting up a model which will describe the effects of a number of management decisions on forests. Each forest is an instance of the forest class and you can see a simplified version below:

class forest():
     instancelist = [] # a list of all the forest instances so I can run functions on all of them at once

     growth_rate = 2 #very simple rate of growth (not realistic!)
     felling_year = 50 #all forest areas are felled at age 50

     def __init__(self, x=0, y=0,age=0,size=0): 
         self.instancelist.append(self) # add the forest area to the instance list
         self.x = x # x coordinate
         self.y = y # y coordinate
         self.age = age # age, not all forests are planted on bare sites, - we have some pre-existing ones to consider.
         self.size = size # very rough - but this is an indicator of the physical volume of timber (not area)

I can now generate a forest object, for example:

f = forest(1,1,20,40)

So, the difficulty I'm having is that I need to generate a number of forest blocks (so we see what the effect is on a wider area). To do this, I'll need to create a lot of areas.

I can do this easily if I don't specify any properties:

forests = [forest() for x in range(20)]

But I can't see how to generate a large number of areas each with unique properties without going through and specifying them all by hand. Is there a method that I can use where I feed in data from another source (lists, tuples, csv, whatever) and use it to build up an inventory of different objects?

Sorry if this is a stupid question (I have been known to ask them from time to time) but it's really confusing me.

Where are the properties stored? It seems there are 4 of them. Do you therefore have a list of lists [with each inner list having 4 items]? Or a generator? — jpp
– jpp, Commented Apr 13, 2018 at 14:26
There are 4 in this example -it's all stripped down, but there will probably be a lot of others. I don't have them all added yet, so I could produce them in whatever format works best really. Eventually, I'll probably need to use some form of configuration file for the starting conditions - but that's a bit of a way off yet. — Will
– Will, Commented Apr 13, 2018 at 14:30
OK, I guess your problem should be reduced to reading the config file into a list of lists, dictionary or other format. Then use a list comprehension. — jpp
– jpp, Commented Apr 13, 2018 at 14:32

jpp · Accepted Answer · 2018-04-13 14:30:39Z

2

Here is one way if you store your properties in a list of lists:

class Forest(object):
    def __init__(self, w, x, y, z):
        self.w = w
        self.x = x
        self.y = y
        self.z = z
        return None

properties = [[1, 2, 3, 4],
              [5, 6, 7, 8],
              [9, 10, 11, 12],
              [13, 14, 15, 16],
              [17, 18, 19, 20]]

forests = [Forest(*p) for p in properties]

print(forests[1].x)  # 6

answered Apr 13, 2018 at 14:30

jpp

166k37 gold badges301 silver badges362 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

Will Over a year ago

Thanks, this looks like what I'm after, quick couple of questions: in forests = [Forest(*p) for p in properties] what is the *p doing? also, why is the object needed in class Forest(object):

Patrick Haugh Over a year ago

@Will *p is called argument unpacking. It takes some iterable and unpacks it to populate the arguments. So Forest(*[1, 2, 3, 4]) becomes Forest(1, 2, 3, 4) after unpacking. The object is the old fashioned/Python 2 way of declaring classes (because every class is a child class of the object class). In Python 3, you can omit that and just do class Forest:, but including it doesn't hurt anything.

abarnert Over a year ago

@PatrickHaugh Actually, in Python 2, every class is not a subclass of object unless you specifically ask for it like this (and you don’t get some modern features without it); in Python 3, every class is a subclass of object (with all the extra features), so you don’t need to specify it anymore.

abarnert Over a year ago

@Will The nice thing about this answer is that the same code will work if properties is a list of dicts with keys w, x, y, and z just by changing the * to a **, and a dict like that stored in a JSON or YAML or CSV config file is a great way to make them human-readable and -editable, so this gives you room to grow into whatever design you want.

Collectives™ on Stack Overflow

Generating new object instances with different properties in Python

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related