Python: Variable not re-initializing after each iteration-CodePudding

I am learning about data handling in python, trying to handle weather data from each day of October. The data is from a local csv. I Iterate for day of the month and iterating for each hour inside of it. I have a class object handling data for each day. The class object is being initialized right after the iteration of each day. The issue is, that this object doesn't get re-initialized after each iteration. I have made som test object below it. The one differens between the test object and the object is, that the object is an object inside of a subfolder containing all my data handling. The object contains a list of data container class objects, whose origin is in the same directory.

-- Main class --

import os 
import csv
from Data.VejrData import *

#Test class
class Ekstra:
    streng: str = ""

class Vejr: 
    currentDirectory: str = os.getcwd()
    dataDirectory: str = 'Data/vejrdata/'
    folder = os.listdir(currentDirectory "/" dataDirectory)
    days: list[VejrData] = []
    
    def __init__(self):
        self.main()
    
    def fetchData(self): 
        for file in self.folder:
            vejrData = VejrData()
            #Error!

            #For each file vejrData should be reset, but doesn't
            #For each file(31 iterations), all rows from the 32 iterations are being added(25 each * 31 = 248000 sets of data) 
            
            #Testing
            string = ""
            string  = "New day"
            print(string)

            #Testing if the same happens with an empty class object. 
            ekstra = Ekstra()
            ekstra.streng  = "New day"
            print(ekstra.streng)
            #This variable from class object does get reset. 
            #VejrData doesn't. 
            
            #Opens datafile from a specifik date. 
            with open(f"{self.dataDirectory}{file}", 'r') as data:      
                csvreader = csv.reader(data)
                number = 0 
                for row in csvreader: 
                    number  = 1
                    index: int = 0 
                    
                    #iterates each element of data in a string. 
                    for textData in row[0].split(";"):
                        vejrData.constructData(index, textData) 
                        index  = 1
            self.days.append(vejrData)

    def main(self): 
        self.fetchData()
        print(len(self.days))
        
        for day in self.days: 
            print(len(day.timeData))
    
vejr = Vejr()

` printing from main function results in: len(self.days) = 32 Length of set of hourly data in each day = 775 for all of the 32 days. 775/25(24 1 header) = 31

-- Data Handling -- `

from .TimeData import TimeData
#https://docs.python.org/3/tutorial/modules.html#intra-package-references

class VejrData: 
    #Lists containing hourly data 
    timeData: list[TimeData] = []
    #Variable retrieving data before being added to list. 
    DataBuilder: None

    #Distributing data from index: index 0 = time data, index 1 = prec data ... 
    def constructData(self,index: int, data): 
        #https://www.freecodecamp.org/news/python-switch-statement-switch-case-example/
        match index:   
            case 0: 
                self.DataBuilder = TimeData()
                self.DataBuilder.tid = data
            case 1: 
                self.DataBuilder.prec = data
            case 2: 
                self.DataBuilder.metp = data
            case 3: 
                self.DataBuilder.megrtp = data
            case 4: 
                self.DataBuilder.mesotp10 = data
            case 5: 
                self.DataBuilder.meanwv = data
                self.timeData.append(self.DataBuilder)
                self.DataBuilder = None

            case _:
                print(f"Error - Index not at index: {index}, is out of range.")

-- Container class -- `

class TimeData: 
    tid: int
    prec: float
    metp: float 
    megrtp: float 
    mesotp10: float
    meanwv: float

The structure is as such /Vejr.py, /Data/VejrData & /Data/TimeData. No errors related to the pathing occurs. I could just give it a new variable at the end of each loop, but that seems off, to be doing what the loop is supposed to do.

I have tried testing whether re-initialization isn't intended to be happening in for loops. I created some objects to see whether or not they would be affected by it. I started with string variable. As the string variable was re-initialized I tried with another class object located in the same file, changed a variable inside of it and saw it re-initialize as well.

So variables and class objects are intended to be re-initialized in each iteration.

CodePudding user response：

...are intended to be re-initialized in each iteration

VejrData.timeData is a class attribute and because it has a mutable default, every instance of VejrData will point-to the exact same list.

>>> v = VejrData() 
>>> w = VejrData()
>>> v.timeData.append('x')
>>> v.timeData
['x']
>>> w.timeData       
['x']
>>> v.timeData is w.timeData
True
>>>

Make timeData an instance attribute.

class VejrData: 
    #Lists containing hourly data 
    # timeData: list[TimeData] = []
    #Variable retrieving data before being added to list. 
    DataBuilder: None

    def __init__(self):
        self.timeData: list[TimeData] = []

>>> v = VejrData()
>>> w = VejrData() 
>>> v.timeData.append('x') 
>>> v.timeData
['x']
>>> w.timeData 
[]
>>> v.timeData is w.timeData
False
>>>

CodePudding user response：

Thanks for the quick response on my question. I was indeed experiencing mutable defaults. After setting my initial value to none and adding it later, I created 32 unique sets of data, instead of one.