使用 python 中的一个 csv 文件比较一列中的两个数据

Question

I am trying to compare two data in one csv file and I cannot use panda. What I am trying to get is the total Unit sold that the two person sell and the sum of all the years then compare who sold more based on the sum of all they sold through out the years. Then also get the least they sold on that particular year.

For example, my.csv is setup like this:
John Smith, 343, 2020
John Smith, 522, 2019
John Smith, 248, 2018
Sherwin Cooper, 412, 2020
Sherwin Cooper, 367, 2019
Sherwin Cooper, 97, 2018
Dorothy Lee, 612, 2020
Dorothy Lee, 687, 2019
Dorothy Lee, 591, 2018

I want to compare John and Dorothy's unit sold and who sold more. So the output should be:
Dorothy Lee sold more units than John smith. A total of 1890 to 1113.
Dorothy Lee sold less in 2018, for only 591.
John Smith sold less in 2018, for only 248.

My code so far is:

import csv

def compare1(employee1):

    with open("employeedata.csv") as file:
    rows = list(csv.DictReader(file, fieldnames = ['c1', 'c2', 'c3']))

    res = {}

       for row in rows:
       if row['c1'] == employee1:
          res[employee1] = res.get(employee1, 0) + int(row['c2'])
        
       print(res)
        
def compare2(employee2):

   with open("employee2.csv") as file:
      rows = list(csv.DictReader(file, fieldnames = ['c1', 'c2', 'c3']))

   res = {}

   for row in rows:
      if row['c1'] == employee2:
         res[employee2] = res.get(employee2, 0) + int(row['c2'])
        
   print(res)

employee1 = input("Enter the first name: ")
employee2 = input("Enter the first name: ")


compare1(employee1)
compare2(employee2)

I don't know the rest. I am stuck. I am a beginner and I can't use Panda. The output I need to have should look like this:

Dorothy Lee sold more units than John smith. A total of 1890 to 1113.
Dorothy Lee sold less in 2018, for only 591.
John Smith sold less in 2018, for only 248.
right now I got the output:
{'John Smith: 1113}
{'Dorothy Lee': 1890}

Answer 1

Suppose my.csv has columns name, sales, year :

import pandas as pd

emp_df = pd.read_csv("my.csv")

emp_gp = emp_df.groupby("name").sales.sum().reset_index(inplace=True)


def compare(saler1, saler2):
    if saler1 in emp_pg.name.values and saler2 in emp_pg.name.values:
         saler1_tol = emp_pg.loc[emp_pg.name == saler1, ["sales"]]
         saler2_tol = emp_pg.loc[emp_pg.name == saler2, ["sales"]]
         if saler1_tol > saler2_tol:
             print(f"{saler1} sold more unit than {saler2}. A total {saler1_tol} to {saler1_tol}")
         else:
             print(f"{saler2} sold more unit than {saler1}. A total {saler2_tol} to {saler2_tol}")
         emp_gb2 = emp_df.groupby("name")
         emp_agg = emp_gb2.agg({
              "sales" : "min"
         })
         emp_agg = emp_agg.reset_index()
         print("{saler1} sold less in {emp_pg.loc[emp_pg.name == saler1, ["year"]].values}, for only {emp_pg.loc[emp_pg.name == saler1, ["sales"]].values}")
         print("{saler2} sold less in {emp_pg.loc[emp_pg.name == saler2, ["year"]].values}, for only {emp_pg.loc[emp_pg.name == saler2, ["sales"]].values}")
    else:
        print("names of salers are not in the table")

Answer 2

Instead of creating a function for each result you want to get, first create a database (a dict is OK) that aggregates the sum of units sold for each name and for each year. Then it is easier to answer to all kind of comparisons without having to repeat code. You can start with something like this,

import csv
from collections import defaultdict

db=defaultdict(lambda: defaultdict(int))

with open('teste.csv', newline='') as csvfile:
    reader = csv.DictReader(csvfile)     
    for row in reader:
        db[row['name']][int(row['year'])]+=int(row['units'])

print(db['Dorothy Lee'][2019]) #Units sold by Dorothy Lee in 2019
print(sum(db['Dorothy Lee'].values())) #Total Units sold by Dorothy Lee

Don't be afraid of the defaultdict module. Check the docs , it is really handy in this kind of scenario. The defaultdict creates a dictionary with a default for every missing key. In this case, the default value of the first defaultdict is another defaultdict, this time with a default value of 0 (the result of calling int()), since we want to compute a sum of units sold (therefore an integer). With this approach, you don't need to check if the key already exists or not, defaultdict takes care of that for you.

PS: the lambda in the first defaultdict is needed to nest a second defaultdict. If you are not familiar with lambda either, check this