Extract two columns sorted from CSV

I have a large csv file, containing multiple values, in the form

Date,Dslam_Name,Card,Port,Ani,DownStream,UpStream,Status
2020-01-03 07:10:01,aart-m1-m1,204,57,302xxxxxxxxx,0,0,down

I want to extract the Dslam_Name and Ani values, sort them by Dslam_name and write them to a new csv in two different columns.

So far my code is as follows:

import csv
import operator

with open('bad_voice_ports.csv') as csvfile:
    readCSV = csv.reader(csvfile, delimiter=',')
    sortedlist = sorted(readCSV, key=operator.itemgetter(1)) 
    for row in sortedlist:
        bad_port = row[1][:4],row[4][2::]
        print(bad_port)
        f = open("bad_voice_portsnew20200103SORTED.csv","a+")
        f.write(row[1][:4] + " " + row[4][2::] + '\n')
        f.close() 

But my Dslam_Name and Ani values are kept in the same column.

As a next step I would like to count how many times the same value appears in the 1st column.

You are forcing them to be a single column. Joining the two into a single string means Python no longer regards them as separate.

But try this instead:

import csv
import operator

with open('bad_voice_ports.csv') as readfile, open('bad_voice_portsnew20200103SORTED.csv', 'w') as writefile:
    readCSV = csv.reader(readfile)
    writeCSV = csv.writer(writefile)
    for row in sorted(readCSV, key=operator.itemgetter(1)):
        bad_port = row[1][:4],row[4][2::]
        print(bad_port)
        writeCSV.writerow(bad_port)

If you want to include the number of times each key occurred, you can easily include that in the program, too. I would refactor slightly to separate the reading and the writing.

import csv
import operator
from collections import Counter

with open('bad_voice_ports.csv') as readfile:
    readCSV = csv.reader(readfile)
    rows = []
    counts = Counter()
    for row in readCSV:
        rows.append([row[1][:4], row[4][2::]])
        counts[row[1][:4]] += 1

with open('bad_voice_portsnew20200103SORTED.csv', 'w') as writefile:
    writeCSV = csv.writer(writefile)
    for row in sorted(rows):
        print(row)
        writeCSV.writerow([counts[row[0]]] + row)

I would recommend to remove the header line from the CSV file entirely; throwing away (or separating out and prepending back) the first line should be an easy change if you want to keep it.
(Also, hard-coding input and output file names is problematic; maybe have the program read them from sys.argv[1:] instead.)

So my suggestion is failry simple. As i stated in a previous comment there is good documentation on CSV read and write in python here: https://realpython.com/python-csv/

As per an example, to read from a csv the columns you need you can simply do this:

>>> file = open('some.csv', mode='r')
>>> csv_reader = csv.DictReader(file)
>>> for line in csv_reader:
...     print(line["Dslam_Name"] + " " + line["Ani"])
...

This would return:

aart-m1-m1 302xxxxxxxxx

Now you can just as easilly create a variable and store the column values there and later write them to a file or just open up a new file wile reading lines and writing the column values in there. I hope this helps you.

After the help from @tripleee and @marxmacher my final code is

import csv
import operator
from collections import Counter

with open('bad_voice_ports.csv') as csv_file:
    readCSV = csv.reader(csv_file, delimiter=',')
    sortedlist = sorted(readCSV, key=operator.itemgetter(1))
    line_count = 0
    rows = []
    counts = Counter()
    for row in sortedlist:
        Dslam = row[1][:4]
        Ani = row[4][2:]
        if line_count == 0:
            print(row[1], row[4])
            line_count += 1
        else:
            rows.append([row[1][:4], row[4][2::]])
            counts[row[1][:4]] += 1
            print(Dslam, Ani)
            line_count += 1
    for row in sorted(rows):
        f = open("bad_voice_portsnew202001061917.xls","a+")
        f.write(row[0] + '\t' + row[1] + '\t' + str(counts[row[0]]) + '\n')
        f.close()
    print('Total of Bad ports =', str(line_count-1))

As with this way the desired values/columns are extracted from the initial csv file and a new xls file is generated with the desired values stored in different columns and the total values per key are counted, along with the total of entries.

Thanks for all the help, please feel free for any improvement suggestions!

You can use sorted:

import csv
_h, *data = csv.reader(open('filename.csv'))
with open('new_csv.csv', 'w') as f:
  write = csv.writer(f)
  csv.writerows([_h, *sorted([(i[1], i[4]) for i in data], key=lambda x:x[0])])