string manipulation python

Question

I am having a problem regarding python b (Backspace), instead of deleting the previous character, it generates a "•" as the image below:

Follow the line of the code that generates this return:

     #-*- coding: utf-8 -*-
import pyodbc
import platform
import json
#------------------------------------------------------#
local_arq = open("C:\Zabbix\prm2.txt", 'r') #Caminho do arquivo cujo encontra-se aa instancias
leitura = local_arq.readlines() #leitura das linhas do arquivo
varhost = platform.node()
x = 0 #var usada para separar os dados recebidos do TXT && Alterar o valor do hostname 
y = 0 
with open("test3.txt","w") as arq:
    print('{"data": [')
    for li in leitura:
        count = li.split('--') #count = contador ele recebe os dados do TXT e separa eles onde tem o -
        count[x].replace(" ", "") #Remove todos os espaços

        for li2 in count:   
            hostname = varhost + '\\' + count[x]
            con = pyodbc.connect("DRIVER={SQL Server};server="+hostname+";uid=sa;pwd=Ww1234567.")
            query = con.cursor()
            query.execute("SELECT name FROM SYSDATABASES")

            for row in query:    
                #print(row)
                print('{"{#INSTANCENAME}": "%s","{#DBNAME}": "%s"},' %(count[x], row[y]))
                lol = ('{"{#INSTANCENAME}": "%s","{#DBNAME}": "%s"},' %(count[x], row[y]))
            x = x + 1
            print(lol + '\b]}')
        y = y + 1

In case, I’m using b to erase that comma.

Thanks in advance!

NOTE: Python version 3.7

Answer 1

There is a concept error there - the " b" is simply a control character, named "Backspace", which has decimal code "8" - when printed in a terminal , the terminal protocol is what causes the previous character to be deleted.

But to write to a file, the " b" is a character like any other - in most text encodings - ASCII, Latin1 and UTF-8 inclusive, will simply generate a byte with value "8" that will stay in the file. And then, if the file is printed, in a terminal, it will look like the previous character has been deleted - try there, instead of opening your file in a text editor, to display its contents (command cat on Linux/Unix/OS X, or type on DOS/windows).

That is, what you need to do is not write the unwanted comma to the file to begin with. You have not added more parts to your program, so you can’t tell exactly where you can do this, but in general it will require you to have a vailable "counter", and record the comma at the beginning of each item, if that item nay is the first.

Or, if you have the entire list of items when recording, the method .join of the lists in Python already puts the commas only as separators of the items. So, let’s assume you have everything you want to write to the file in a "values" list, you can do it like this:

arq.write("""{"chave":[%s]}""" % ", ".join(valores))

(Here I used the old way % to interpolate strings to avoid having to escape the dictionary keys from the outside, using .format or f-strings)

An even better alternative, if you are creating a JSON file is to let Python transform a "live" data structure, composed of dictionaries and lists into the JSON string.

import json
arq.write(json.dumps(meus_dados))

where "meus_data" is a dictionary, with the necessary lists and sub-dictionaries. Using JSON Encoder is much better than trying to generate the string manually, because it accounts for all the other cases where JSON would error, such as quotes inside the strings, very large numbers, Python "None" values, etc...

In short:

• Use the json.dumps to create your JSON string - do not try code it manually.
• The " b" doesn’t even work because it na truth prints a character the most. The terminal protocol is that uses this character as a control code to return a position.

Answer 2

There are two suggestions for you to use, the first is to save straight to the file without the "," which would make things easier. If it is not possible, the second is to make a "copy" of each line and simply cut their "," thus:

with open('dbname.txt', 'r+') as Arq:
    for line in Arq:
        # .split() corta a linha do arquivo exatamente no caracter ',' e forma uma lista,
        # que no seu caso tem duas partes, antes e após a virgula.
        novalinha = line.split(',')
        # Refaz tudo em outro arquivo, de maneira correta.
        with open('tmp.txt', 'w+') as NewArq:
            NewArq.write(novalinha[0] + novalinha[1])

So each line would be the same, except that you would delete the "," from them, then just pass all the lines to the original file again.