Remove all last commas between brackets

Question

I'm trying to format the following as JSON (in Python):

{
   "sessionTimeout":"3600.0",
   "serverVersion":"LK_LK-NL-7_188-176-419",
   "worldDawn":"2018-10-09 06:00:00 Etc/GMT",
   "Data":{
      "Player":[
         {
            "nick":"Player11226400",
            "points":"44",
            "alliancePermission":"0",
            "isOnVacation":"false",
            "id":"5048",
            "rank":"561",
            "entityName":"Player",
         },
         {
            "nick":"Player11230580",
            "points":"15",
            "alliancePermission":"0",
            "isOnVacation":"false",
            "id":"5215",
            "rank":"2081",
            "entityName":"Player",
         },
         {
            "nick":"Player11291581",
            "points":"15",
            "alliancePermission":"0",
            "isOnVacation":"false",
            "id":"5942",
            "rank":"2081",
            "entityName":"Player",
         }
      ]
   }
}

Every JSON validator of course tells me that this is not valid JSON because of the trailing "," before each }, so I want to remove that comma. I tried removing them with .replace('"Player",", '"Player"') but I don't consider this as a good solution. This because I also have trailing "," in for example a Alliance or Habitat string ("Habitat", & "Alliance",)

Could anyone help me with finding a better solution to this problem?

Answer 1

Two solutions:

First one, cool if your json file doesn't have null or false/true booleans would be to read the input using ast.literal_eval which can process those commas, then dump back the dict as json if needed:

d = ast.literal_eval(txt)
print(json.dumps(d,indent=4))

if you can't, you can use a regex to remove commas that happen before a newline if the next line starts with spaces + closing braces/brackets:

import re

print(json.loads(re.sub(",(\n\s+[\}\]])",r"\1",txt)))

here's a full snippet containing the 2 approaches with a cut-down version of your input:

import ast,json

txt = """{
   "sessionTimeout":"3600.0",
   "serverVersion":"LK_LK-NL-7_188-176-419",
   "worldDawn":"2018-10-09 06:00:00 Etc/GMT",
   "Data":{
      "Player":[
         {
            "nick":"Player11226400",
            "rank":"561",
            "entityName":"Player",
         },
         {
            "nick":"Player11230580",
            "rank":"2081",
            "entityName":"Player",
         },
         {
            "nick":"Player11291581",
            "rank":"2081",
            "entityName":"Player",
         }
      ]
   }
}"""

print("ast literal eval:")
d = ast.literal_eval(txt)
print(json.dumps(d,indent=4))

import re

print("regex:")
d = json.loads(re.sub(",(\n\s+[\}\]])",r"\1",txt))
print(json.dumps(d,indent=4))

both methods successfully parse your input file. The first method will work flawlessly as long as there are no null pointers & booleans, the second method may fail in some obscure formatting cases. In case of bad luck, you can hack something in between.

Answer 2

Since JSON is a subset of YAML, where dangling commas are permitted, you can use a yaml parser here.

pip install pyyaml

The library has a load function, similar to the standard library json.loads.

import json, yaml
json.dumps(yaml.load(jsondata))

You can also use ast.literal_eval for this particular data. But unlike pyyaml that doesn't work if your json contains literal false, true or null values.