CODEWITHSUNDEEP
pythonjsondictionary
nonuser
nonuser

Reputation: 103

How can I convert a dictionary-like string to a dictionary?

I have a dictionary like string as below:

str = "Access AR1:\n\tTargets: \n\t\tManagement Name:csw_1\n\t\tObject Name:csw_obj_1\n\t\tdetails:103\n\t\tManagement Name:csw_123\n\t\tObject Name:csw_obj_134\n\t\tdetails:123\n\tSources: \n\t\tIP:10.20.30.40\n\t\tSubnet Mask:255.255.255.255\nAccess AR2:\n\tTargets: \n\t\tManagement Name:csw_2\n\t\tObject Name:csw_obj_2\n\t\tdetails:110\n\t\tManagement Name:csw_431\n\t\tObject Name:csw_obj_21\n\t\tdetails:134\n\tSources: \n\t\tIP:10.20.10.10\n\t\tSubnet Mask:255.255.255.192"

and this goes on in the same format, which would look like:

Access AR1:
    Targets: 
            Management Name:csw_1
            Object Name:csw_obj_1
            details:103
            Management Name:csw_123
            Object Name:csw_obj_134
            details:123
    Sources: 
            IP:10.20.30.40
            Subnet Mask:255.255.255.255
Access AR2:
    Targets: 
            Management Name:csw_2
            Object Name:csw_obj_2
            details:110
            Management Name:csw_431
            Object Name:csw_obj_21
            details:134
    Sources: 
            IP:10.20.10.10
            Subnet Mask:255.255.255.192

and this need to be changed to:

str = {"Access AR1": { "Targets": [{"Management Name:csw_1", "Object Name":"csw_obj_1", "details":"103"}, {"Management Name:csw_123", "Object Name":"csw_obj_134", "details":"123"}]
                      "Sources": {"IP":"10.20.30.40", "Subnet Mask": "255.255.255.255"}
                    },
      "Access AR2": { "Targets": [{"Management Name:csw_2", "Object Name":"csw_obj_2", "details":"110"}, {"Management Name:csw_431", "Object Name":"csw_obj_21", "details":"133"}]
                      "Sources": {"IP":"10.20.10.10", "Subnet Mask": "255.255.255.192"}
                    }
      }

I tried ast.literal_eval, eval and ended up with an error.

Upvotes: 7

Views: 391

Answers (3)

Ajax1234
Ajax1234

Reputation: 71451

You can use recursion to convert the input to a dictionary, without having to make any adjustments to the string itself beforehand:

import re
s = "Access AR1:\n\tTargets: \n\t\tManagement Name:csw_1\n\t\tObject Name:csw_obj_1\n\t\tdetails:103\n\tSources: \n\t\tIP:10.20.30.40\n\t\tSubnet Mask:255.255.255.255\nAccess AR2:\n\tTargets: \n\t\tManagement Name:csw_2\n\t\tObject Name:csw_obj_2\n\t\tdetails:110\n\tSources: \n\t\tIP:10.20.10.10\n\t\tSubnet Mask:255.255.255.192"
def to_dict(d):
   k, v, r = None, [], {}
   for *b, a in d:
     if not b:
        if k is not None:
           if not v:
              r[j[0]] = (j:=re.split(':\s*', k))[-1]
           else:
              r[re.split(':\s*', k)[0]] = to_dict(v)
        k, v = a, []
     else:
         v.append([*b[1:], a])
   if k is not None:
       if not v:
          r[j[0]] = (j:=re.split(':\s*', k))[-1]
       else:
          r[re.split(':\s*', k)[0]] = to_dict(v)
   return r

import json       
new_s = [re.findall('\t|[^\t]+$', i) for i in s.split('\n')]
print(json.dumps(to_dict(new_s), indent=4))

Output:

{
    "Access AR1": {
        "Targets": {
            "Management Name": "csw_1",
            "Object Name": "csw_obj_1",
            "details": "103"
        },
        "Sources": {
            "IP": "10.20.30.40",
            "Subnet Mask": "255.255.255.255"
        }
    },
    "Access AR2": {
        "Targets": {
            "Management Name": "csw_2",
            "Object Name": "csw_obj_2",
            "details": "110"
        },
        "Sources": {
            "IP": "10.20.10.10",
            "Subnet Mask": "255.255.255.192"
        }
    }
}

Upvotes: 1

DisappointedByUnaccountableMod
DisappointedByUnaccountableMod

Reputation: 6826

Your string is nearly yaml - replace the \t with one or more spaces and it loads as YAML.

First:

pip install pyyaml

Then this code works:

import yaml
import pprint

str = "Access AR1:\n\tTargets: \n\t\tManagement Name:csw_1\n\t\tObject Name:csw_obj_1\n\t\tdetails:103\n\tSources: \n\t\tIP:10.20.30.40\n\t\tSubnet Mask:255.255.255.255\nAccess AR2:\n\tTargets: \n\t\tManagement Name:csw_2\n\t\tObject Name:csw_obj_2\n\t\tdetails:110\n\tSources: \n\t\tIP:10.20.10.10\n\t\tSubnet Mask:255.255.255.192"

str1 = str.replace( '\t', '    ' )

res = yaml.load(str1)

pprint.pprint( res )

Output:

{'Access AR1': {'Sources': 'IP:10.20.30.40 Subnet Mask:255.255.255.255',
                'Targets': 'Management Name:csw_1 Object Name:csw_obj_1 '
                           'details:103'},
 'Access AR2': {'Sources': 'IP:10.20.10.10 Subnet Mask:255.255.255.192',
                'Targets': 'Management Name:csw_2 Object Name:csw_obj_2 '
                           'details:110'}}

Or, if you really want this as a string then:

str = repr(res)

UPDATE

Just realised that e.g. 'Managent Name:csw_1' isn't being detected as a key:value. Needed a regex re.sub() to split these onto separate lines to fix this:

import yaml
import pprint
import re

str = "Access AR1:\n\tTargets: \n\t\tManagement Name:csw_1\n\t\tObject Name:csw_obj_1\n\t\tdetails:103\n\tSources: \n\t\tIP:10.20.30.40\n\t\tSubnet Mask:255.255.255.255\nAccess AR2:\n\tTargets: \n\t\tManagement Name:csw_2\n\t\tObject Name:csw_obj_2\n\t\tdetails:110\n\tSources: \n\t\tIP:10.20.10.10\n\t\tSubnet Mask:255.255.255.192"

# replace \t with four-space indent
str1 = str.replace( '\t', '    ' )

# further tweak to split sub-keys like " Management Name:csw_1" onto separate lines
str1 = re.sub(r"^(\s+)(.*?\S:)(\S.*)", r"\1\2\n\1    \3",str1,flags=re.MULTILINE )
   
res = yaml.load(str1)

pprint.pprint( res )

This is the string after tweaks:

Access AR1:
    Targets:
        Management Name:
            csw_1
        Object Name:
            csw_obj_1
        details:
            103
    Sources:
        IP:
            10.20.30.40
        Subnet Mask:
            255.255.255.255
Access AR2:
    Targets:
        Management Name:
            csw_2
        Object Name:
            csw_obj_2
        details:
            110
    Sources:
        IP:
            10.20.10.10
        Subnet Mask:
            255.255.255.192

Result:

{'Access AR1': {'Sources': {'IP': '10.20.30.40',
                            'Subnet Mask': '255.255.255.255'},
                'Targets': {'Management Name': 'csw_1',
                            'Object Name': 'csw_obj_1',
                            'details': 103}},
 'Access AR2': {'Sources': {'IP': '10.20.10.10',
                            'Subnet Mask': '255.255.255.192'},
                'Targets': {'Management Name': 'csw_2',
                            'Object Name': 'csw_obj_2',
                            'details': 110}}}

Upvotes: 8

Bahae El Hmimdi
Bahae El Hmimdi

Reputation: 368

Hello this is the solution you need

strr = "Access AR1:\n\tTargets: \n\t\tManagement Name:csw_1\n\t\tObject Name:csw_obj_1\n\t\tdetails:103\n\tSources: \n\t\tIP:10.20.30.40\n\t\tSubnet Mask:255.255.255.255\nAccess AR2:\n\tTargets: \n\t\tManagement Name:csw_2\n\t\tObject Name:csw_obj_2\n\t\tdetails:110\n\tSources: \n\t\tIP:10.20.10.10\n\t\tSubnet Mask:255.255.255.192"
nvstr=strr.replace("\n\t\t","-").replace("\n\t","+")
#print(nvstr)
nvdd={}
for u in nvstr.split("\n"):
   #print(u) 
   dts=u.split("+") 
   
   nvdd[dts[0]]={}
   
   for el in dts[1:]:
       
     dts1=el.split("-")
     nvdd[dts[0]][dts1[0][:-1]]={}
     for el1 in dts1[1:]:
         
       k,v=el1.split(":")
       nvdd[dts[0]][dts1[0][:-1]][k]=v
       
print(nvdd)

Upvotes: 2

Related Questions