如何将类似字典的字符串转换为字典？

Question

我有一个类似字典的字符串，如下所示：

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"

这是以相同的格式进行的，它看起来像：

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

这需要更改为：

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"}
                    }
      }

我试过了ast.literal_eval, eval，结果却出错了

Answer 1

您的字符串几乎是yaml-将\t替换为一个或多个空格，它将作为yaml加载

第一:

pip install pyyaml

那么这个代码就起作用了：

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 )

输出：

{'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'}}

或者，如果您确实希望将其作为字符串，则：

str = repr(res)

更新

刚刚意识到例如“Managent Name:csw_1”没有被检测为key:value。需要一个regex re.sub()将它们拆分为单独的行以修复此问题：

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 )

这是调整后的字符串：

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

结果:

{'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}}}

Answer 2

您好，这是您需要的解决方案

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)    

Answer 3

您可以使用递归将输入转换为字典，而无需事先对字符串本身进行任何调整：

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))

输出：

{
    "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"
        }
    }
}