Nikita Grigorev
Nikita Grigorev

Reputation: 23

Parse JSON array in Python

I have some JSON file:

{
  "cis" : [ {
    "ucmdbId" : "835cfedfaabc32a1358b322ff3bae056",
    "type" : "running_software",
    "properties" : {
      "display_label" : "jboss (site1.ru)"
    }
  }, {
    "ucmdbId" : "7ef9f21c132c12b3d8d2af0964cc5970",
    "type" : "node",
    "properties" : {
      "display_label" : "site2.ru"
    }
  } ],
  "relations" : [ {
    "ucmdbId" : "80c42edbe32fbb4c25621756ec9e09d2",
    "type" : "compound_f",
    "properties" : null,
    "end1Id" : "23e30baf2320a3274d0aa1e7f56cdaef",
    "end2Id" : "15af0ba134327d32a0c5c72450e63fcd"
  }, {
    "ucmdbId" : "7fe9fb15d4462d1212aeee4aef2f32b4",
    "type" : "compound_f",
    "properties" : null,
    "end1Id" : "23e30baf2320a3274d0aa327f56cdaef",
    "end2Id" : "9232dd2621b814da632932e8cd33ffc8"
  } ]
}

I only need the cis array. So this is what I need to parse:

[{
  "ucmdbId" : "835cfedfaabc32a1358b322ff3bae056",
  "type" : "running_software",
  "display_label" : "jboss (site1.ru)"
 }, {
  "ucmdbId" : "7ef9f21c132c12b3d8d2af0964cc5970",
  "type" : "node",
  "display_label" : "site2.ru"
}]

Python script:

#!/usr/bin/python
import sys
import os
import tablib
import pandas as pd
import json
from pandas.io.json import json_normalize

f = open('/home/nik/test.json', 'rw')
jsonArray = f.read()
f.close
data = json.dumps(json.loads(jsonArray)['cis'])
jsonResult = pd.read_json(data)
array = json.loads(jsonArray)

print jsonArray
jsonResult.to_excel('/home/nik/output.xlsx', sheet_name='Sheet1')

But how can I get key parameters? I try to use:

print data['type'].keys()
print data['type']

But it gives me error:

AttributeError: 'str' object has no attribute 'keys'

How can I get the proper JSON format?

Update. Solution:

Thanks, it works. My complete code to export JSON into xlsx file:

#!/usr/bin/python
import subprocess
import sys
import os
import tablib
import pandas as pd
import json
import glob
import string

path = '/home/nik/json'
for jsonfile in glob.glob(os.path.join(path, '*.json')):
#jsonfile = '/home/nik/test.json'
    with open(jsonfile) as data_file:
        data = json.load(data_file)

    JSON = '[{ \n'
    for index, item in enumerate(data['cis']):
        ucmdbId = (item['ucmdbId'])
        type = (item['type'])
        display_label = (item['properties']['display_label'])
        Text1 = '  \"ucmdbId\" : \"%s\",' %(ucmdbId)
        Text2 = '  \"type\" : \"%s\",' %(type)
        Text3 = '  \"display_label\" : \"%s\",' %(display_label)
        if index==(len(data['cis'])-1):
            End = '}]'
        else:
            End = '}, {'
        JSON += Text3+'\n'+Text2+'\n'+Text1+'\n'+End+'\n'

    JSON = JSON.translate({ord(c): None for c in '\/'})
    jsonResult = pd.read_json(JSON)
    jsonResult = jsonResult.sort_values(by='type')
    jsonResult.to_excel(jsonfile+'.xlsx', sheet_name='Object monitoring', index=False)

Upvotes: 2

Views: 11669

Answers (1)

Gajendra D Ambi
Gajendra D Ambi

Reputation: 4233

import json
from pprint import pprint
jsonfile = 'C:\\temp\\temp.json' # path to your json file
with open(jsonfile) as data_file:    
    data = json.load(data_file)
pprint(data['cis'])

The above will give you just the cis array. Below is a more granular output

for item in data['cis']:
    ucmdbId = (item['ucmdbId'])
    type = (item['type'])
    display_label = (item['properties']['display_label'])
    print(ucmdbId)
    print(type)
    print(display_label)

If you want it with key labels then use

for item in data['cis']:
    ucmdbId = (item['ucmdbId'])
    type = (item['type'])
    display_label = (item['properties']['display_label'])
    print('ucmdbId:{}'.format(ucmdbId))
    print('type:{}'.format(type))
    print('display_label:{}'.format(display_label))

Upvotes: 1

Related Questions