Parsing a json (a response from freebase) in python

Question

Parsing a json (a response from freebase) in python This question is regarding JSON parsing. I am querying freebase for some data and getting a JSON object. The response has a following structure

{
  "result": [
    {
      "attribute0": [
        "attrbVal0"
      ],
      "attribute1": [],
      "attribute2": "attrbVal1",
      "attribute3": [
        "val1",
        "val2"
      ],
      "creator": "abc",
      "key": [
        "val2",
        "val3"
      ]
    }
  ]
}

Please note that, the attributes can have zero values to any number of values. When there are no values it is represented as [] or null. The set of attributes is not known to me. It changes as the query changes,So I can't hard code the values like

result['attribute2'];

From the above JSON I want to get the attributes where the value is either [] or null . I have tried following things to get attributes and values,

print response.keys()

prints result

for r in response['result']: 
    print r

This prints everything inside result in one go. That is

print len(result) #prints 1

I have tried the following to get the list of attributes, but no luck.

result = response['result']
elem = json.loads(result);
keys = elem.keys()

So I am looking for the code to get all key value pairs from the above json and some explanation pointing out my mistake.

Answer 1

You can loop over the items of a dictionary, giving you both the keys and the values in pairs; this then lets you filter on the value:

for result in response['result']:
    for attrname, value in result.items():
        if not value:  # empty list or None
            print attrname

Note that response['result'] is a list, containing (presumably) one or more dictionary objects.

In Python both the empty list and None (the Python equivalent of the JSON null ) are considered false in a boolean context, so not value is True for those attributes whose values are either an empty list or null in the original JSON response.

Demo:

>>> response = {
...   "result": [
...     {
...       "attribute0": [
...         "attrbVal0"
...       ],
...       "attribute1": [],
...       "attribute2": "attrbVal1",
...       "attribute3": [
...         "val1",
...         "val2"
...       ],
...       "creator": "abc",
...       "key": [
...         "val2",
...         "val3"
...       ]
...     }
...   ]
... }
>>> for result in response['result']:
...     for attrname, value in result.items():
...         if not value:  # empty list or None
...             print attrname
... 
attribute1

So in your sample input, only attribute1 has an empty value.