Reading JSON from Files with json.load()

From Text to Usable Data

A JSON response may begin as ordinary text, but your Python program usually needs a list, dictionary, or primitive value that it can work with. The json.loads() function parses a JSON string and builds a corresponding Python data structure. The original string remains unchanged, so assign the returned Python object to a variable.

Parsing the Outer Structure

json.loads(data) reads the JSON string and builds a new Python object. The outer square brackets in the JSON text indicate a list. Curly braces inside that list indicate dictionary objects. Colons separate keys from values, while commas separate elements. After parsing a JSON string representing two users, the result is a Python list containing two dictionaries.

import json

data = '[{"id": 1, "x": "chuck.example", "name": "Chuck"}, {"id": 2, "x": "ana.example", "name": "Ana"}]'
info = json.loads(data)

print(type(data))
print(type(info))
print(len(info))
print(info[0]["name"])

<class 'str'>
<class 'list'>
2
Chuck

The variable data still refers to the original JSON string. The variable info refers to the new Python object returned by json.loads(). Because the outer JSON value is a list, len(info) counts the two user dictionaries. The expression info[0] selects the first dictionary, and info[0]["name"] selects the value associated with its name key.

Following Nested Access

first_user = info[0] user_name = first_user["name"] print(user_name)

Knowing the Expected Shape

Parsing JSON does not tell you which path you intended to follow. You need to know the structure in advance: whether the outer value is a list or dictionary, whether a list index is required, and which keys exist in each dictionary. In the user example, you must know that the result is a list of users, that each user is a dictionary, and that the dictionaries contain keys such as id, x, and name.

Tracing Parsing Problems

When parsing or access produces an unexpected result, inspect the process in layers. First verify that the input is a string containing valid JSON. Next confirm the structure of the parsed result. Finally check that every index and key matches that structure. Printing each layer makes it easier to identify where your expectation and the actual data diverged.

print(type(data))
print(type(info))
print(len(info))
print(info[0])
print(info[0]["name"])

• Forgetting to import json

  The name json has not been made available to the program.

  Fix: Import the json module before calling json.loads().

• Passing a non-string object to json.loads()

  json.loads() is used to parse a JSON string, not an already parsed Python object.

  Fix: Pass the JSON string to json.loads(), then assign its returned Python object to a separate variable.

• Using a field that does not exist

  The source structure contains fields such as id, x, and name; accessing an absent field can produce a KeyError.

  Fix: Check the documented structure and use the exact key that exists.

• Skipping the structural trace

  You cannot reliably navigate the value if you do not know whether each layer is a list or dictionary.

  Fix: Print the type and contents of each relevant layer before changing the access expression.

Practice the Transformation

Key Takeaways

1. json.loads() parses a JSON string and returns a new Python object.
2. The original JSON string is unchanged, so store the returned object in a variable.
3. JSON arrays become Python lists in the parsed structure, while JSON objects become Python dictionaries.
4. Navigate lists with indices and dictionaries with keys, following the structure documented for the data.
5. When debugging, inspect the input, parsed type, length, intermediate values, and final key or index access.