API-shaped data

Decode a response; do not confuse absence with success

These are mock JSON strings, not HTTP requests. json.loads converts JSON into Python data; parsing alone does not validate the response shape. The trace tasks assume a JSON object with status text and a list of records containing item_id. Test status and each record before using a real service.

dictionary.get("data", []) supplies a fallback only when the key is missing. A present null becomes None, not an empty list. The change task assumes data, when present, is a list. The extract_records filter returns a list for status ok and [] for non-ok envelopes; [] alone cannot tell a failed request from a successful empty result. Keep the original status/error evidence when integrating a service.

The independent helper validates its text/object/list boundary, leaves malformed JSON errors visible and starts fresh per call. The complete example keeps status beside the list so the caller can distinguish error from empty success. Neither local code nor tutor login sends an automatic API request.

# Preserve response status beside the typed records.
import json

def decode_response(text):
    response = json.loads(text)
    if not isinstance(response, dict):
        raise ValueError("Expected a JSON object")
    status = response.get("status")
    rows = response.get("data", [])
    if not isinstance(rows, list):
        raise ValueError("Expected a data list")
    return status, rows

for text in [
    '{"status":"ok","data":[]}',
    '{"status":"error","message":"timeout"}',
]:
    status, rows = decode_response(text)
    print(status, len(rows))

Refresh JSON parsing and typed validation on Days 21 and 24. A small extraction filter is not a network reliability policy.

Official reference for this lesson