Reading Structured JSON Responses

No parsing required — it's already structured

"Running Show Commands with Netmiko" made a point of this: send_command()'s output is raw CLI text, and turning it into a usable fact requires writing a parser (is_ntp_synced(), is_vlan_present()). A JSON API response skips that step entirely — the structure is already there.

CLI text (Netmiko)

show ip interface briefraw text output
Write a parseris_ntp_synced(), is_vlan_present()
Now it's usable data

JSON API response

GET /devices/SW1/interfacesreturns JSON
response.json()already a dict/list
Usable data immediately

Quick checkConcept check

Why doesn't a JSON API response need a custom parser the way send_command()'s output did?

It does need one — JSON always requires custom parsing tooBecause JSON is already structured data — response.json() turns it directly into Python dicts/lists, with no text-scanning step in betweenBecause APIs never return useful dataBecause JSON only works for interface data specifically

response.json(), directly

parse_response.py

import requests

response = requests.get("http://127.0.0.1:8765/devices/SW1/interfaces")
data = response.json()

print(type(data))
print(data)

response.json() returns real Python data — dicts and lists, exactly the shapes 'Working with JSON in Python' and 'Lists & Dictionaries for Network Data' already taught.

output

<class 'dict'>
{'hostname': 'SW1', 'interfaces': [{'name': 'GigabitEthernet1/0/1', 'status': 'up'}, {'name': 'GigabitEthernet1/0/2', 'status': 'down'}]}

A dict with a hostname key and an interfaces key holding a list of dicts — the exact 'list of dictionaries' shape this curriculum has used since Python Lesson 3, arriving over HTTP instead of being built by hand.

Quick checkPredict output

Given the parsed data above, what does data["interfaces"][1]["status"] evaluate to?

"up""down"GigabitEthernet1/0/2An error — interfaces only has one item

Extracting the field you actually need

extract_field.py

response = requests.get("http://127.0.0.1:8765/devices/SW1/interfaces")
data = response.json()

for interface in data["interfaces"]:
  print(interface["name"], "-", interface["status"])

Looping over data['interfaces'] is exactly the same for-loop pattern used on every list of dicts since Python Lesson 4 — devices.csv rows, VLAN lists, and now API response items all follow the same shape.

output

GigabitEthernet1/0/1 - up
GigabitEthernet1/0/2 - down

Real, verified output from this module's mock API.

Quick checkIdentify the correct code

Which line correctly prints just the names of every 'up' interface from the parsed response?

for interface in data["interfaces"]: if interface["status"] == "up": print(interface["name"])for interface in data: print(interface["name"])print(data["interfaces"]["name"])if data["status"] == "up": print(data["name"])

The response's structure is still just a claim

Parsed successfully is not the same as correct

response.json() will happily parse whatever JSON the server actually sent — including a shape that's missing a field your code expects. A KeyError from data["interfaces"] on a response that doesn't have that key is a real, common bug — not a sign that JSON parsing itself is unreliable. This is the same theme Configuration & Template Automation raised about rendered templates: successfully processing something is never the same claim as the something being correct.

Quick checkScenario

A script calls response.json() and then reads data["interfaces"], but the API actually returned {"error": "device not found"} — a 404 response with a different shape entirely. What happens?

response.json() raises an error immediately, before any field accessresponse.json() parses successfully (it's valid JSON), but then data["interfaces"] raises a KeyError, since that key doesn't exist in this particular responsePython silently returns an empty list for the missing keyThe script automatically detects the 404 and skips the field access

Key takeaways

response.json() turns an HTTP response body directly into Python dicts/lists — no custom parser needed, unlike raw CLI text from send_command().
The resulting shape is the exact 'list of dictionaries' pattern this curriculum has used since Python Lesson 3 — looping, indexing, and key access all work exactly as already learned.
A response parsing successfully only proves it was valid JSON — not that it has the fields your code expects. Checking the status code first (Lesson 5) is what actually protects against this.

What you'll learn

No parsing required — it's already structured

response.json(), directly

Extracting the field you actually need

The response's structure is still just a claim