Curious builder0 XP earned · 300 to level 2

0 daysFinish a lesson to begin

Badge collection0 of 6 unlocked

33 small wins to finish your path

The ticket triage program, end to end

Lesson 0 promised a program that reads support tickets, asks a model to sort each one, checks every answer and saves the results. Here it is.

Make a folder with a virtual environment, as in lesson 29, and install pydantic and pytest into it. Put tickets.json from lesson 16 in the folder. Then write triage.py in four pieces.

The shape of an answer

Exampletriage.py, part 1

import asyncio
import json
from typing import Literal

from pydantic import BaseModel, Field, ValidationError


class Triage(BaseModel):
    category: Literal["billing", "shipping", "other"]
    priority: int = Field(ge=1, le=5)

The imports, then the Triage model from lesson 24: three allowed categories and a priority from 1 to 5.

The model

Exampletriage.py, part 2

async def ask_model(text: str) -> str:
    await asyncio.sleep(0.5)
    text = text.lower()
    if "charged" in text or "refund" in text:
        return '{"category": "billing", "priority": 4}'
    if "parcel" in text or "arrived" in text:
        return '{"category": "shipping", "priority": 3}'
    return "I am not sure how to sort this one."

The stand-in from lesson 17, now written with async def and a half-second wait, as in lesson 26, so it behaves like a call over the network.

One ticket

Exampletriage.py, part 3

async def triage(ticket: dict) -> dict:
    answer = await ask_model(ticket["text"])
    try:
        result = Triage.model_validate_json(answer)
    except ValidationError:
        return {"id": ticket["id"], "category": None, "priority": None, "checked": False}
    return {"id": ticket["id"], **result.model_dump(), "checked": True}

Ask, then check with model_validate_json from lesson 25. An answer that fails the check is not guessed at: the row records checked: False and no category, so nothing downstream treats it as sorted.

**result.model_dump() inside the dictionary copies every field of the checked answer into it. It is the same pair of stars as **settings in lesson 12, used to spread a dictionary out instead of collecting one.

Every ticket

Exampletriage.py, part 4

async def main():
    with open("tickets.json") as f:
        tickets = json.load(f)

    results = await asyncio.gather(*[triage(ticket) for ticket in tickets])

    with open("results.json", "w") as f:
        json.dump(results, f, indent=2)

    unchecked = [row["id"] for row in results if not row["checked"]]
    print(f"Sorted {len(results)} tickets. For a person: {unchecked}")


if __name__ == "__main__":
    asyncio.run(main())

Read the file (lesson 16), sort every ticket at once with gather (lesson 27), write the results, and list the ids a person should look at.

if __name__ == "__main__": is true only when the file is run directly with python triage.py. When the tests import triage, main does not run, so importing the file does not start sorting tickets.

Run it

Example

python triage.py
cat results.json

All five tickets in about half a second. Ticket 4, the password question, is the one the model could not sort, and it is listed for a person instead of being filed as something it is not.

Prove it

Exampletest_triage.py

import asyncio

from triage import triage


def test_billing_ticket_is_checked():
    row = asyncio.run(triage({"id": 1, "text": "I was charged twice"}))
    assert row["category"] == "billing"
    assert row["checked"] is True


def test_unclear_ticket_goes_to_a_person():
    row = asyncio.run(triage({"id": 4, "text": "How do I change my password?"}))
    assert row["checked"] is False
    assert row["category"] is None

One test for a ticket that sorts, one for the failure. asyncio.run calls the async triage from an ordinary test function.

Example

pytest -q

Where each piece came from

triage.py, by lesson

Things to add

Try it yourself

Wrap the call in ask_with_retries from lesson 28, and give the stand-in a FlakyModel-style failure.
Add a semaphore so no more than two tickets are asked at once, and time the run.
Add a test that loads results.json after running main and checks it has five rows.

What this course left out

These are worth learning next, when a project needs them:

Topic	What it is for
Tuples and sets	A tuple is a list that cannot change; a set holds each value once and checks membership fast.
`while`, `break` and `match`	Loops that run until a condition changes, leaving a loop early, and matching a value against several shapes.
Docstrings and `*args`	Describing a function inside it, and collecting any number of positional arguments.
`pathlib`	Working with file paths that behave the same on Windows, macOS and Linux.
`logging`	Recording what a program did, with levels, instead of `print`.
Environment variables	Keeping API keys out of your code, read with `os.environ`.
Generators and `yield`	Producing values one at a time, which is how streamed model output arrives.
Writing decorators	You have used `@dataclass` and `@pytest.mark.parametrize`; writing your own comes later.
HTTP requests	Calling a real API with `httpx` or `requests`.
Regular expressions	Finding patterns in text with the `re` module.

Previouspytest.raises and parametrize: testing the failures

You understood something today that you didn't yesterday.