Exercise 1: Method Chain — Logging Pipeline

Goal

Build a logging pipeline using the Chain of Responsibility method chain pattern.

A log message has a level:

DEBUG = 1
INFO  = 2
WARNING = 3
ERROR = 4

A handler is configured with a minimum level. When a message arrives, the handler processes it if the message level is exactly its own level, and passes it along otherwise. At the end of the chain, unhandled messages are silently dropped.

---

Part A — build the chain

Implement:

• A LogLevel enum with DEBUG, INFO, WARNING, ERROR.
• A LogMessage dataclass with level: LogLevel and text: str.
• An abstract LogHandler base class with:
  • set_next(handler) -> LogHandler — wires the chain, returns the handler passed in.
  • handle(message: LogMessage) -> None — processes or passes along.
• Three concrete handlers: DebugHandler, InfoHandler, WarningHandler, ErrorHandler — each prints a formatted line when it processes a message.

Wire them and verify the output matches the expected section below.

---

Part B — trace the chain

Add a name attribute to the base handler. Before each handler either processes or passes a message, print a trace line:

[DebugHandler] received DEBUG — processing
[DebugHandler] received INFO — passing along
[InfoHandler] received INFO — processing

This makes the chain’s decision-making visible. Run the same messages from Part A with tracing on.

---

Expected output (Part A, no tracing)

[DEBUG] hello from debug
[INFO]  user logged in
[WARN]  disk usage at 80%
[ERROR] database connection failed

---

Skeleton

See exercise1.py.

---

Hints

• set_next should store the next handler and return it — this enables the fluent wiring style: debug.set_next(info).set_next(warning).set_next(error).
• The base handle method should call self._next.handle(message) if self._next is not None.
• Each concrete handler should override handle, check if the message is “theirs”, process or pass.
• In Part B, print the trace line inside handle before deciding what to do.

---

Back to notes · Exercise 2