Exercise 1: Factory Methods as Named Constructors

Goal

Practice using factory methods when a class has several natural ways to create itself.

The focus of this exercise is not choosing between different classes. The focus is naming different construction paths clearly.

You will create a Document class that stores a normalized document:

Document(title: str, body: str)

The final object should always store plain text in body, but callers may provide input in different formats.

---

Task

Open exercise1.py and complete the Document class.

Implement these class methods:

Document.from_plain_text(title, text)
Document.from_markdown(title, markdown)
Document.from_html(title, html)

Each method should return a Document.

---

Requirements

1. from_plain_text

Should:

• strip leading/trailing whitespace from the title
• strip leading/trailing whitespace from the body
• return a Document

Example:

doc = Document.from_plain_text(" Notes ", " Hello world ")

assert doc.title == "Notes"
assert doc.body == "Hello world"

2. from_markdown

Should:

• strip leading/trailing whitespace from the title
• remove simple Markdown heading markers such as #, ##, and ###
• remove simple bold markers **
• strip leading/trailing whitespace from the final body
• return a Document

Example:

doc = Document.from_markdown(
    " Notes ",
    "# Heading\n\nThis is **important**.",
)

assert doc.title == "Notes"
assert "#" not in doc.body
assert "**" not in doc.body
assert "Heading" in doc.body
assert "important" in doc.body

You do not need to build a real Markdown parser. Keep the conversion simple.

3. from_html

Should:

• strip leading/trailing whitespace from the title
• remove simple HTML tags
• unescape basic HTML entities, such as &
• strip leading/trailing whitespace from the final body
• return a Document

Example:

doc = Document.from_html(
    " Notes ",
    "<h1>Heading</h1><p>Alice &amp; Bob</p>",
)

assert doc.title == "Notes"
assert "<h1>" not in doc.body
assert "Alice & Bob" in doc.body

---

Test examples

The skeleton file includes a small run_tests() function. After completing the exercise, this command should pass:

python exercise1.py

The tests check behavior like this:

def test_plain_text_document_creation():
    doc = Document.from_plain_text(" Notes ", " Hello world ")
    assert doc.title == "Notes"
    assert doc.body == "Hello world"


def test_markdown_document_creation():
    doc = Document.from_markdown(
        " Notes ",
        "# Heading\n\nThis is **important**.",
    )
    assert doc.title == "Notes"
    assert "#" not in doc.body
    assert "**" not in doc.body
    assert "Heading" in doc.body
    assert "important" in doc.body


def test_html_document_creation():
    doc = Document.from_html(
        " Notes ",
        "<h1>Heading</h1><p>Alice &amp; Bob</p>",
    )
    assert doc.title == "Notes"
    assert "<h1>" not in doc.body
    assert "Alice & Bob" in doc.body

You can add more tests if your implementation handles more Markdown or HTML cases.

---

Questions to answer after completing it

1. Why is Document.from_markdown(...) clearer than Document(...)?
2. What does the method name communicate to the reader?
3. Would Document.create(...) be a good name here? Why or why not?
4. Should these methods be @staticmethod or @classmethod?

---

Expected takeaway

Factory methods are useful when the same class has several meaningful ways to be created.

In this exercise:

Document.from_plain_text(...)
Document.from_markdown(...)
Document.from_html(...)

all return a Document, but each method explains how the input should be interpreted.

---

Back to Factory Methods as Named Constructors · Script · Solution