0Pricing
Web Scraping & Bots · レッスン

複雑なHTML構造のナビゲーション

深くネストされたHTMLドキュメントや不規則な構造のHTMLドキュメントを効率的にたどる技法を学びます。

「複雑なHTML構造のナビゲーション」はCoddyKit上の無料Web Scraping & Botsレッスンです。 これはレッスン1/4です。 下記で完全なレッスンを無料で読むことができます。その後、ブラウザ内の組み込みコードエディタと24時間対応のAIチューターでハンズオン演習できます。 これはWeb Scraping & Bots学習パスの一部であり、ウェブとCoddyKitアプリ全体で進捗が同期されます。 Web Scraping & Botsコースには全4レッスンが含まれています。

このレッスンの一部はまだ翻訳されておらず、英語で表示されています。

Beyond Simple Extraction

What happens when the data you need isn't neatly tucked into an element with a unique ID or class? Sometimes, web pages have complex layouts that require a more sophisticated approach.

In this lesson, we'll learn how to "walk" through the HTML structure, finding elements based on their relationships to others. This technique is called HTML tree traversal.

The HTML Tree Structure

Think of an HTML document like a family tree. Every element (like a <div>, <p>, or <h1>) is a "node."

  • Parent: An element that contains other elements.
  • Child: An element directly inside another element.
  • Sibling: Elements at the same level, sharing the same parent.
  • Descendant: Any element inside a parent, directly or indirectly.

Understanding these relationships is key to navigating complex pages.

Children: Direct Descendants

To get the direct children of an element, BeautifulSoup offers a couple of ways. The .contents property returns a list of all children, including NavigableStrings (text nodes).

The .children property returns an iterator, which is often more memory-efficient for large documents. Let's see an example.

from bs4 import BeautifulSoup

html_doc = """
<div class="parent">
  <p>Child 1</p>
  <span>Child 2</span>
  Text node
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')
parent_div = soup.find('div', class_='parent')

print("Using .contents:")
for child in parent_div.contents:
  print(f"- {child.name or 'Text'}: {repr(child)[:20]}...")

print("\nUsing .children:")
for child in parent_div.children:
  print(f"- {child.name or 'Text'}: {repr(child)[:20]}...")

Moving Up: Parents

Sometimes you find an element, but need information from its containing element. BeautifulSoup allows you to easily move "up" the tree.

  • The .parent property gives you the direct parent of an element.
  • The .parents property gives you an iterator for all ancestor elements, all the way up to the document root.
from bs4 import BeautifulSoup

html_doc = """
<div class="grandparent">
  <div class="parent">
    <p class="child">Hello</p>
  </div>
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')
child_p = soup.find('p', class_='child')

print(f"Child: {child_p.name}")
print(f"Direct Parent: {child_p.parent.name}")

print("All Ancestors:")
for p in child_p.parents:
  if p.name: # Filter out [document] and other non-tag elements
    print(f"- {p.name}")

Side-by-Side: Single Siblings

Elements at the same level are called siblings. You can move between them using .next_sibling and .previous_sibling.

Be aware that these properties will also include "NavigableString" objects if there's whitespace or text directly between your tags. You might need to check if the result is a tag.

from bs4 import BeautifulSoup

html_doc = """
<div class="items">
  <p>Item 1</p>
  <span>Item 2</span>
  <p>Item 3</p>
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')
item_one = soup.find('p', string='Item 1')

# The next sibling after <p>Item 1</p> is usually a newline character (NavigableString)
# We need to find the *next tag* sibling
next_tag = item_one.next_sibling
while next_tag and not next_tag.name:
    next_tag = next_tag.next_sibling

print(f"Item 1: {item_one.text}")
if next_tag:
  print(f"Next tag sibling: {next_tag.text}")
else:
  print("No next tag sibling found.")

All Siblings: next_siblings

To get all the siblings that come after or before a particular tag, you can use the .next_siblings and .previous_siblings iterators.

This is extremely useful when you've found one element and need to extract data from all related elements at the same level.

from bs4 import BeautifulSoup

html_doc = """
<div class="menu">
  <a href="#home">Home</a>
  <a href="#about">About</a>
  <a href="#contact">Contact</a>
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')
home_link = soup.find('a', href='#home')

print("All next siblings after Home:")
for sibling in home_link.next_siblings:
  if sibling.name == 'a': # Only interested in <a> tags
    print(f"- {sibling.text}: {sibling['href']}")

Beyond Siblings: find_next

Sometimes, the data you need isn't a direct child or sibling, but appears somewhere later in the HTML document relative to your current element. This is where .find_next() and .find_all_next() come in.

These methods search the rest of the document after the current tag, regardless of parent-child-sibling relationships. Similarly, .find_previous() and .find_all_previous() search before.

from bs4 import BeautifulSoup

html_doc = """
<div class="header">
  <h2>Section Title</h2>
</div>
<p>Some introductory text.</p>
<div class="content">
  <p>First paragraph.</p>
  <span>Important data!</span>
  <p>Second paragraph.</p>
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')
section_title = soup.find('h2')

# Find the next <span> tag anywhere after the h2
important_data_span = section_title.find_next('span')

print(f"Section Title: {section_title.text}")
if important_data_span:
  print(f"Data found after title: {important_data_span.text}")

# Find all <p> tags after the h2
all_next_paragraphs = section_title.find_all_next('p')
print("All paragraphs after title:")
for p in all_next_paragraphs:
    print(f"- {p.text}")

Chaining Traversal Methods

The real power of traversal comes from chaining these methods. You can start at one point, move to a parent, then find a sibling, and then extract data from its children.

This allows you to navigate very specific, complex pathways to reach exactly the data you need, even if it's not directly addressable by a simple selector.

from bs4 import BeautifulSoup

html_doc = """
<div class="product-listing">
  <div class="product">
    <h3>Laptop X</h3>
    <p class="price">$1200</p>
  </div>
  <div class="product">
    <h3>Mouse Y</h3>
    <p class="price">$50</p>
  </div>
</div>
<div class="summary">
  <p>Total items: 2</p>
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')

# Find "Laptop X" and then get its price
laptop_h3 = soup.find('h3', string='Laptop X')
if laptop_h3:
  # The price is a sibling of the h3
  price_tag = laptop_h3.find_next_sibling('p', class_='price')
  if price_tag:
    print(f"{laptop_h3.text} price: {price_tag.text}")

# Find the product listing, then get the text of the summary's paragraph
product_listing_div = soup.find('div', class_='product-listing')
if product_listing_div:
  summary_div = product_listing_div.find_next_sibling('div', class_='summary')
  if summary_div:
    summary_text = summary_div.find('p').text
    print(f"Summary: {summary_text}")

Handling Text Nodes

When traversing, you'll often encounter NavigableString objects. These represent the text content that isn't wrapped in its own HTML tag, like the whitespace (newlines and spaces) between tags.

BeautifulSoup treats these as distinct nodes in the tree. When using traversal methods like .next_sibling, you might get a NavigableString before you get the next actual tag. Always check .name or type() if you only want tags.

from bs4 import BeautifulSoup

html_doc = """
<div>
  Hello
  <p>World</p>
  !
</div>
"""
soup = BeautifulSoup(html_doc, 'html.parser')
div_tag = soup.find('div')

print("Children of the div:")
for child in div_tag.children:
  if child.name:
    print(f"- Tag: {child.name}, Text: {child.text}")
  else:
    print(f"- Text Node: {repr(child)}")

# Accessing a specific NavigableString
hello_text_node = list(div_tag.children)[0]
print(f"\nFirst child (text node): '{hello_text_node.strip()}'")

Traversal Challenge

Consider the following HTML snippet. You want to extract the "Availability" status for "Product B". Which sequence of BeautifulSoup traversal methods would correctly get you the text "In Stock" starting from the <h3> tag for "Product B"?

<html>
<body>
<div class="catalog">
  <div class="item">
    <h3>Product A</h3>
    <p>Price: $10</p>
    <span>Status: Out of Stock</span>
  </div>
  <div class="item">
    <h3>Product B</h3>
    <p>Price: $20</p>
    <span>Availability: In Stock</span>
  </div>
  <div class="item">
    <h3>Product C</h3>
    <p>Price: $30</p>
    <span>Status: Low Stock</span>
  </div>
</div>
</body>
</html>

Recap: Mastering Traversal

Congratulations! You've learned how to navigate complex HTML structures using BeautifulSoup's powerful traversal methods.

  • We explored moving up (.parent, .parents), down (.contents, .children), and sideways (.next_sibling, .previous_sibling, .next_siblings, .previous_siblings) in the HTML tree.
  • You also saw how to find elements anywhere after or before a current tag using .find_next() and .find_previous().
  • Mastering these techniques allows you to extract data from even the most challenging and irregularly structured web pages.

Next, we'll dive into using CSS selectors for even more precise data extraction!

よくある質問

「複雑なHTML構造のナビゲーション」レッスンは無料ですか？

はい。「複雑なHTML構造のナビゲーション」の完全なテキストはこのウェブで無料で読めます。インタラクティブに演習し（組み込みコードエディタと24時間対応のAIチューター）、Web Scraping & Botsコースの残りをアンロックするには、CoddyKit PROにアップグレードしてください。 Web Scraping & Botsコースには全4レッスンが含まれています。

「複雑なHTML構造のナビゲーション」で何を学びますか？

深くネストされたHTMLドキュメントや不規則な構造のHTMLドキュメントを効率的にたどる技法を学びます。 ブラウザで直接実行するハンズオンコードでWeb Scraping & Botsを演習し、24時間対応のAIチューターがレッスンを進める中での質問に答えます。

Web Scraping & Botsを始めるのに経験は必要ですか？

事前経験は必要ありません。CoddyKitのWeb Scraping & Botsは初級者から上級者向けに構成されているため、ここから始めるか最初から始めて、自分のペースで進むことができます。 これはレッスン1/4です。

「複雑なHTML構造のナビゲーション」レッスンにはどのくらい時間がかかりますか？

ほとんどのCoddyKitレッスンは約5～10分かかります。各レッスンはコンパクトでインタラクティブなので、着実に進歩し、ウェブとアプリ全体で正確に前回の場所から再開できます。

このWeb Scraping & Botsレッスンでコードを書いて実行できますか？

はい。すべてのWeb Scraping & Botsレッスンに組み込みコードエディタが含まれているため、ブラウザでリアルコードを書いて実行し、即座のAIフィードバックを取得できます。ローカル設定は不要です。

このコースのすべてのレッスン

  1. 複雑なHTML構造のナビゲーション
  2. 精密な抽出のためのCSSセレクター
  3. 堅牢な選択のためのXPath
  4. HTML テーブルからデータを抽出する
← Web Scraping & Botsに戻る