extract, decompose and clear

Three methods remove content from the tree, with different consequences: extract() lifts a node out but keeps it usable in Python, decompose() removes it and lets the garbage collector reclaim it, and clear() empties a tag's contents in place.

.extract() - Remove but Keep

from bs4 import BeautifulSoup

soup = BeautifulSoup("<div><p>a</p><p>b</p></div>", "html.parser")
p = soup.find("p")
removed = p.extract()
print(soup)        # <div><p>b</p></div>
print(removed)     # <p>a</p>  - still a valid object
print(removed.parent)  # None

Use extract() when you want the removed node for later re-insertion or separate processing.

.decompose() - Remove and Forget

soup = BeautifulSoup("<div><script>var x=1;</script><p>a</p></div>", "html.parser")
soup.script.decompose()
print(soup)  # <div><p>a</p></div>

decompose() is the right tool for noise you never want again: scripts, styles, comments, ad containers. After decompose() the object is no longer usable.

.clear() - Empty a Tag

soup = BeautifulSoup("<p><b>x</b> y</p>", "html.parser")
soup.p.clear()
print(soup)  # <p/>

clear() removes all children of a tag but keeps the tag itself, which is handy before filling it with new content.

Choosing the Right Method

• Remove and reuse the node: extract().
• Remove noise permanently: decompose().
• Empty a container in place: clear().

Summary

extract() removes a node from the tree while keeping it as a usable object, decompose() removes it permanently (ideal for script and style cleanup), and clear() empties a tag's contents while keeping the tag. Pick by whether you need the removed content afterward.