htmltree-view

Visualize HTML DOM structure as a depth-limited, colorized ASCII tree — like the tree command, but for HTML files.

<html> lang="en"  [ L0 2ch ]
├── <head>  [ L0 4ch ]
│   ├── <meta> charset="utf-8"  [ L1 empty ]
│   ├── <meta> name="viewport" content="width=device-width"  [ L1 empty ]
│   ├── <title>  [ L1 empty ]
│   │   └── "My Page"
│   └── <link> rel="stylesheet" href="style.css"  [ L1 empty ]
└── <body>  [ L1 3ch ]
    ├── <header>  [ L2 2ch ]
    │   └── … (2 children hidden)
    ├── <main> id="main-content"  [ L2 2ch ]
    │   └── … (2 children hidden)
    └── <footer>  [ L2 2ch ]
        └── … (2 children hidden)

────────────────────────────────────────────────────
  Tags: 8  Text nodes: 1  Max depth: 2 (capped at 2)
  Top tags: meta×2, html×1, head×1, title×1, link×1

Features

• Depth limiting — -d N stops at level N; truncated sub-trees show a … (X children hidden) hint
• CSS selector zoom — -s "#app" or -s "body > main" focuses any sub-tree
• Semantic tag colors — headings in amber, structural in blue, forms in pink, links in cyan, etc.
• Depth-cycling pipe colors — guide lines change shade per nesting level
• [L3 5ch] badges — depth level + direct child-tag count on every node
• Text nodes — quoted inline, with --text-limit truncation and whitespace collapsing
• Attribute filtering — --attrs id class href shows only what you care about; --attrs hides all
• Attribute value truncation — --attr-limit 80 prevents base64/data-URI blowout
• HTML comments — hidden by default, shown with --show-comments
• URL fetching — htmltree https://example.com -d 3
• stdin pipe — curl ... | htmltree - or echo '<div/>' | htmltree -
• Output to file — -o tree.txt (auto-disables color)
• Auto color detection — ANSI disabled when stdout is not a TTY; respects NO_COLOR / FORCE_COLOR env vars
• Streaming output — iter_lines() yields one line at a time; never builds the full string unless you ask
• No recursion — iterative DFS walk; handles arbitrarily deep HTML without RecursionError
• Stats summary — total tags, text nodes, comments, max depth seen, top-5 tag frequencies

Install

pip install htmltree-view

# With faster lxml parser:
pip install "htmltree-view[lxml]"

# With html5lib (most spec-accurate):
pip install "htmltree-view[html5lib]"

CLI

# Full tree
htmltree index.html

# Limit depth to 3 levels
htmltree index.html -d 3

# Focus on a CSS-selected sub-tree
htmltree index.html -s "body > main"
htmltree index.html -s "#app"
htmltree index.html -s ".container"

# Fetch from URL
htmltree https://example.com -d 4

# Read from stdin
curl https://example.com | htmltree -
echo '<div><p>hi</p></div>' | htmltree -

# Show only id and class attributes
htmltree index.html --attrs id class

# Hide all attributes
htmltree index.html --attrs

# Hide text nodes (structure only)
htmltree index.html --no-text

# Show HTML comments
htmltree index.html --show-comments

# Truncate text/attr at 40 chars
htmltree index.html --text-limit 40 --attr-limit 40

# Save to file (color auto-disabled)
htmltree index.html -o structure.txt

# Pipe to less with color preserved
htmltree index.html --force-color | less -R

# Use lxml backend (faster)
htmltree index.html --parser lxml

# Plain output (no ANSI)
htmltree index.html --no-color

Python API

from htmltree import HtmlTree

html = open("index.html").read()

# Basic usage
tree = HtmlTree(html)
tree.print()

# Limit depth, filter attributes
tree = HtmlTree(html, max_depth=3, show_attrs=["id", "class"])
tree.print()

# Zoom into a sub-tree
tree = HtmlTree(html, max_depth=5, show_text=False)
tree.print(root_selector="body > main")

# Render to string
tree = HtmlTree(html, max_depth=2, force_color=False)
output = tree.render(root_selector="body")
print(output)

# Stream line by line (memory-efficient for large pages)
tree = HtmlTree(html, max_depth=4)
for line in tree.iter_lines(root_selector="#content"):
    print(line)

# Access stats after render
tree.render()
print(tree.stats.total_tags)
print(tree.stats.tag_counts)      # dict: tag name → count
print(tree.stats.max_depth_seen)
print(tree.stats.total_text_nodes)
print(tree.stats.total_comments)

CLI reference

Tree legend

Environment variables

Requirements

• Python ≥ 3.8
• beautifulsoup4 ≥ 4.12
• Optional: lxml, html5lib

License

MIT

👤 Author

Hadi Cahyadi

Support me on Patreon