Rosettes vs Pygments

A detailed comparison of Rosettes and Pygments for syntax highlighting.

Summary

When to Choose Rosettes

✅ Choose Rosettes when:

• You need guaranteed O(n) performance
• Security is critical (no ReDoS risk)
• You're using Python 3.14t with free-threading
• You're highlighting common languages (Python, JS, Rust, etc.)
• You want parallel highlighting for many code blocks

❌ Choose Pygments when:

• You need obscure language support (500+ languages)
• You need output formats beyond HTML and terminal (LaTeX, RTF, SVG)
• You're on Python < 3.14
• You need battle-tested stability

Feature Comparison

Language Support

Rosettes covers: Python, JavaScript, TypeScript, Rust, Go, C, C++, Java, Ruby, PHP, and 45 more popular languages.

Pygments adds: COBOL, Fortran, Ada, dozens of DSLs, legacy languages, and specialized formats.

Output Formats

Rosettes supports HTML and terminal (ANSI) output. For LaTeX, RTF, SVG, or IRC, use Pygments.

CSS Compatibility

Rosettes can use any Pygments CSS theme withcss_class_style="pygments".

Performance Comparison

Single Block Highlighting

Parallel Highlighting (8 blocks)

Pygments doesn't have built-in parallel support, requiring manual thread pool setup.

Security Comparison

ReDoS Vulnerability

Rosettes: Not vulnerable. State machine lexers process each character exactly once.

Pygments: Some lexers are vulnerable. Crafted input can cause exponential processing time.

Example vulnerable pattern (simplified):

# Pygments regex pattern (hypothetical)
r'(a+)+$'

# Malicious input
"aaaaaaaaaaaaaaaaaaaaaaaaaaaa!"

# Result: Exponential backtracking

Rosettes eliminates this entire category of vulnerability by design.

API Comparison

Basic Highlighting

Rosettes:

from rosettes import highlight

# HTML output (default)
html = highlight("def foo(): pass", "python")

# Terminal output (ANSI colors)
ansi = highlight("def foo(): pass", "python", formatter="terminal")

Pygments:

from pygments import highlight
from pygments.lexers import PythonLexer
from pygments.formatters import HtmlFormatter

html = highlight("def foo(): pass", PythonLexer(), HtmlFormatter())

Tokenization

Rosettes:

from rosettes import tokenize
tokens = tokenize("x = 1", "python")

Pygments:

from pygments.lexers import PythonLexer
tokens = list(PythonLexer().get_tokens("x = 1"))

Parallel Highlighting

Rosettes:

from rosettes import highlight_many
results = highlight_many(blocks)  # Built-in

Pygments:

from concurrent.futures import ThreadPoolExecutor
from pygments import highlight
# ... manual setup required

Migration Path

Switching from Pygments to Rosettes is straightforward for HTML output:

# Before (Pygments)
from pygments import highlight
from pygments.lexers import get_lexer_by_name
from pygments.formatters import HtmlFormatter

lexer = get_lexer_by_name("python")
formatter = HtmlFormatter()
html = highlight(code, lexer, formatter)

# After (Rosettes)
from rosettes import highlight

html = highlight(code, "python", css_class_style="pygments")

Your existing CSS themes work without changes.

See Migration Guide for a complete walkthrough.

Ecosystem

Pygments Ecosystem

• Sphinx integration
• Jupyter integration
• Many editor plugins
• 18+ years of community themes
• Extensive documentation

Rosettes Ecosystem

• Bengal static site generator
• Pygments CSS theme compatibility
• Growing documentation
• Modern Python 3.14+ focus

Conclusion

For most web applications highlighting popular languages, Rosettes provides better performance and security. For specialized needs (obscure languages, LaTeX/RTF output), Pygments remains the comprehensive choice.