HTML parsing benchmark

benchmark.py

import timeit
import requests

import html_text
import justext 
from selectolax.parser import HTMLParser

# There's a difference between these two
html_long = requests.get("http://planet.python.org/").content.decode("utf-8")
html_short = "<p><b>This</b> is just a small example.</p>"

# Run benchmark
htmls = {"long": html_long, "short": html_short}

results = {} 

for html_name, html in htmls.items():
    methods = {
        "justext": lambda: justext.justext(html, tuple()),
        "htmltext": lambda: html_text.extract_text(html, guess_layout=False),
        "selectolax": lambda: HTMLParser(html).text()
    }
    for method, func in methods.items():
        print(method, html_name)
        results[(html_name, method)] = timeit.timeit(func, number=500)
        
# Pretty Print 
from rich.console import Console
from rich.table import Table

table = Table(title="Benchmark Results")

table.add_column("HTML Variant", style="cyan")
table.add_column("Library", justify="right", style="magenta")
table.add_column("Time (s)", style="green")

for (variant, library), speed in results.items():
    table.add_row(variant, library, str(speed))

console = Console()
console.print(table)

"""
                 Benchmark Results                  
┏━━━━━━━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━┓
┃ HTML Variant ┃    Library ┃ Time (s)             ┃
┡━━━━━━━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━┩
│ long         │    justext │ 0.49477454099996976  │
│ long         │   htmltext │ 10.072820916999945   │
│ long         │ selectolax │ 14.882792749999908   │
│ short        │    justext │ 0.025378625000030297 │
│ short        │   htmltext │ 0.012803834000010283 │
│ short        │ selectolax │ 0.004145417000017915 │
└──────────────┴────────────┴──────────────────────┘

"""