A simple compression test of a book in TXT UTF-8 format

text-compression-test.py

#!/bin/python3
import lzma
import gzip
import deflate #pip install deflate
import brotli  #pip install brotlipy

# From https://www.gutenberg.org/files/1342/1342-0.txt
BOOK_NAME = "pride_and_prejudice-jane_austen.txt"


def compress_brotli(txt_bytes):
    bytes_ = brotli.compress(txt_bytes, brotli.BrotliEncoderMode.TEXT, 11) # 11 is maximum compression rate
    print_datasize("Brotli", bytes_)


def compress_lzma(txt_bytes):
    bytes_ = lzma.compress(txt_bytes, format=lzma.FORMAT_ALONE, preset=9) # 9 is maximum compression rate
    print_datasize("LZMA", bytes_)


def compress_deflategzip(txt_bytes):
    bytes_ = deflate.gzip_compress(txt_bytes, 12) # 12 is maximum compression rate
    print_datasize("Deflate+GZip", bytes_)


def compress_gzip(txt_bytes):
    bytes_ = gzip.compress(txt_bytes)
    print_datasize("GZip", bytes_)


def print_datasize(algo, bytes_):
    bytes_count = len(bytes_) / 1000
    print(f"Size using {algo}: {bytes_count:.3f}KB")


### MAIN ###
with open(BOOK_NAME, "rb") as book:
    txt_bytes = book.read()

# Print uncompressed file size
print_datasize("NO COMPRESSION", txt_bytes)

# Test compression algorithms
compress_brotli(txt_bytes)     # 1°
compress_lzma(txt_bytes)       # 2°
compress_deflategzip(txt_bytes) # 3°
compress_gzip(txt_bytes)