Clean HTML files generated by the Windows Kindle app when exporting highlights.

kindle_highlight_clean.py

import sys
import re

# TODO: The main code should be inside a __main__, I think.

FILE = sys.argv[1]

file = open(FILE, "r", encoding="utf8")
data = file.read()
file.close()

if (
    ("class='bookTitle'" not in data)
    or ("class='noteHeading'" not in data)
    or (not FILE.endswith(".html"))
):
    raise Exception("It seems this file is not a Kindle highlight HTML file.")

original_data = data

replacement_rules = {
    " .": ".",
    " ,": ",",
    "( ": "(",
    " )": ")",
    " :": ":",
    " ;": ";",
    " ?": "?",
    " !": "!",
    " - ": "-",
    " / ": "/",
    "“ ": "“",
    " ”": "”",
}

replacement_rules_regex = {
    "# ([0-9])": "#\\1"
}

for key, value in replacement_rules.items():
    data = data.replace(key, value)

for key, value in replacement_rules_regex.items():
    data = re.sub(key, value, data)

if original_data == data:
    print("Data didn't change.")
else:
    print("Data changed.")
    file = open(FILE, "w", encoding="utf8")
    file.write(data)
    file.close()

print("OK")