This script decomposes the Unicode characters in a text file based on UnicodeData.txt and then creates a json file which can be used in other programming languages for decomposing. I used this script to fix the Persian characters extracted from PDF files.

unicode-decompose.py

import re
import json
fixset = {}
# Get this file from http://ftp.unicode.org/Public/UNIDATA/UnicodeData.txt
with open('UnicodeData.txt', 'r') as data:
	for line in data:
		lineData = line.split(';')
		if lineData[5] != '':
			g = re.search(r'^<.+> (.+)$', lineData[5])
			if g is not None and g.group(1) is not None:
				fixed = ""
				for unicode in g.group(1).split():
					fixed += chr(int(unicode, 16))
				fixset[chr(int(lineData[0], 16))] = fixed
# Pirnt to json for other apps
with open('decomposition.json', 'w') as decomposition:
	json.dump(fixset, decomposition, separators=(',', ":"))
# Fix file
with open('in.txt', 'r', encoding='utf-8') as input:
	with open('out.txt', 'w', encoding='utf-8') as output:
		while True:
			char = input.read(1)         
			if not char:
				break
			output.write(fixset.get(char, char))