fix_json.py

import json_fix

def fix_instructions(file_path):
    with open(file_path, 'r', encoding='utf-8-sig') as f:
        lines = f.readlines()

    for i, line in enumerate(lines):
        if line.strip().startswith('"instruction":'):
            first_quote = line.find('"', line.find(':'))
            last_quote = line.rfind('"')
            # Remove all problematic characters between the quotes
            new_line = line[:first_quote+1] + ''.join(c for c in line[first_quote+1:last_quote] if c.isalnum() or c.isspace()) + line[last_quote:]
            lines[i] = new_line

    with open(file_path, 'w', encoding='utf-8-sig') as f:
        f.writelines(lines)

file_path = r"C:\SD\updated_updated_combined.json"
fix_instructions(file_path)