import re
import argparse




def remove_html_tag(html_string):
    """
    remove html tag
    :param html_string:
    :return:
    """
    return re.sub('<.*?>', '', html_string)

def convert_html2FAQ(html_string):
    """
    convert html to FAQ
    :param html_string:
    :return: question and answer tuple list

    input example : 满足什么条件才能进行全息幻视?	<div dir="ltr" data-zone-id="0" data-line-index="0" data-line="true">您好，全息幻视需要满足以下两个基本条件：</div> <div data-zone-id="0" data-line-index="1" data-line="true"> </div> <div data-zone-id="0" data-line-index="2" data-line="true">1)您已经拥有了想要幻视的目标城市外观，且当前城市外观和目标城市外观都需要是永久城市外观；</div> <div data-zone-id="0" data-line-index="3" data-line="true">2)目标城市外观假如为可升级类型的城堡皮肤，需要达到最高等级后，才能够被幻视。</div>
    """
    lines = html_string.split('\n')
    lines = [line.strip() for line in lines if line.strip() != '']
    for line in lines:
        tuple = line.split('\t', 2)
        question, answer = tuple[0], tuple[1]
        answer = remove_html_tag(answer)
        yield f"Question: {question}", f"Answer: {answer}"

def filter_makenosence_lines(conversion_content:str):
    lines = conversion_content.splitlines()
    filtered_lines = [ line for line in lines if line.startswith('@Jarvis') or line.startswith('亲爱的玩家') ]
    return '\n'.join(filtered_lines)
    
"""
implement a main entry, and use different flag to execute different functions
"""
if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('--input_file', type=str, default='./JY_FAQ_Origin.txt', help='input file')
    parser.add_argument('--output_file', type=str, default='./JY_FAQ.txt', help='output file')
    parser.add_argument('--command', type=str, default='process_faq', help='use process_faq or process_queries')
    args = parser.parse_args()
    input_file = args.input_file
    output_file = args.output_file
    if args.command == 'process_faq':
        with open(output_file, 'w', encoding='utf-8') as out_f:
            with open(input_file, 'r', encoding='utf-8') as input_f:
                all_html = input_f.read()
                for question, answer in convert_html2FAQ(all_html):
                    out_f.write(f"\n{question}\n{answer}\n")

    elif args.command == 'process_queries':
        with open(output_file, 'w', encoding='utf-8') as out_f:
            with open(input_file, 'r', encoding='utf-8') as input_f:
                conversation_content = input_f.read()
                filterd_content = filter_makenosence_lines(conversation_content)
                out_f.write(filterd_content)