Greedが生成したhtmlから問題文を抜き出して，日本語に翻訳&挿入したhtmlを作成

generateHtml.py

import sys
import subprocess
from html.parser import HTMLParser
from microsofttranslator import Translator

client_id = ""
client_secret = ""


class TestHTMLParser(HTMLParser):
    def __init__(self):
        HTMLParser.__init__(self)
        self.in_intro = False
        self.problem = ""

    def handle_starttag(self, tag, attrs):
        if 'intro' == tag:
            self.in_intro = True

    def handle_endtag(self, tag):
        if 'intro' == tag:
            self.in_intro = False

    def handle_data(self, data):
        if self.in_intro and data.strip():
            self.problem += data


def main():
    argvs = sys.argv

    if len(argvs) == 2:
        input_file_name = argvs[1]
        output_file_name = input_file_name.replace(".html", "") + "_JP.html"
    elif len(argvs) == 3:
        input_file_name = argvs[1]
        output_file_name = argvs[2]
    else:
        print("input html")
        return

    with open(input_file_name) as f:
        html = f.read()

    parser = TestHTMLParser()
    parser.feed(html)
    parser.close()

    problem = parser.problem

    # 翻訳
    try:
        translator = Translator(client_id=client_id, client_secret=client_secret)
        problem_jp = translator.translate(problem, "ja")
    except:
        ex, ms, tb = sys.exc_info()
        problem_jp = "An unexpected error has occurred.<br/>"
        problem_jp += str(ex) + str(ms) + str(tb)

    # 作成した日本語訳の挿入
    block = '<h2 class="section-title">Problem Statement(日本語)</h2>'
    block += "<div>" + problem_jp.replace("。", "<br/>") + "</div>"
    i = html.find('<h2 class="section-title">Problem Statement</h2>')
    html = html[:i] + block + html[i:]
    
    with open(output_file_name, "w", encoding="utf-8") as f:
        f.write(html)

    # operaでhtmlを開く
    subprocess.check_call(["C:/Program Files (x86)/Opera/launcher.exe", output_file_name])

if __name__ == '__main__':
    main()