import os import sys from pathlib import Path import chardet from dotenv import load_dotenv import re import requests from urllib.parse import urlparse from bs4 import BeautifulSoup import codecs # BOM付き保存用ライブラリ # ------------------------------------------------------ # 初期設定 # ------------------------------------------------------ account_inf_path = "d:/MyWebs/Database/accounts.env" config_path = "translate.env" limit_to_multibyte_str = False source_lang = "JA" target_lang = "EN" url_to_translate = "http://d2mate.mdxes.iir.isct.ac.jp/D2MatE/excel/vba_setup.html" # ------------------------------------------------------ # 設定ファイル読み込み # ------------------------------------------------------ load_dotenv(dotenv_path = account_inf_path) api_key = os.getenv("DEEPL_API_KEY") if not api_key: print("ERROR: DeepL APIキーが見つかりません。環境変数 'DEEPL_API_KEY' を設定してください。") sys.exit(1) endpoint = os.getenv("endpoint") if not api_key: print("ERROR: endpointが見つかりません。環境変数 'endpoint' を設定してください。") sys.exit(1) load_dotenv(dotenv_path = config_path) force_server_charcode = os.getenv("force_server_charcode", 'utf-8') out_html = os.getenv("out_html", '') # ''ならurlから自動生成 # ------------------------------------------------------ # 起動時引数設定 # ------------------------------------------------------ argv = sys.argv nargs = len(argv) if nargs > 1: url_to_translate = argv[1] if nargs > 2: source_lang = argv[2] if nargs > 3: target_lang = argv[3] print() print(f"Translate [{url_to_translate}] from [{source_lang}] to [{target_lang}]") print() # ------------------------------------------------------ # 関数 # ------------------------------------------------------ def usage(): print() print("usage: python translate.py url (source_lang target_lang)") print(" source_lang, target_lang: e.g., JA, EN") def split_url(url): parsed = urlparse(url) path_segments = parsed.path.split('/') # パスを分割してセグメント化 file_name = path_segments[-1] if path_segments[-1] else None # ファイル名（空の場合はNone） if file_name: if '.' in file_name: file_body, extension = file_name.rsplit('.', 1) else: file_body, extension = file_name, None else: file_body, extension = None, None directory = '/'.join(path_segments[:-1]) if len(path_segments) > 1 else None return { "protocol": parsed.scheme, "server": parsed.netloc, "directory": directory, "file_name": file_name, "file_body": file_body, "extension": extension } def get_filename_from_url(url): print(f"Fetching HTML header from {url}...") filename_candidates = ['', 'index.html', 'index.php', 'Default.html', 'Default.htm'] for f in filename_candidates: response = None try: if f == "": response = requests.head(url, allow_redirects=True) elif url.endswith('/'): response = requests.head(url + f, allow_redirects=True) else: response = requests.head(url + '/' + f, allow_redirects=True) except Exception as e: # return f"Error: {e}" pass if response: final_url = response.url url_dict = split_url(final_url) if url_dict.get('file_name', None) is not None: return final_url else: # print("else") pass return None def get_from_url(url, force_server_charcode = ""): print(f"Fetching HTML from {url}...") filal_url = get_filename_from_url(url) try: response = requests.get(filal_url, allow_redirects=True) except Exception as e: return f"Error: {e}", "" if response.status_code != 200: return f"Failed to fetch HTML: {response.status_code}", "" print("response.encoding=", response.encoding) print(f"Content-Type Header: {response.headers.get('Content-Type')}") # print("response.content=", response.content) # print("response.text=", response.text) detected_encoding = chardet.detect(response.content) print(f"Detected Encoding: {detected_encoding}") if force_server_charcode != "": response.encoding = force_server_charcode if response.encoding is None: response.encoding = 'utf-8' # 最終的なリダイレクト先URL final_url = response.url # print(f"Final URL: {final_url}") return response.text, final_url def is_multibyte_str(text): # 2バイト文字（日本語を含む）を検出する正規表現 pattern = re.compile(r'[\u0800-\uFFFF]') return bool(pattern.search(text)) def translate_text(text, api_key): """ DeepL APIを使用してテキストを翻訳する関数 """ if not is_multibyte_str(text): return text headers = {"Authorization": f"DeepL-Auth-Key {api_key}"} params = { "text": text, "source_lang": source_lang, "target_lang": target_lang } response = requests.post(endpoint, headers=headers, data=params) if response.status_code == 200: result = response.json() return result["translations"][0]["text"] else: print(f"Error: {response.status_code}, {response.text}") return text # エラーの場合、元のテキストを返す def translate_html(text, api_key): """ HTMLの日本語部分を英語に翻訳 """ print("Parsing HTML...") soup = BeautifulSoup(text, 'html.parser') print("Translating content...") for element in soup.find_all(string=True): # 全てのテキストノードを取得 if element.strip(): # 空白でない場合のみ処理 try: print(f" Original: {element}") # 元のテキストを出力（デバッグ用） translated_text = translate_text(element, api_key) print(f" Translated: {translated_text}") # 翻訳後のテキストを出力 element.replace_with(translated_text) # 翻訳済みテキストに置換 except Exception as e: print(f"Error while translating: {e}") return str(soup) def main(): global out_html text, final_url = get_from_url(url_to_translate, force_server_charcode = force_server_charcode) if final_url == "": print(f"Error in main(): {text}") exit() print("final_url:", final_url) url_dict = split_url(final_url) #print("url_dict:", url_dict) file_name = url_dict.get('file_name', None) if file_name is None: print(f"Error: URL [{final_url}] does not include file name.") exit() if out_html == '': if re.search(r'_jp\.', file_name): out_html = re.sub(r'_jp\.', r'_en\.', url_dict['file_name']) else: out_html = f"{url_dict['file_body']}_en.{url_dict['extension']}" print("out_html:", out_html) if os.path.exists(out_html): print(f"\nError: out_html [{out_html}] exists. Terminate.\n") return translated_html = translate_html(text, api_key) print(f"Save translated HTML to {out_html}") with open(out_html, "w", encoding="utf-8") as file: file.write(translated_html) if __name__ == "__main__": main() usage()