古いメール *.eml を読める形のテキストに変換。ただしこれで展開できなかったものもある

メールをtxtに変換.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import email\n",
    "from email import policy\n",
    "from email.parser import BytesParser\n",
    "import glob\n",
    "from datetime import datetime\n",
    "\n",
    "\n",
    "def read_eml(file_path):\n",
    "  \"\"\"\n",
    "  .eml形式のメールファイルを読み込み、その内容を抽出します。\n",
    "  引数:\n",
    "    file_path (str): .emlファイルのパス。\n",
    "  戻り値:\n",
    "    dict: メールの件名、送信者、受信者、日付、本文を含む辞書。\n",
    "      - 'subject' (str): メールの件名。\n",
    "      - 'from' (str): メールの送信者。\n",
    "      - 'to' (str): メールの受信者。\n",
    "      - 'date' (str): メールが送信された日付。\n",
    "      - 'body' (str): メールのプレーンテキスト本文。\n",
    "  \"\"\"\n",
    "  with open(file_path, 'rb') as f:\n",
    "    msg = BytesParser(policy=policy.default).parse(f)\n",
    "  \n",
    "  # Extract the email content\n",
    "  subject = msg['subject']\n",
    "  from_ = msg['from']\n",
    "  to = msg['to']\n",
    "  date = msg['date']\n",
    "  body = msg.get_body(preferencelist=('plain'))\n",
    "  if body:\n",
    "    body = body.get_content()\n",
    "  else:\n",
    "    body = \"\"\n",
    "  \n",
    "  return {\n",
    "    'subject': subject,\n",
    "    'from': from_,\n",
    "    'to': to,\n",
    "    'date': date,\n",
    "    'body': body\n",
    "  }\n",
    "\n",
    "def save_email_content_to_txt(file_path, email_content):\n",
    "  \"\"\"\n",
    "  メールの内容をテキストファイルに保存します。\n",
    "  引数:\n",
    "    file_path (str): .emlファイルのパス。\n",
    "    email_content (dict): メールの内容を含む辞書。\n",
    "  \"\"\"\n",
    "\n",
    "  # メールの日付を解析して yyyymmdd 形式に変換\n",
    "  email_date = datetime.strptime(email_content['date'], '%a, %d %b %Y %H:%M:%S %z')\n",
    "  date_str = email_date.strftime('%Y%m%d')\n",
    "\n",
    "  # 新しいファイル名を作成\n",
    "  txt_file_path = date_str + '_' + file_path + '.txt'\n",
    "  # ディレクトリが存在しない場合は作成\n",
    "  if not os.path.exists('text'):\n",
    "    os.makedirs('text')\n",
    "\n",
    "  # 新しいファイルパスを text/ ディレクトリに変更\n",
    "  txt_file_path = os.path.join('text', txt_file_path)\n",
    "\n",
    "  with open(txt_file_path, 'w') as f:\n",
    "    f.write('Subject: ' + email_content['subject'] + '\\n')\n",
    "    f.write('From: ' + email_content['from'] + '\\n')\n",
    "    f.write('To: ' + email_content['to'] + '\\n')\n",
    "    f.write('Date: ' + email_content['date'] + '\\n')\n",
    "    f.write('---------------------\\n')\n",
    "    f.write(email_content['body'])\n",
    "\n",
    "# .emlファイルをすべて取得\n",
    "eml_files = glob.glob('*.eml')\n",
    "\n",
    "for file_path in eml_files:\n",
    "  # メール内容を読み込む\n",
    "  email_content = read_eml(file_path)\n",
    "\n",
    "  # 関数を使用してメール内容をテキストファイルに保存\n",
    "  save_email_content_to_txt(file_path, email_content)\n",
    "\n",
    "  # メール内容を表示\n",
    "  print(email_content)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import glob\n",
    "\n",
    "# Get all .txt files in the current directory\n",
    "txt_files = glob.glob('*.txt')\n",
    "\n",
    "# Delete each .txt file\n",
    "for txt_file in txt_files:\n",
    "  os.remove(txt_file)\n",
    "  print(f\"Deleted {txt_file}\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}