import re
import os
import email
from email.header import decode_header
from email.utils import parsedate_to_datetime


def parse_eml(eml_fp, attr_dir):
    """
    eml文件解析
    :params eml_fp: eml文件路徑
    :params attr_dir: 附件保存目錄
    """
    if not os.path.exists(attr_dir):
        os.makedirs(attr_dir)

    # 讀取eml文件
    with open(eml_fp, "r") as file:
        eml_content = file.read()
    # 轉為email對象
    msg = email.message_from_string(eml_content)

    # 郵件主題
    subject_bytes, subject_encode = decode_header(msg["Subject"])[0]
    if subject_encode:
        subject = subject_bytes.decode(subject_encode)
    else:
        subject = subject_bytes
    print("主題：", subject)

    # 郵件發件人
    from_ip = re.search("<(.*)>", msg["from"]).group(1)
    print("發件人郵箱：", from_ip)
    from_name = decode_header(msg["from"].split("<")[0].strip())
    if from_name:
        if from_name[0] and from_name[0][1]:
            from_n = from_name[0][0].decode(from_name[0][1])
        else:
            from_n = from_name[0][0]
    print("發件人名稱：", from_n)

    # 郵件時間
    received_date = parsedate_to_datetime(msg["date"])
    print("接收時間：", received_date)

    # 郵件正文及附件
    for par in msg.walk():
        if not par.is_multipart():  # 判斷是否為multipart，里面的數據不需要
            name = par.get_param("name")  # 獲取附件的文件名
            if name:  
                # 附件
                fname = decode_header(name)[0]
                if fname[1]:
                    attr_name = fname[0].decode(fname[1])
                else:
                    attr_name = fname[0]
                print("附件名:", attr_name)
                # 解碼附件內容
                attr_data = par.get_payload(decode=True)
                attr_fp = os.path.join(attr_dir, attr_name)
                with open(attr_fp, 'wb') as f_write:
                    f_write.write(attr_data)
            else:  
                # 正文
                text_char = par.get_content_charset()
                if "text/plain" in par["content-type"]:  # 文本正文
                    body = par.get_payload(decode=True).decode(text_char)
                    print("郵件正文：", body)
                else:  # html格式正文
                    html_body = par.get_payload(decode=True).decode(text_char)
                    print("HTML正文：", html_body)
            print("-" * 60)