PTT 文章網址轉換成看板與aid

get_aid_from_url.py

def get_aid_from_url(url: str) -> (str, str):

    # from get_aid_from_url in PyPtt

    # 檢查是否符合 PTT BBS 文章網址格式
    pattern = re.compile('https://www.ptt.cc/bbs/[-.\w]+/M.[\d]+.A[.\w]*.html')
    r = pattern.search(url)
    if r is None:
        raise ValueError('url must be www.ptt.cc article url')

    # 演算法參考 https://www.ptt.cc/man/C_Chat/DE98/DFF5/DB61/M.1419434423.A.DF0.html
    # aid 字元表
    aid_table = '0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz-_'

    board = url[23:]
    board = board[:board.find('/')]

    temp = url[url.rfind('/') + 1:].split('.')
    # print(temp)

    id_0 = int(temp[1])  # dec

    aid_0 = ''
    for _ in range(6):
        index = id_0 % 64
        aid_0 = f'{aid_table[index]}{aid_0}'
        id_0 = int(id_0 / 64)

    if temp[3] != 'html':
        id_1 = int(temp[3], 16)  # hex
        aid_1 = ''
        for _ in range(2):
            index = id_1 % 64
            aid_1 = f'{aid_table[index]}{aid_1}'
            id_1 = int(id_1 / 64)
    else:
        aid_1 = '00'

    aid = f'{aid_0}{aid_1}'

    return board, aid