libe.py

from typing import Any
from urllib import request, parse as urlparse
from re import compile, Pattern
from json import dumps
from sys import stderr
from signal import signal, SIGPIPE, SIG_DFL

from bs4 import BeautifulSoup


signal(SIGPIPE, SIG_DFL)


BASE_URL = 'https://libe-tokyo.com/'
PATTERN_GENERAL = compile(r'.+[：:](?P<text>.+)')
PATTERN_NUMBER = compile(r'.+[：:](?P<text>\d+)')
PATTERN_NAME = compile(r'名前[：:](?:\[(?P<tag>.+)\])?(?P<name>[^(（)]+)(?:[（(](?P<ruby>.+)[)）])?')


def soup_open(path: str) -> BeautifulSoup:
    url = urlparse.urljoin(BASE_URL, path)
    print('Opening %s' % url, file=stderr)

    with request.urlopen(url) as response:
        return BeautifulSoup(response, 'html.parser')


def soup_find(src: BeautifulSoup, pattern: Pattern, prefix: str = None, group: str = 'text') -> Any:
    for tag in src.find_all('dd'):
        if prefix != None and not tag.text.startswith(prefix):
            continue

        m = pattern.match(tag.text)
        if m == None:
            continue

        if group != None:
            return m.group(group)
        
        return m.groupdict()


def int_safe(src: str) -> int:
    if src == None:
        return None

    return int(src)


def parse_bool(src: BeautifulSoup, prefix: str) -> bool:
    value = soup_find(src, PATTERN_GENERAL, prefix)
    if value == None:
        return None
    
    return value == '○' or value == '◎'


def parse_girl(path: str) -> dict:
    soup = soup_open(path)
    profile_section = soup.find('dt', attrs={'class': 'profile'}, text='プロフィール').parent
    options_section = soup.find('dt', attrs={'class': 'profile'}, text='可能プレイ&オプション').parent
    
    girl = {
        'name': soup_find(profile_section, PATTERN_NAME, group=None),
        'age': int_safe(soup_find(profile_section, PATTERN_NUMBER, '年齢')),
        'sizes': {
            't': int_safe(soup_find(profile_section, PATTERN_NUMBER, 'T')),
            'b': int_safe(soup_find(profile_section, PATTERN_NUMBER, 'B')),
            'p': int_safe(soup_find(profile_section, PATTERN_NUMBER, 'P')),
        },
        'types': soup_find(profile_section, PATTERN_GENERAL, 'タイプ').split('/'),
        'options': {
            'af': parse_bool(options_section, 'アナル受け'),
            'reversed_af': parse_bool(options_section, '逆アナル'),
            '3p': parse_bool(options_section, '3P'),
            'reversed_3p': parse_bool(options_section, '逆3P'),
            'soft_s': parse_bool(options_section, 'ソフトS'),
            'soft_m': parse_bool(options_section, 'ソフトM'),
        },
    }

    return girl


soup = soup_open('/index2.html')
sections = soup.find_all('div',attrs={
    "class": ["girls-left", "girls-right"],
})

girls = []
for section in sections:
    for anchor in section.find_all('a'):
        girls.append(parse_girl(anchor['href']))

print(dumps(girls))

run.sh

curl "https://gist.githubusercontent.com/siketyan/a43cbf286d4c70b2abf2dc40bc04900a/raw/e1b4440a35a41250d59dd55d5a92d5a804ff9f62/libe.py" | python3 | jq "map(select(.options.reversed_af)) | sort_by(.sizes.p) | reverse | map(.name.name)"