Interactive game listing from Metacritics by user score

main.py

import streamlit as st
import pandas
import requests
from bs4 import BeautifulSoup
from datetime import datetime, timedelta
from IPython.core.display import display, HTML


def main():
    st.sidebar.title('Select page')
    st.sidebar.number_input('Page', step=1, min_value=0, on_change=load, key='number')


def image(path):
    return f'<img src="{path}" width="120" >'


def load():
    url = 'https://www.metacritic.com/browse/games/score/userscore/all/pc/filtered'
    query = {
        'sort': 'desc',
        'page': st.session_state.number or 0
    }
    headers = {
        'authority': 'www.metacritic.com',
        'accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.9',
        'accept-encoding': 'gzip, deflate, br',
        'accept-language': 'en-US,en;q=0.9,tr;q=0.8,vi;q=0.7,de;q=0.6,ru;q=0.5',
        'cache-control': 'no-cache',
        'cookie': 'mc_s_s=d_1; AMCVS_3C66570E5FE1A4AB0A495FFC%40AdobeOrg=1; s_cc=true; mc_ag=1; ctk=NjM0MTc1OWE1OGU2YWIzMGE3ODQxYzNhMjc4OQ%3D%3D; tmpid=1667017030218164; dw-tag=content_header_wrapper%3Bmain_content; prevPageType=list_page; s_sq=%5B%5BB%5D%5D; OptanonConsent=isIABGlobal=false&datestamp=Sat+Oct+29+2022+07%3A34%3A40+GMT%2B0300+(GMT%2B03%3A00)&version=6.20.0&hosts=&consentId=a09860bf-db3b-4a30-a0a0-6e39ccffa8c8&interactionCount=1&landingPath=NotLandingPage&groups=C0001%3A1%2CC0002%3A1%2CC0003%3A1%2CC0004%3A0&AwaitingReconsent=false; AMCV_3C66570E5FE1A4AB0A495FFC%40AdobeOrg=1585540135%7CMCMID%7C36401659451030133907163307877303367266%7CMCOPTOUT-1667025280s%7CNONE%7CvVersion%7C4.4.0; metapv=71; utag_main=v_id:0183b7b032520019bc6df6a1c0c805075001b06d00bd0$_sn:5$_ss:0$_st:1667019881830$vapi_domain:metacritic.com$_pn:71%3Bexp-session$ses_id:1667016422075%3Bexp-session',
        'pragma': 'no-cache',
        'referer': 'https://www.metacritic.com/browse/games/score/userscore/all/pc/filtered?sort=desc&page=5',
        'user-agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/106.0.0.0 Safari/537.36',
    }
    res = requests.get(url, params=query, headers=headers)
    soup = BeautifulSoup(res.content, 'lxml')
    games = soup.select('table.clamp-list tr')
    selected = []
    for game in games:
        if game.has_attr('class'):
            continue

        date_text = game.select('div.clamp-details span')[-1].text
        date = datetime.strptime(date_text, '%B %d, %Y')
        if date < datetime.now() - timedelta(days=6 * 365):
            continue

        row = {
            'image': game.select_one('td.clamp-image-wrap img').get('src'),
            'date': date.strftime('%B %d, %Y'),
            'name': game.select_one('a.title').text,
            'user_score': game.select_one('a.metascore_anchor div').text,
            'url': 'https://www.metacritic.com' + game.select_one('a.title').get('href'),
        }

        selected.append(row)

    df = pandas.DataFrame(selected)
    html = df.to_html(escape=False, formatters=dict(image=image))
    st.markdown(html, unsafe_allow_html=True)


if __name__ == '__main__':
    main()
    load()