#!/usr/bin/env python3

# LOGIN

# https://steam.readthedocs.io/en/latest/api/steam.webauth.html

import steam.webauth as wa
from os import environ

user = wa.WebAuth(environ['STEAM_USERNAME'])

try:
    user.login(environ['STEAM_PASSWORD'])
except (wa.LoginIncorrect) as exp:
    raise
except (wa.CaptchaRequired) as exp:
    print(user.captcha_url)
    user.login(password=environ['STEAM_PASSWORD'], captcha=input("Captcha: "))
except wa.EmailCodeRequired:
    user.login(email_code=input("Email Code: "))
except wa.TwoFactorCodeRequired:
    user.login(twofactor_code=input("2FA Code: "))

# CRAWL

from bs4 import BeautifulSoup
from datetime import datetime, timezone
import pytz
import re
from urllib.parse import urlparse
import requests

pdt = pytz.timezone("Us/Pacific")

def parse_trs(trs):
    for tr in trs:
        tds = tr.find_all('td')
        from_url = tds[0].find('a')['href']
        from_name = tds[0].text
        to_url = tds[1].find('a')['href']
        to_name = tds[1].text
        date = datetime.strptime(tds[2].text, '%b %d, %Y @ %I:%M%p PDT').replace(tzinfo=pdt)
        text = tds[3].text
        print(f'({date}) {from_name} -> {to_name}: {text}')
        # download steamuserimages
        for url_string in re.findall(r'(https?://\S+)', text):
            url = urlparse(url_string)
            if url.netloc.startswith('steamuserimages'):
                response = requests.get(url_string)
                with open(url.path.strip('/').replace('/', '_'), "wb") as f:
                    f.write(response.content)

# get first page

r = user.session.get('https://help.steampowered.com/en/accountdata/GetFriendMessagesLog')

soup = BeautifulSoup(r.text, 'html.parser')

continue_value = soup.find(class_='AccountDataLoadMore')['data-continuevalue']

account_data_table = soup.find(id='AccountDataTable_1')
trs = account_data_table.find_all('tr')[1:]

parse_trs(trs)

# get further pages

while True:
    r = user.session.get(f'https://help.steampowered.com/en/accountdata/AjaxLoadMoreData/?url=GetFriendMessagesLog&continue={continue_value}')
    continue_value = r.json()['continue']
    html = r.json()['html']
    trs = BeautifulSoup(html, 'html.parser').find_all('tr')
    parse_trs(trs)