cast/ChannelProvider/SVT/__init__.py

#!/usr/bin/env python3

import hashlib
import io
import json
import os
import pickle

import ChannelProvider
import feedparser
import requests
import wx
from bs4 import BeautifulSoup
from Items import Item


class SVT(ChannelProvider.ChannelProvider):
    m_cache: dict = dict()
    m_cachefile = '/tmp/svt_cache'

    def __init__(self) -> None:
        super().__init__('SVT', 'http://www.svtplay.se/rss.xml')

        if os.path.exists(self.m_cachefile):
            with open(self.m_cachefile, 'rb') as cachehandle:
                self.m_cache = pickle.load(cachehandle)
        self.m_items: list[Item] = self.parse_feed()

    def refresh_items(self):
        self.m_items: list[Item] = self.parse_feed()

    def parse_feed(self) -> list[Item]:
        feed = feedparser.parse(self.get_feed())
        entries = feed['entries']
        items: list[Item] = list()

        for entry in entries:
            key = hashlib.sha256(entry['link'].encode('utf-8')).hexdigest()

            if key in self.m_cache.keys():
                thumbnail_link = self.m_cache[key]['thumbnail_link']
                resolved_link = self.m_cache[key]['resolved_link']
                description = self.m_cache[key]['description']
                published_parsed = self.m_cache[key]['published_parsed']
                title = self.m_cache[key]['title']
            else:
                svt_id = ''

                for link in entry['links']:
                    if str(link['type']).startswith('image/'):
                        thumbnail_link = str(link['href'])
                        break
                page = requests.get(str(entry['link']))
                soup = BeautifulSoup(page.text, 'html.parser')

                for element in soup.find_all('a'):
                    href = element.get('href')
                    datart = element.get('data-rt')

                    if datart == 'top-area-play-button':
                        svt_id = href.split('=')[1].split('&')[0]

                api = json.loads(
                    requests.get(
                        'https://api.svt.se/video/{}'.format(svt_id)).text)
                resolved_link = ''

                for reference in api['videoReferences']:
                    if reference['format'] == "dashhbbtv":
                        resolved_link = reference['url']
                        print(resolved_link)
                description = str(entry['description'])
                published_parsed = entry['published_parsed']
                title = str(entry['title'])
            self.m_cache[key] = {'thumbnail_link': thumbnail_link}
            self.m_cache[key]['resolved_link'] = resolved_link
            self.m_cache[key]['description'] = description
            self.m_cache[key]['published_parsed'] = published_parsed
            self.m_cache[key]['title'] = title

            res = requests.get(thumbnail_link)
            content_bytes = io.BytesIO(res.content)
            image = wx.Image(content_bytes,
                             type=wx.BITMAP_TYPE_ANY,
                             index=-1)
            thumbnail = wx.Bitmap(image)
            item = Item(description, resolved_link, self.m_provider_name,
                        published_parsed, thumbnail, title)
            items.append(item)

        # write to cache file
        with open(self.m_cachefile, 'wb') as cachehandle:
            pickle.dump(self.m_cache, cachehandle)

        return items