svtplay-dl/lib/svtplay_dl/service/dr.py

# ex:ts=4:sw=4:sts=4:et
# -*- tab-width: 4; c-basic-offset: 4; indent-tabs-mode: nil -*-
from __future__ import absolute_import
import base64
import re
import json
import copy
import binascii
import hashlib
from urllib.parse import urljoin, urlparse

from cryptography.hazmat.primitives.ciphers import Cipher, algorithms, modes
from cryptography.hazmat.backends import default_backend

from svtplay_dl.service import Service, OpenGraphThumbMixin
from svtplay_dl.fetcher.hls import hlsparse
from svtplay_dl.fetcher.hds import hdsparse
from svtplay_dl.subtitle import subtitle
from svtplay_dl.error import ServiceError


class Dr(Service, OpenGraphThumbMixin):
    supported_domains = ['dr.dk']

    def get(self):
        data = self.get_urldata()

        match = re.search(r'resource:[ ]*"([^"]*)",', data)
        if match:
            resource_url = match.group(1)
            resource_data = self.http.request("get", resource_url).content
            resource = json.loads(resource_data)
            streams = self.find_stream(self.config, resource)
            for i in streams:
                yield i
        else:
            match = re.search(r'resource="([^"]*)"', data)
            if not match:
                yield ServiceError("Cant find resource info for this video")
                return
            if match.group(1)[:4] != "http":
                resource_url = "http:{0}".format(match.group(1))
            else:
                resource_url = match.group(1)
            resource_data = self.http.request("get", resource_url).text
            resource = json.loads(resource_data)

            if "Links" not in resource:
                yield ServiceError("Cant access this video. its geoblocked.")
                return
            if "SubtitlesList" in resource and len(resource["SubtitlesList"]) > 0:
                suburl = resource["SubtitlesList"][0]["Uri"]
                yield subtitle(copy.copy(self.config), "wrst", suburl, output=self.output)
            if "Data" in resource:
                streams = self.find_stream(self.config, resource)
                for i in streams:
                    yield i
            else:
                for stream in resource['Links']:
                    uri = stream["Uri"]
                    if uri is None:
                        uri = self._decrypt(stream["EncryptedUri"])

                    if stream["Target"] == "HDS":
                        streams = hdsparse(copy.copy(self.config),
                                           self.http.request("get", uri, params={"hdcore": "3.7.0"}),
                                           uri, output=self.output)
                        if streams:
                            for n in list(streams.keys()):
                                yield streams[n]
                    if stream["Target"] == "HLS":
                        streams = hlsparse(self.config, self.http.request("get", uri), uri, output=self.output)
                        for n in list(streams.keys()):
                            yield streams[n]

    def find_all_episodes(self, config):
        episodes = []
        matches = re.findall(r'<button class="show-more" data-url="([^"]+)" data-partial="([^"]+)"',
                             self.get_urldata())
        for encpath, enccomp in matches:
            newstyle = '_' in encpath
            if newstyle:
                encbasepath = encpath.split('_')[0]
                path = base64.b64decode(encbasepath + '===').decode('latin1')
            else:
                path = base64.b64decode(encpath + '===').decode('latin1')

            if '/view/' in path:
                continue

            params = 'offset=0&limit=1000'
            if newstyle:
                encparams = base64.b64encode(params.encode('latin1')).decode('latin1').rstrip('=')
                encpath = '{0}_{1}'.format(encbasepath, encparams)
            else:
                path = '{0}?{1}'.format(urlparse(path).path, params)
                encpath = base64.b64encode(path.encode('latin1')).decode('latin1').rstrip('=')

            url = urljoin('https://www.dr.dk/tv/partial/',
                          '{0}/{1}'.format(enccomp, encpath))
            data = self.http.request('get', url).content.decode('latin1')

            matches = re.findall(r'"program-link" href="([^"]+)">', data)
            episodes = [urljoin('https://www.dr.dk/', url) for url in matches]
            break

        if not episodes:
            prefix = '/'.join(urlparse(self.url).path.rstrip('/').split('/')[:-1])
            matches = re.findall(r'"program-link" href="([^"]+)">', self.get_urldata())
            episodes = [urljoin('https://www.dr.dk/', url)
                        for url in matches
                        if url.startswith(prefix)]

        if config.get("all_last") != -1:
            episodes = episodes[:config.get("all_last")]
        else:
            episodes.reverse()

        return episodes

    def find_stream(self, config, resource):
        tempresource = resource['Data'][0]['Assets']
        # To find the VideoResource, they have Images as well
        for resources in tempresource:
            if resources['Kind'] == 'VideoResource':
                links = resources['Links']
                break
        for i in links:
            if i["Target"] == "Ios" or i["Target"] == "HLS":
                streams = hlsparse(config, self.http.request("get", i["Uri"]), i["Uri"], output=self.output)
                for n in list(streams.keys()):
                    yield streams[n]

    def _decrypt(self, url):
        n = int(url[2:10], 16)
        iv_hex = url[10 + n:]
        data = binascii.a2b_hex(url[10:10 + n].encode('ascii'))
        key = hashlib.sha256(('%s:sRBzYNXBzkKgnjj8pGtkACch' % iv_hex).encode('utf-8')).digest()
        iv = bytes.fromhex(iv_hex)

        backend = default_backend()
        cipher = Cipher(algorithms.AES(key), modes.CBC(iv), backend=backend)
        decryptor = cipher.decryptor()
        decrypted = decryptor.update(data)
        return decrypted[:-decrypted[-1]].decode('utf-8')