debian/control: Update list of extractors in long description.

[youtubedl] / youtube_dl / extractor / canvas.py

diff --git a/youtube_dl/extractor/canvas.py b/youtube_dl/extractor/canvas.py
index 6899f8431788fad62e4615a33d95f3ffa65f4ea1..174fd9e2b4b00b384b1737a1df058050e4603de9 100644 (file)
--- a/youtube_dl/extractor/canvas.py
+++ b/youtube_dl/extractor/canvas.py
@@ -1,16 +1,23 @@
 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

+import json

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

+from .gigya import GigyaBaseIE
+from ..compat import compat_HTTPError

 from ..utils import (
 from ..utils import (

-    float_or_none,
+    ExtractorError,

     strip_or_none,
     strip_or_none,

+    float_or_none,
+    int_or_none,
+    merge_dicts,
+    parse_iso8601,

 )
 
 
 class CanvasIE(InfoExtractor):
 )
 
 
 class CanvasIE(InfoExtractor):

-    _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet)/assets/(?P<id>m[dz]-ast-[^/?#&]+)'
+    _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrtvideo)/assets/(?P<id>[^/?#&]+)'

     _TESTS = [{
         'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
         'md5': '90139b746a0a9bd7bb631283f6e2a64e',
     _TESTS = [{
         'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475',
         'md5': '90139b746a0a9bd7bb631283f6e2a64e',


@@ -166,3 +173,143 @@ class CanvasEenIE(InfoExtractor):
             'title': title,
             'description': self._og_search_description(webpage),
         }
             'title': title,
             'description': self._og_search_description(webpage),
         }

+
+
+class VrtNUIE(GigyaBaseIE):
+    IE_DESC = 'VrtNU.be'
+    _VALID_URL = r'https?://(?:www\.)?vrt\.be/(?P<site_id>vrtnu)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.vrt.be/vrtnu/a-z/postbus-x/1/postbus-x-s1a1/',
+        'info_dict': {
+            'id': 'pbs-pub-2e2d8c27-df26-45c9-9dc6-90c78153044d$vid-90c932b1-e21d-4fb8-99b1-db7b49cf74de',
+            'ext': 'flv',
+            'title': 'De zwarte weduwe',
+            'description': 'md5:d90c21dced7db869a85db89a623998d4',
+            'duration': 1457.04,
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'season': '1',
+            'season_number': 1,
+            'episode_number': 1,
+        },
+        'skip': 'This video is only available for registered users'
+    }]
+    _NETRC_MACHINE = 'vrtnu'
+    _APIKEY = '3_0Z2HujMtiWq_pkAjgnS2Md2E11a1AwZjYiBETtwNE-EoEHDINgtnvcAOpNgmrVGy'
+    _CONTEXT_ID = 'R3595707040'
+
+    def _real_initialize(self):
+        self._login()
+
+    def _login(self):
+        username, password = self._get_login_info()
+        if username is None:
+            return
+
+        auth_data = {
+            'APIKey': self._APIKEY,
+            'targetEnv': 'jssdk',
+            'loginID': username,
+            'password': password,
+            'authMode': 'cookie',
+        }
+
+        auth_info = self._gigya_login(auth_data)
+
+        # Sometimes authentication fails for no good reason, retry
+        login_attempt = 1
+        while login_attempt <= 3:
+            try:
+                # When requesting a token, no actual token is returned, but the
+                # necessary cookies are set.
+                self._request_webpage(
+                    'https://token.vrt.be',
+                    None, note='Requesting a token', errnote='Could not get a token',
+                    headers={
+                        'Content-Type': 'application/json',
+                        'Referer': 'https://www.vrt.be/vrtnu/',
+                    },
+                    data=json.dumps({
+                        'uid': auth_info['UID'],
+                        'uidsig': auth_info['UIDSignature'],
+                        'ts': auth_info['signatureTimestamp'],
+                        'email': auth_info['profile']['email'],
+                    }).encode('utf-8'))
+            except ExtractorError as e:
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                    login_attempt += 1
+                    self.report_warning('Authentication failed')
+                    self._sleep(1, None, msg_template='Waiting for %(timeout)s seconds before trying again')
+                else:
+                    raise e
+            else:
+                break
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage, urlh = self._download_webpage_handle(url, display_id)
+
+        info = self._search_json_ld(webpage, display_id, default={})
+
+        # title is optional here since it may be extracted by extractor
+        # that is delegated from here
+        title = strip_or_none(self._html_search_regex(
+            r'(?ms)<h1 class="content__heading">(.+?)</h1>',
+            webpage, 'title', default=None))
+
+        description = self._html_search_regex(
+            r'(?ms)<div class="content__description">(.+?)</div>',
+            webpage, 'description', default=None)
+
+        season = self._html_search_regex(
+            [r'''(?xms)<div\ class="tabs__tab\ tabs__tab--active">\s*
+                    <span>seizoen\ (.+?)</span>\s*
+                </div>''',
+             r'<option value="seizoen (\d{1,3})" data-href="[^"]+?" selected>'],
+            webpage, 'season', default=None)
+
+        season_number = int_or_none(season)
+
+        episode_number = int_or_none(self._html_search_regex(
+            r'''(?xms)<div\ class="content__episode">\s*
+                    <abbr\ title="aflevering">afl</abbr>\s*<span>(\d+)</span>
+                </div>''',
+            webpage, 'episode_number', default=None))
+
+        release_date = parse_iso8601(self._html_search_regex(
+            r'(?ms)<div class="content__broadcastdate">\s*<time\ datetime="(.+?)"',
+            webpage, 'release_date', default=None))
+
+        # If there's a ? or a # in the URL, remove them and everything after
+        clean_url = urlh.geturl().split('?')[0].split('#')[0].strip('/')
+        securevideo_url = clean_url + '.mssecurevideo.json'
+
+        try:
+            video = self._download_json(securevideo_url, display_id)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                self.raise_login_required()
+            raise
+
+        # We are dealing with a '../<show>.relevant' URL
+        redirect_url = video.get('url')
+        if redirect_url:
+            return self.url_result(self._proto_relative_url(redirect_url, 'https:'))
+
+        # There is only one entry, but with an unknown key, so just get
+        # the first one
+        video_id = list(video.values())[0].get('videoid')
+
+        return merge_dicts(info, {
+            '_type': 'url_transparent',
+            'url': 'https://mediazone.vrt.be/api/v1/vrtvideo/assets/%s' % video_id,
+            'ie_key': CanvasIE.ie_key(),
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'season': season,
+            'season_number': season_number,
+            'episode_number': episode_number,
+            'release_date': release_date,
+        })