[vrv] Add new extractor

2017-03-31 23:28:24 +01:00
parent 7453999580
commit 77c8ebe631
2 changed files with 152 additions and 0 deletions
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@@ -1182,6 +1182,7 @@ from .voxmedia import VoxMediaIE
 from .vporn import VpornIE
 from .vrt import VRTIE
 from .vrak import VrakIE
+from .vrv import VRVIE
 from .medialaan import MedialaanIE
 from .vube import VubeIE
 from .vuclip import VuClipIE
--- a/youtube_dl/extractor/vrv.py
+++ b/youtube_dl/extractor/vrv.py
@@ -0,0 +1,151 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+import json
+import hashlib
+import hmac
+import random
+import string
+import time
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_urlencode,
+    compat_urlparse,
+)
+from ..utils import (
+    float_or_none,
+    int_or_none,
+)
+
+
+class VRVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
+    _TEST = {
+        'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
+        'info_dict': {
+            'id': 'GR9PNZ396',
+            'ext': 'mp4',
+            'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
+            'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
+            'uploader_id': 'seeso',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+    _API_DOMAIN = None
+    _API_PARAMS = {}
+    _CMS_SIGNING = {}
+
+    def _call_api(self, path, video_id, note, data=None):
+        base_url = self._API_DOMAIN + '/core/' + path
+        encoded_query = compat_urllib_parse_urlencode({
+            'oauth_consumer_key': self._API_PARAMS['oAuthKey'],
+            'oauth_nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
+            'oauth_signature_method': 'HMAC-SHA1',
+            'oauth_timestamp': int(time.time()),
+            'oauth_version': '1.0',
+        })
+        headers = self.geo_verification_headers()
+        if data:
+            data = json.dumps(data).encode()
+            headers['Content-Type'] = 'application/json'
+        method = 'POST' if data else 'GET'
+        base_string = '&'.join([method, compat_urlparse.quote(base_url, ''), compat_urlparse.quote(encoded_query, '')])
+        oauth_signature = base64.b64encode(hmac.new(
+            (self._API_PARAMS['oAuthSecret'] + '&').encode('ascii'),
+            base_string.encode(), hashlib.sha1).digest()).decode()
+        encoded_query += '&oauth_signature=' + compat_urlparse.quote(oauth_signature, '')
+        return self._download_json(
+            '?'.join([base_url, encoded_query]), video_id,
+            note='Downloading %s JSON metadata' % note, headers=headers, data=data)
+
+    def _call_cms(self, path, video_id, note):
+        return self._download_json(
+            self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
+            note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
+
+    def _set_api_params(self, webpage, video_id):
+        if not self._API_PARAMS:
+            self._API_PARAMS = self._parse_json(self._search_regex(
+                r'window\.__APP_CONFIG__\s*=\s*({.+?})</script>',
+                webpage, 'api config'), video_id)['cxApiParams']
+            self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
+
+    def _set_cms_signing(self, video_id):
+        if not self._CMS_SIGNING:
+            self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, video_id,
+            headers=self.geo_verification_headers())
+        media_resource = self._parse_json(self._search_regex(
+            r'window\.__INITIAL_STATE__\s*=\s*({.+?})</script>',
+            webpage, 'inital state'), video_id).get('watch', {}).get('mediaResource') or {}
+
+        video_data = media_resource.get('json')
+        if not video_data:
+            self._set_api_params(webpage, video_id)
+            episode_path = self._call_api('cms_resource', video_id, 'episode resource path', data={
+                'resource_key': 'cms:/episodes/' + video_id,
+            })['__links__']['cms_resource']['href']
+            self._set_cms_signing(video_id)
+            video_data = self._call_cms(episode_path, video_id, 'video')
+        title = video_data['title']
+
+        streams_json = media_resource.get('streams', {}).get('json', {})
+        if not streams_json:
+            self._set_api_params(webpage, video_id)
+            streams_path = video_data['__links__']['streams']['href']
+            self._set_cms_signing(video_id)
+            streams_json = self._call_cms(streams_path, video_id, 'streams')
+
+        audio_locale = streams_json.get('audio_locale')
+        formats = []
+        for stream_id, stream in streams_json.get('streams', {}).get('adaptive_hls', {}).items():
+            stream_url = stream.get('url')
+            if not stream_url:
+                continue
+            stream_id = stream_id or audio_locale
+            m3u8_formats = self._extract_m3u8_formats(
+                stream_url, video_id, 'mp4', m3u8_id=stream_id,
+                note='Downloading %s m3u8 information' % stream_id,
+                fatal=False)
+            if audio_locale:
+                for f in m3u8_formats:
+                    f['language'] = audio_locale
+            formats.extend(m3u8_formats)
+        self._sort_formats(formats)
+
+        thumbnails = []
+        for thumbnail in video_data.get('images', {}).get('thumbnails', []):
+            thumbnail_url = thumbnail.get('source')
+            if not thumbnail_url:
+                continue
+            thumbnails.append({
+                'url': thumbnail_url,
+                'width': int_or_none(thumbnail.get('width')),
+                'height': int_or_none(thumbnail.get('height')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'description': video_data.get('description'),
+            'duration': float_or_none(video_data.get('duration_ms'), 1000),
+            'uploader_id': video_data.get('channel_id'),
+            'series': video_data.get('series_title'),
+            'season': video_data.get('season_title'),
+            'season_number': int_or_none(video_data.get('season_number')),
+            'season_id': video_data.get('season_id'),
+            'episode': title,
+            'episode_number': int_or_none(video_data.get('episode_number')),
+            'episode_id': video_data.get('production_episode_id'),
+        }