update from upstream

1 files changed, 220 insertions, 14 deletions

diff --git a/hypervideo_dl/extractor/vidio.py b/hypervideo_dl/extractor/vidio.py
index b1243e8..571448b 100644
--- a/hypervideo_dl/extractor/vidio.py
+++ b/hypervideo_dl/extractor/vidio.py
@@ -1,19 +1,80 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
+    ExtractorError,
+    get_element_by_class,
     int_or_none,
     parse_iso8601,
+    smuggle_url,
     str_or_none,
     strip_or_none,
     try_get,
+    unsmuggle_url,
+    urlencode_postdata,
 )
 
 
-class VidioIE(InfoExtractor):
+class VidioBaseIE(InfoExtractor):
+    _LOGIN_URL = 'https://www.vidio.com/users/login'
+    _NETRC_MACHINE = 'vidio'
+
+    def _login(self):
+        username, password = self._get_login_info()
+        if username is None:
+            return
+
+        def is_logged_in():
+            res = self._download_json(
+                'https://www.vidio.com/interactions.json', None, 'Checking if logged in', fatal=False) or {}
+            return bool(res.get('current_user'))
+
+        if is_logged_in():
+            return
+
+        login_page = self._download_webpage(
+            self._LOGIN_URL, None, 'Downloading log in page')
+
+        login_form = self._form_hidden_inputs("login-form", login_page)
+        login_form.update({
+            'user[login]': username,
+            'user[password]': password,
+        })
+        login_post, login_post_urlh = self._download_webpage_handle(
+            self._LOGIN_URL, None, 'Logging in', data=urlencode_postdata(login_form), expected_status=[302, 401])
+
+        if login_post_urlh.status == 401:
+            if get_element_by_class('onboarding-content-register-popup__title', login_post):
+                raise ExtractorError(
+                    'Unable to log in: The provided email has not registered yet.', expected=True)
+
+            reason = get_element_by_class('onboarding-form__general-error', login_post) or get_element_by_class('onboarding-modal__title', login_post)
+            if 'Akun terhubung ke' in reason:
+                raise ExtractorError(
+                    'Unable to log in: Your account is linked to a social media account. '
+                    'Use --cookies to provide account credentials instead', expected=True)
+            elif reason:
+                subreason = get_element_by_class('onboarding-modal__description-text', login_post) or ''
+                raise ExtractorError(
+                    'Unable to log in: %s. %s' % (reason, clean_html(subreason)), expected=True)
+            raise ExtractorError('Unable to log in')
+
+    def _real_initialize(self):
+        self._api_key = self._download_json(
+            'https://www.vidio.com/auth', None, data=b'')['api_key']
+        self._login()
+
+    def _call_api(self, url, video_id, note=None):
+        return self._download_json(url, video_id, note=note, headers={
+            'Content-Type': 'application/vnd.api+json',
+            'X-API-KEY': self._api_key,
+        })
+
+
+class VidioIE(VidioBaseIE):
     _VALID_URL = r'https?://(?:www\.)?vidio\.com/watch/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
     _TESTS = [{
         'url': 'http://www.vidio.com/watch/165683-dj_ambred-booyah-live-2015',
@@ -41,24 +102,43 @@ class VidioIE(InfoExtractor):
     }, {
         'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north',
         'only_matching': True,
+    }, {
+        # Premier-exclusive video
+        'url': 'https://www.vidio.com/watch/1550718-stand-by-me-doraemon',
+        'only_matching': True
     }]
 
-    def _real_initialize(self):
-        self._api_key = self._download_json(
-            'https://www.vidio.com/auth', None, data=b'')['api_key']
-
     def _real_extract(self, url):
-        video_id, display_id = re.match(self._VALID_URL, url).groups()
-        data = self._download_json(
-            'https://api.vidio.com/videos/' + video_id, display_id, headers={
-                'Content-Type': 'application/vnd.api+json',
-                'X-API-KEY': self._api_key,
-            })
+        match = self._match_valid_url(url).groupdict()
+        video_id, display_id = match.get('id'), match.get('display_id')
+        data = self._call_api('https://api.vidio.com/videos/' + video_id, display_id)
         video = data['videos'][0]
         title = video['title'].strip()
+        is_premium = video.get('is_premium')
+
+        if is_premium:
+            sources = self._download_json(
+                'https://www.vidio.com/interactions_stream.json?video_id=%s&type=videos' % video_id,
+                display_id, note='Downloading premier API JSON')
+            if not (sources.get('source') or sources.get('source_dash')):
+                self.raise_login_required('This video is only available for registered users with the appropriate subscription')
+
+            formats, subs = [], {}
+            if sources.get('source'):
+                hls_formats, hls_subs = self._extract_m3u8_formats_and_subtitles(
+                    sources['source'], display_id, 'mp4', 'm3u8_native')
+                formats.extend(hls_formats)
+                subs.update(hls_subs)
+            if sources.get('source_dash'):  # TODO: Find video example with source_dash
+                dash_formats, dash_subs = self._extract_mpd_formats_and_subtitles(
+                    sources['source_dash'], display_id, 'dash')
+                formats.extend(dash_formats)
+                subs.update(dash_subs)
+        else:
+            hls_url = data['clips'][0]['hls_url']
+            formats, subs = self._extract_m3u8_formats_and_subtitles(
+                hls_url, display_id, 'mp4', 'm3u8_native')
 
-        formats = self._extract_m3u8_formats(
-            data['clips'][0]['hls_url'], display_id, 'mp4', 'm3u8_native')
         self._sort_formats(formats)
 
         get_first = lambda x: try_get(data, lambda y: y[x + 's'][0], dict) or {}
@@ -76,6 +156,7 @@ class VidioIE(InfoExtractor):
             'duration': int_or_none(video.get('duration')),
             'like_count': get_count('likes'),
             'formats': formats,
+            'subtitles': subs,
             'uploader': user.get('name'),
             'timestamp': parse_iso8601(video.get('created_at')),
             'uploader_id': username,
@@ -87,3 +168,128 @@ class VidioIE(InfoExtractor):
             'comment_count': get_count('comments'),
             'tags': video.get('tag_list'),
         }
+
+
+class VidioPremierIE(VidioBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?vidio\.com/premier/(?P<id>\d+)/(?P<display_id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.vidio.com/premier/2885/badai-pasti-berlalu',
+        'playlist_mincount': 14,
+    }, {
+        # Series with both free and premier-exclusive videos
+        'url': 'https://www.vidio.com/premier/2567/sosmed',
+        'only_matching': True,
+    }]
+
+    def _playlist_entries(self, playlist_url, display_id):
+        index = 1
+        while playlist_url:
+            playlist_json = self._call_api(playlist_url, display_id, 'Downloading API JSON page %s' % index)
+            for video_json in playlist_json.get('data', []):
+                link = video_json['links']['watchpage']
+                yield self.url_result(link, 'Vidio', video_json['id'])
+            playlist_url = try_get(playlist_json, lambda x: x['links']['next'])
+            index += 1
+
+    def _real_extract(self, url):
+        url, idata = unsmuggle_url(url, {})
+        playlist_id, display_id = self._match_valid_url(url).groups()
+
+        playlist_url = idata.get('url')
+        if playlist_url:  # Smuggled data contains an API URL. Download only that playlist
+            playlist_id = idata['id']
+            return self.playlist_result(
+                self._playlist_entries(playlist_url, playlist_id),
+                playlist_id=playlist_id, playlist_title=idata.get('title'))
+
+        playlist_data = self._call_api('https://api.vidio.com/content_profiles/%s/playlists' % playlist_id, display_id)
+
+        return self.playlist_from_matches(
+            playlist_data.get('data', []), playlist_id=playlist_id, ie=self.ie_key(),
+            getter=lambda data: smuggle_url(url, {
+                'url': data['relationships']['videos']['links']['related'],
+                'id': data['id'],
+                'title': try_get(data, lambda x: x['attributes']['name'])
+            }))
+
+
+class VidioLiveIE(VidioBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?vidio\.com/live/(?P<id>\d+)-(?P<display_id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.vidio.com/live/204-sctv',
+        'info_dict': {
+            'id': '204',
+            'title': 'SCTV',
+            'uploader': 'SCTV',
+            'uploader_id': 'sctv',
+            'thumbnail': r're:^https?://.*\.jpg$',
+        },
+    }, {
+        # Premier-exclusive livestream
+        'url': 'https://www.vidio.com/live/6362-tvn',
+        'only_matching': True,
+    }, {
+        # DRM premier-exclusive livestream
+        'url': 'https://www.vidio.com/live/6299-bein-1',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id, display_id = self._match_valid_url(url).groups()
+        stream_data = self._call_api(
+            'https://www.vidio.com/api/livestreamings/%s/detail' % video_id, display_id)
+        stream_meta = stream_data['livestreamings'][0]
+        user = stream_data.get('users', [{}])[0]
+
+        title = stream_meta.get('title')
+        username = user.get('username')
+
+        formats = []
+        if stream_meta.get('is_drm'):
+            if not self.get_param('allow_unplayable_formats'):
+                self.report_drm(video_id)
+        if stream_meta.get('is_premium'):
+            sources = self._download_json(
+                'https://www.vidio.com/interactions_stream.json?video_id=%s&type=livestreamings' % video_id,
+                display_id, note='Downloading premier API JSON')
+            if not (sources.get('source') or sources.get('source_dash')):
+                self.raise_login_required('This video is only available for registered users with the appropriate subscription')
+
+            if str_or_none(sources.get('source')):
+                token_json = self._download_json(
+                    'https://www.vidio.com/live/%s/tokens' % video_id,
+                    display_id, note='Downloading HLS token JSON', data=b'')
+                formats.extend(self._extract_m3u8_formats(
+                    sources['source'] + '?' + token_json.get('token', ''), display_id, 'mp4', 'm3u8_native'))
+            if str_or_none(sources.get('source_dash')):
+                pass
+        else:
+            if stream_meta.get('stream_token_url'):
+                token_json = self._download_json(
+                    'https://www.vidio.com/live/%s/tokens' % video_id,
+                    display_id, note='Downloading HLS token JSON', data=b'')
+                formats.extend(self._extract_m3u8_formats(
+                    stream_meta['stream_token_url'] + '?' + token_json.get('token', ''),
+                    display_id, 'mp4', 'm3u8_native'))
+            if stream_meta.get('stream_dash_url'):
+                pass
+            if stream_meta.get('stream_url'):
+                formats.extend(self._extract_m3u8_formats(
+                    stream_meta['stream_url'], display_id, 'mp4', 'm3u8_native'))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'is_live': True,
+            'description': strip_or_none(stream_meta.get('description')),
+            'thumbnail': stream_meta.get('image'),
+            'like_count': int_or_none(stream_meta.get('like')),
+            'dislike_count': int_or_none(stream_meta.get('dislike')),
+            'formats': formats,
+            'uploader': user.get('name'),
+            'timestamp': parse_iso8601(stream_meta.get('start_time')),
+            'uploader_id': username,
+            'uploader_url': 'https://www.vidio.com/@' + username if username else None,
+        }