[gaia] Add new extractor(#14605)

2 files changed, 99 insertions, 0 deletions

diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
index ddeb70284..c7d04d366 100644
--- a/youtube_dl/extractor/extractors.py
+++ b/youtube_dl/extractor/extractors.py
@@ -411,6 +411,7 @@ from .funk import (
 from .funnyordie import FunnyOrDieIE
 from .fusion import FusionIE
 from .fxnetworks import FXNetworksIE
+from .gaia import GaiaIE
 from .gameinformer import GameInformerIE
 from .gameone import (
     GameOneIE,
diff --git a/youtube_dl/extractor/gaia.py b/youtube_dl/extractor/gaia.py
new file mode 100644
index 000000000..f2eef3f4c
--- /dev/null
+++ b/youtube_dl/extractor/gaia.py
@@ -0,0 +1,98 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    str_or_none,
+    strip_or_none,
+    try_get,
+)
+
+
+class GaiaIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?gaia\.com/video/(?P<id>[^/?]+).*?\bfullplayer=(?P<type>feature|preview)'
+    _TESTS = [{
+        'url': 'https://www.gaia.com/video/connecting-universal-consciousness?fullplayer=feature',
+        'info_dict': {
+            'id': '89356',
+            'ext': 'mp4',
+            'title': 'Connecting with Universal Consciousness',
+            'description': 'md5:844e209ad31b7d31345f5ed689e3df6f',
+            'upload_date': '20151116',
+            'timestamp': 1447707266,
+            'duration': 936,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.gaia.com/video/connecting-universal-consciousness?fullplayer=preview',
+        'info_dict': {
+            'id': '89351',
+            'ext': 'mp4',
+            'title': 'Connecting with Universal Consciousness',
+            'description': 'md5:844e209ad31b7d31345f5ed689e3df6f',
+            'upload_date': '20151116',
+            'timestamp': 1447707266,
+            'duration': 53,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }]
+
+    def _real_extract(self, url):
+        display_id, vtype = re.search(self._VALID_URL, url).groups()
+        node_id = self._download_json(
+            'https://brooklyn.gaia.com/pathinfo', display_id, query={
+                'path': 'video/' + display_id,
+            })['id']
+        node = self._download_json(
+            'https://brooklyn.gaia.com/node/%d' % node_id, node_id)
+        vdata = node[vtype]
+        media_id = compat_str(vdata['nid'])
+        title = node['title']
+
+        media = self._download_json(
+            'https://brooklyn.gaia.com/media/' + media_id, media_id)
+        formats = self._extract_m3u8_formats(
+            media['mediaUrls']['bcHLS'], media_id, 'mp4')
+        self._sort_formats(formats)
+
+        subtitles = {}
+        text_tracks = media.get('textTracks', {})
+        for key in ('captions', 'subtitles'):
+            for lang, sub_url in text_tracks.get(key, {}).items():
+                subtitles.setdefault(lang, []).append({
+                    'url': sub_url,
+                })
+
+        fivestar = node.get('fivestar', {})
+        fields = node.get('fields', {})
+
+        def get_field_value(key, value_key='value'):
+            return try_get(fields, lambda x: x[key][0][value_key])
+
+        return {
+            'id': media_id,
+            'display_id': display_id,
+            'title': title,
+            'formats': formats,
+            'description': strip_or_none(get_field_value('body') or get_field_value('teaser')),
+            'timestamp': int_or_none(node.get('created')),
+            'subtitles': subtitles,
+            'duration': int_or_none(vdata.get('duration')),
+            'like_count': int_or_none(try_get(fivestar, lambda x: x['up_count']['value'])),
+            'dislike_count': int_or_none(try_get(fivestar, lambda x: x['down_count']['value'])),
+            'comment_count': int_or_none(node.get('comment_count')),
+            'series': try_get(node, lambda x: x['series']['title'], compat_str),
+            'season_number': int_or_none(get_field_value('season')),
+            'season_id': str_or_none(get_field_value('series_nid', 'nid')),
+            'episode_number': int_or_none(get_field_value('episode')),
+        }