[revision3] Add new extractor(closes #6388)

- revision3.com
- testtube.com
- animalist.com

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index b3f7059e441a79107a8791ae2a140e11e7199f5e..7adce5499ba248740d6501e82e3f623f7a610fa2 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -557,6 +557,7 @@ from .redtube import RedTubeIE
 from .regiotv import RegioTVIE
 from .restudy import RestudyIE
 from .reverbnation import ReverbNationIE
+from .revision3 import Revision3IE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE

diff --git a/youtube_dl/extractor/revision3.py b/youtube_dl/extractor/revision3.py
new file mode 100644 (file)
index 0000000..25fe4ef
--- /dev/null
+++ b/youtube_dl/extractor/revision3.py
@@ -0,0 +1,103 @@
+# 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,
+    parse_iso8601,
+    unescapeHTML,
+)
+
+
+class Revision3IE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|testtube|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)'
+    _TESTS = [{
+        'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016',
+        'md5': 'd94a72d85d0a829766de4deb8daaf7df',
+        'info_dict': {
+            'id': '73034',
+            'ext': 'webm',
+            'title': '5 Google Predictions for 2016',
+            'description': 'Google had a great 2015, but it\'s already time to look ahead. Here are our five predictions for 2016.',
+            'upload_date': '20151228',
+            'timestamp': 1451325600,
+            'duration': 187,
+        }
+    }, {
+        'url': 'http://testtube.com/brainstuff',
+        'info_dict': {
+            'id': '251',
+            'title': 'BrainStuff',
+            'description': 'Whether the topic is popcorn or particle physics, you can count on the HowStuffWorks team to explore-and explain-the everyday science in the world around us on BrainStuff.',
+        },
+        'playlist_mincount': 93,
+    }]
+    _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s'
+    _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62'
+
+    def _real_extract(self, url):
+        domain, display_id = re.match(self._VALID_URL, url).groups()
+        page_info = self._download_json(
+            self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id)
+
+        if page_info['data']['type'] == 'episode':
+            episode_data = page_info['data']
+            video_id = compat_str(episode_data['video']['data']['id'])
+            video_data = self._download_json(
+                'http://revision3.com/api/getPlaylist.json?api_key=%s&codecs=h264,vp8,theora&video_id=%s' % (self._API_KEY, video_id),
+                video_id)['items'][0]
+
+            formats = []
+            for media_type, media in video_data['media'].items():
+                for quality_id, quality in media.items():
+                    if quality_id == 'hls':
+                        formats.extend(self._extract_m3u8_formats(
+                            quality['url'], video_id, 'mp4',
+                            'm3u8_native', m3u8_id='hls', fatal=False))
+                    else:
+                        formats.append({
+                            'url': quality['url'],
+                            'format_id': '%s-%s' % (media_type, quality_id),
+                            'tbr': int_or_none(quality.get('bitrate')),
+                        })
+            self._sort_formats(formats)
+
+            thumbnails = [{
+                'url': image_url,
+                'id': image_id,
+            } for image_id, image_url in video_data.get('images', {}).items()]
+
+            return {
+                'id': video_id,
+                'title': unescapeHTML(video_data['title']),
+                'description': unescapeHTML(video_data.get('summary')),
+                'timestamp': parse_iso8601(episode_data.get('publishTime'), ' '),
+                'author': episode_data.get('author'),
+                'duration': int_or_none(video_data.get('duration')),
+                'thumbnails': thumbnails,
+                'formats': formats,
+            }
+        else:
+            show_data = page_info['show']['data']
+            episodes_data = page_info['episodes']['data']
+            num_episodes = page_info['meta']['totalEpisodes']
+            processed_episodes = 0
+            entries = []
+            page_num = 1
+            while True:
+                entries.extend([self.url_result(
+                    url + '/%s' % episode['slug']) for episode in episodes_data])
+                processed_episodes += len(episodes_data)
+                if processed_episodes == num_episodes:
+                    break
+                page_num += 1
+                episodes_data = self._download_json(self._PAGE_DATA_TEMPLATE % (
+                    domain, display_id + '/' + compat_str(page_num), domain),
+                    display_id)['episodes']['data']
+
+            return self.playlist_result(
+                entries, compat_str(show_data['id']),
+                show_data.get('name'), show_data.get('summary'))