Merge remote-tracking branch 'duncankl/airmozilla'
[youtube-dl] / youtube_dl / extractor / hitbox.py
index 239da3cd4738d27c62c990a7a72c1dea1ef26f07..84bd7c0804eb2966c7062288a1a190c8509f62fc 100644 (file)
@@ -1,35 +1,22 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
-    unified_strdate,
+    clean_html,
+    parse_iso8601,
+    float_or_none,
+    int_or_none,
+    compat_str,
 )
 
 
 class HitboxIE(InfoExtractor):
+    IE_NAME = 'hitbox'
     _VALID_URL = r'https?://(?:www\.)?hitbox\.tv/video/(?P<id>[0-9]+)'
-    _TESTS = [{
-        'url': 'http://www.hitbox.tv/video/358062',
-        'info_dict': {
-            'id': '358062',
-            'title': 'Megaman',
-            'alt_title': 'Megaman',
-            'description': '',
-            'ext': 'mp4',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': 3834,
-            'resolution': 'SD 480p',
-            'uploader_id': 'supergreatfriend',
-            'view_count': int,
-            'upload_date': '20141225',
-            'categories': [None],
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }, {
+    _TEST = {
         'url': 'http://www.hitbox.tv/video/203213',
         'info_dict': {
             'id': '203213',
@@ -38,10 +25,11 @@ class HitboxIE(InfoExtractor):
             'description': '',
             'ext': 'mp4',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': 215,
+            'duration': 215.1666,
             'resolution': 'HD 720p',
-            'uploader_id': 'hitboxlive',
+            'uploader': 'hitboxlive',
             'view_count': int,
+            'timestamp': 1407576133,
             'upload_date': '20140809',
             'categories': ['Live Show'],
         },
@@ -49,24 +37,29 @@ class HitboxIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
-    }]
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
+    }
 
+    def _extract_metadata(self, url, video_id):
         thumb_base = 'https://edge.sf.hitbox.tv'
         metadata = self._download_json(
-            'https://www.hitbox.tv/api/media/video/%s' % (video_id), video_id
-        )
+            '%s/%s' % (url, video_id), video_id)
+
+        date = 'media_live_since'
+        media_type = 'livestream'
+        if metadata.get('media_type') == 'video':
+            media_type = 'video'
+            date = 'media_date_added'
 
-        video_meta = metadata.get('video', [])[0]
+        video_meta = metadata.get(media_type, [])[0]
         title = video_meta.get('media_status')
         alt_title = video_meta.get('media_title')
-        description = video_meta.get('media_description')
-        duration = int(float(video_meta.get('media_duration')))
+        description = clean_html(
+            video_meta.get('media_description') or
+            video_meta.get('media_description_md'))
+        duration = float_or_none(video_meta.get('media_duration'))
         uploader = video_meta.get('media_user_name')
-        views = int(video_meta.get('media_views'))
-        upload_date = unified_strdate(video_meta.get('media_date_added'))
+        views = int_or_none(video_meta.get('media_views'))
+        timestamp = parse_iso8601(video_meta.get(date), ' ')
         categories = [video_meta.get('category_name')]
         thumbs = [
             {'url': thumb_base + video_meta.get('media_thumbnail'),
@@ -77,28 +70,97 @@ class HitboxIE(InfoExtractor):
              'height': 432},
         ]
 
-        player_config = self._download_json(
-            'https://www.hitbox.tv/api/player/config/video/%s' % (video_id),
-            video_id
-        )
-
-        clip = player_config.get('clip')
-        video_url = clip.get('url')
-        res = clip.get('bitrates', [])[0].get('label')
-
         return {
             'id': video_id,
             'title': title,
             'alt_title': alt_title,
             'description': description,
-            'url': video_url,
             'ext': 'mp4',
             'thumbnails': thumbs,
             'duration': duration,
-            'resolution': res,
-            'uploader_id': uploader,
+            'uploader': uploader,
             'view_count': views,
-            'upload_date': upload_date,
+            'timestamp': timestamp,
             'categories': categories,
-            'protocol': 'm3u8',
         }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        metadata = self._extract_metadata(
+            'https://www.hitbox.tv/api/media/video',
+            video_id)
+
+        player_config = self._download_json(
+            'https://www.hitbox.tv/api/player/config/video/%s' % video_id,
+            video_id)
+
+        clip = player_config.get('clip')
+        video_url = clip.get('url')
+        res = clip.get('bitrates', [])[0].get('label')
+
+        metadata['resolution'] = res
+        metadata['url'] = video_url
+        metadata['protocol'] = 'm3u8'
+
+        return metadata
+
+
+class HitboxLiveIE(HitboxIE):
+    IE_NAME = 'hitbox:live'
+    _VALID_URL = r'https?://(?:www\.)?hitbox\.tv/(?!video)(?P<id>.+)'
+    _TEST = {
+        'url': 'http://www.hitbox.tv/dimak',
+        'info_dict': {
+            'id': 'dimak',
+            'ext': 'mp4',
+            'description': 'md5:c9f80fa4410bc588d7faa40003fc7d0e',
+            'timestamp': int,
+            'upload_date': compat_str,
+            'title': compat_str,
+            'uploader': 'Dimak',
+        },
+        'params': {
+            # live
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        metadata = self._extract_metadata(
+            'https://www.hitbox.tv/api/media/live',
+            video_id)
+
+        player_config = self._download_json(
+            'https://www.hitbox.tv/api/player/config/live/%s' % video_id,
+            video_id)
+
+        formats = []
+        cdns = player_config.get('cdns')
+        servers = []
+        for cdn in cdns:
+            base_url = cdn.get('netConnectionUrl')
+            host = re.search('.+\.([^\.]+\.[^\./]+)/.+', base_url).group(1)
+            if base_url not in servers:
+                servers.append(base_url)
+                for stream in cdn.get('bitrates'):
+                    label = stream.get('label')
+                    if label != 'Auto':
+                        formats.append({
+                            'url': '%s/%s' % (base_url, stream.get('url')),
+                            'ext': 'mp4',
+                            'vbr': stream.get('bitrate'),
+                            'resolution': label,
+                            'rtmp_live': True,
+                            'format_note': host,
+                            'page_url': url,
+                            'player_url': 'http://www.hitbox.tv/static/player/flowplayer/flowplayer.commercial-3.2.16.swf',
+                        })
+
+        self._sort_formats(formats)
+        metadata['formats'] = formats
+        metadata['is_live'] = True
+        metadata['title'] = self._live_title(metadata.get('title'))
+        return metadata