youtube-dl

commit e498758b9cfa983d05b10a185ecdef480a93cf30
parent 5fc8d893612285ddba082f381006337573b36afa
Author: Sergey M․ <dstftw@gmail.com>
Date:   Sat, 25 Feb 2017 22:56:42 +0700

[freshlive] Fix issues and improve (closes #12175)

Diffstat:
M
youtube_dl/extractor/extractors.py
 | 
2
+-
M
youtube_dl/extractor/freshlive.py
 | 
65
++++++++++++++++++++++++++++++++++++++++++-----------------------

2 files changed, 43 insertions(+), 24 deletions(-)
diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py
@@ -339,7 +339,7 @@ from .francetv import (
 )
 from .freesound import FreesoundIE
 from .freespeech import FreespeechIE
-from .freshlive import FreshliveIE
+from .freshlive import FreshLiveIE
 from .funimation import FunimationIE
 from .funnyordie import FunnyOrDieIE
 from .fusion import FusionIE
diff --git a/youtube_dl/extractor/freshlive.py b/youtube_dl/extractor/freshlive.py
@@ -2,34 +2,40 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-
+from ..compat import compat_str
 from ..utils import (
+    ExtractorError,
     int_or_none,
-    parse_iso8601
+    try_get,
+    unified_timestamp,
 )
 
-class FreshliveIE(InfoExtractor):
-    _VALID_URL = r'https?://freshlive\.tv/(?P<streamer>[^/]+)/(?P<id>[0-9]+)'
+
+class FreshLiveIE(InfoExtractor):
+    _VALID_URL = r'https?://freshlive\.tv/[^/]+/(?P<id>\d+)'
     _TEST = {
         'url': 'https://freshlive.tv/satotv/74712',
-        'md5': '224f50d268b6b9f94e4198deccd55d6d',
+        'md5': '9f0cf5516979c4454ce982df3d97f352',
         'info_dict': {
-            'description': 'テスト',
-            'duration': 1511,
             'id': '74712',
             'ext': 'mp4',
-            'timestamp': 1483621764,
             'title': 'テスト',
+            'description': 'テスト',
             'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1511,
+            'timestamp': 1483619655,
             'upload_date': '20170105',
             'uploader': 'サトTV',
             'uploader_id': 'satotv',
             'view_count': int,
+            'comment_count': int,
+            'is_live': False,
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+
         webpage = self._download_webpage(url, video_id)
 
         options = self._parse_json(
@@ -38,27 +44,41 @@ class FreshliveIE(InfoExtractor):
                 webpage, 'initial context'),
             video_id)
 
-        programs = options['context']['dispatcher']['stores']['ProgramStore']['programs']
-        info = programs.get(video_id, {})
+        info = options['context']['dispatcher']['stores']['ProgramStore']['programs'][video_id]
 
-        video_url = info.get('liveStreamUrl') or info.get('archiveStreamUrl')
-        if not video_url:
-            raise ExtractorError('%s not a valid broadcast ID' % video_id, expected=True)
+        title = info['title']
+
+        if info.get('status') == 'upcoming':
+            raise ExtractorError('Stream %s is upcoming' % video_id, expected=True)
+
+        stream_url = info.get('liveStreamUrl') or info['archiveStreamUrl']
+
+        is_live = info.get('liveStreamUrl') is not None
 
         formats = self._extract_m3u8_formats(
-            video_url, video_id, ext='mp4', m3u8_id='hls')
+            stream_url, video_id, ext='mp4',
+            entry_protocol='m3u8' if is_live else 'm3u8_native',
+            m3u8_id='hls')
+
+        if is_live:
+            title = self._live_title(title)
 
         return {
             'id': video_id,
             'formats': formats,
-            'title': info.get('title'),
+            'title': title,
             'description': info.get('description'),
-            'duration': int_or_none(info.get('airTime')),
-            'is_live': int_or_none(info.get('airTime')) == None,
             'thumbnail': info.get('thumbnailUrl'),
-            'uploader': info.get('channel', {}).get('title'),
-            'uploader_id': info.get('channel', {}).get('code'),
-            'uploader_url': info.get('channel', {}).get('permalink'),
-            'timestamp': parse_iso8601(info.get('startAt')),
+            'duration': int_or_none(info.get('airTime')),
+            'timestamp': unified_timestamp(info.get('createdAt')),
+            'uploader': try_get(
+                info, lambda x: x['channel']['title'], compat_str),
+            'uploader_id': try_get(
+                info, lambda x: x['channel']['code'], compat_str),
+            'uploader_url': try_get(
+                info, lambda x: x['channel']['permalink'], compat_str),
             'view_count': int_or_none(info.get('viewCount')),
-        }-
\ No newline at end of file
+            'comment_count': int_or_none(info.get('commentCount')),
+            'tags': info.get('tags', []),
+            'is_live': is_live,
+        }