Update changelog with bugs being fixed.

[youtubedl] / youtube_dl / extractor / streamcz.py

diff --git a/youtube_dl/extractor/streamcz.py b/youtube_dl/extractor/streamcz.py
index 7362904db50588acb6f1f988b7fc6665f9b8ce2e..58e0b4c802781728cf9d4ba17a19236bc3b54565 100644 (file)
--- a/youtube_dl/extractor/streamcz.py
+++ b/youtube_dl/extractor/streamcz.py
@@ -1,67 +1,105 @@
-# -*- coding: utf-8 -*-
+# coding: utf-8
 from __future__ import unicode_literals
 
-import re
-import json
+import hashlib
+import time
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    sanitized_Request,
+)
+
+
+def _get_api_key(api_path):
+    if api_path.endswith('?'):
+        api_path = api_path[:-1]
+
+    api_key = 'fb5f58a820353bd7095de526253c14fd'
+    a = '{0:}{1:}{2:}'.format(api_key, api_path, int(round(time.time() / 24 / 3600)))
+    return hashlib.md5(a.encode('ascii')).hexdigest()
 
 
 class StreamCZIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?stream\.cz/.+/(?P<videoid>.+)'
+    _VALID_URL = r'https?://(?:www\.)?stream\.cz/.+/(?P<id>[0-9]+)'
+    _API_URL = 'http://www.stream.cz/API'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.stream.cz/peklonataliri/765767-ecka-pro-deti',
-        'md5': '6d3ca61a8d0633c9c542b92fcb936b0c',
+        'md5': '934bb6a6d220d99c010783c9719960d5',
         'info_dict': {
             'id': '765767',
             'ext': 'mp4',
             'title': 'Peklo na talíři: Éčka pro děti',
-            'description': 'md5:49ace0df986e95e331d0fe239d421519',
-            'thumbnail': 'http://im.stream.cz/episode/52961d7e19d423f8f06f0100',
+            'description': 'Taška s grónskou pomazánkou a další pekelnosti ZDE',
+            'thumbnail': 're:^http://im.stream.cz/episode/52961d7e19d423f8f06f0100',
             'duration': 256,
         },
-    }
+    }, {
+        'url': 'http://www.stream.cz/blanik/10002447-tri-roky-pro-mazanka',
+        'md5': '849a88c1e1ca47d41403c2ba5e59e261',
+        'info_dict': {
+            'id': '10002447',
+            'ext': 'mp4',
+            'title': 'Kancelář Blaník: Tři roky pro Mazánka',
+            'description': 'md5:3862a00ba7bf0b3e44806b544032c859',
+            'thumbnail': 're:^http://im.stream.cz/episode/537f838c50c11f8d21320000',
+            'duration': 368,
+        },
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('videoid')
-
-        webpage = self._download_webpage(url, video_id)
+        video_id = self._match_id(url)
+        api_path = '/episode/%s' % video_id
 
-        data = self._html_search_regex(r'Stream\.Data\.Episode\((.+?)\);', webpage, 'stream data')
-
-        jsonData = json.loads(data)
+        req = sanitized_Request(self._API_URL + api_path)
+        req.add_header('Api-Password', _get_api_key(api_path))
+        data = self._download_json(req, video_id)
 
         formats = []
-        for video in jsonData['instances']:
-            for video_format in video['instances']:
-                format_id = video_format['quality']
-
-                if format_id == '240p':
-                    quality = 0
-                elif format_id == '360p':
-                    quality = 1
-                elif format_id == '480p':
-                    quality = 2
-                elif format_id == '720p':
-                    quality = 3
-
+        for quality, video in enumerate(data['video_qualities']):
+            for f in video['formats']:
+                typ = f['type'].partition('/')[2]
+                qlabel = video.get('quality_label')
                 formats.append({
-                    'format_id': '%s-%s' % (video_format['type'].split('/')[1], format_id),
-                    'url': video_format['source'],
+                    'format_note': '%s-%s' % (qlabel, typ) if qlabel else typ,
+                    'format_id': '%s-%s' % (typ, f['quality']),
+                    'url': f['source'],
+                    'height': int_or_none(f['quality'].rstrip('p')),
                     'quality': quality,
                 })
-
         self._sort_formats(formats)
 
+        image = data.get('image')
+        if image:
+            thumbnail = self._proto_relative_url(
+                image.replace('{width}', '1240').replace('{height}', '697'),
+                scheme='http:',
+            )
+        else:
+            thumbnail = None
+
+        stream = data.get('_embedded', {}).get('stream:show', {}).get('name')
+        if stream:
+            title = '%s: %s' % (stream, data['name'])
+        else:
+            title = data['name']
+
+        subtitles = {}
+        srt_url = data.get('subtitles_srt')
+        if srt_url:
+            subtitles['cs'] = [{
+                'ext': 'srt',
+                'url': srt_url,
+            }]
+
         return {
-            'id': str(jsonData['id']),
-            'title': self._og_search_title(webpage),
-            'thumbnail': jsonData['episode_image_original_url'].replace('//', 'http://'),
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
             'formats': formats,
-            'description': self._og_search_description(webpage),
-            'duration': int_or_none(jsonData['duration']),
-            'view_count': int_or_none(jsonData['stats_total']),
+            'description': data.get('web_site_text'),
+            'duration': int_or_none(data.get('duration')),
+            'view_count': int_or_none(data.get('views')),
+            'subtitles': subtitles,
         }