Merge tag 'upstream/2017.05.18.1'

[youtubedl] / youtube_dl / extractor / eagleplatform.py

diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py
index 688dfc2f7f34d15e712481351a6073987325add5..76d39adac5faa9912f42d271def60a46128be3f7 100644 (file)
--- a/youtube_dl/extractor/eagleplatform.py
+++ b/youtube_dl/extractor/eagleplatform.py
@@ -4,6 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+)

 from ..utils import (
     ExtractorError,
     int_or_none,
 from ..utils import (
     ExtractorError,
     int_or_none,


@@ -21,13 +25,13 @@ class EaglePlatformIE(InfoExtractor):
     _TESTS = [{
         # http://lenta.ru/news/2015/03/06/navalny/
         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
     _TESTS = [{
         # http://lenta.ru/news/2015/03/06/navalny/
         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',

-        'md5': '0b7994faa2bd5c0f69a3db6db28d078d',
+        # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used

         'info_dict': {
             'id': '227304',
             'ext': 'mp4',
             'title': 'Навальный вышел на свободу',
             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
         'info_dict': {
             'id': '227304',
             'ext': 'mp4',
             'title': 'Навальный вышел на свободу',
             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',

-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',

             'duration': 87,
             'view_count': int,
             'age_limit': 0,
             'duration': 87,
             'view_count': int,
             'age_limit': 0,


@@ -36,28 +40,58 @@ class EaglePlatformIE(InfoExtractor):
         # http://muz-tv.ru/play/7129/
         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
         'url': 'eagleplatform:media.clipyou.ru:12820',
         # http://muz-tv.ru/play/7129/
         # http://media.clipyou.ru/index/player?record_id=12820&width=730&height=415&autoplay=true
         'url': 'eagleplatform:media.clipyou.ru:12820',

-        'md5': '6c2ebeab03b739597ce8d86339d5a905',
+        'md5': '358597369cf8ba56675c1df15e7af624',

         'info_dict': {
             'id': '12820',
             'ext': 'mp4',
             'title': "'O Sole Mio",
         'info_dict': {
             'id': '12820',
             'ext': 'mp4',
             'title': "'O Sole Mio",

-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',

             'duration': 216,
             'view_count': int,
         },
         'skip': 'Georestricted',
     }]
 
             'duration': 216,
             'view_count': int,
         },
         'skip': 'Georestricted',
     }]
 

-    def _handle_error(self, response):
+    @staticmethod
+    def _extract_url(webpage):
+        # Regular iframe embedding
+        mobj = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
+            webpage)
+        if mobj is not None:
+            return mobj.group('url')
+        # Basic usage embedding (see http://dultonmedia.github.io/eplayer/)
+        mobj = re.search(
+            r'''(?xs)
+                    <script[^>]+
+                        src=(?P<q1>["\'])(?:https?:)?//(?P<host>.+?\.media\.eagleplatform\.com)/player/player\.js(?P=q1)
+                    .+?
+                    <div[^>]+
+                        class=(?P<q2>["\'])eagleplayer(?P=q2)[^>]+
+                        data-id=["\'](?P<id>\d+)
+            ''', webpage)
+        if mobj is not None:
+            return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
+
+    @staticmethod
+    def _handle_error(response):

         status = int_or_none(response.get('status', 200))
         if status != 200:
             raise ExtractorError(' '.join(response['errors']), expected=True)
 
         status = int_or_none(response.get('status', 200))
         if status != 200:
             raise ExtractorError(' '.join(response['errors']), expected=True)
 

-    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
-        response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
-        self._handle_error(response)
+    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata', *args, **kwargs):
+        try:
+            response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
+        except ExtractorError as ee:
+            if isinstance(ee.cause, compat_HTTPError):
+                response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
+                self._handle_error(response)
+            raise

         return response
 
         return response
 

+    def _get_video_url(self, url_or_request, video_id, note='Downloading JSON metadata'):
+        return self._download_json(url_or_request, video_id, note)['data'][0]
+

     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         host, video_id = mobj.group('custom_host') or mobj.group('host'), mobj.group('id')


@@ -69,7 +103,7 @@ class EaglePlatformIE(InfoExtractor):
 
         title = media['title']
         description = media.get('description')
 
         title = media['title']
         description = media.get('description')

-        thumbnail = media.get('snapshot')
+        thumbnail = self._proto_relative_url(media.get('snapshot'), 'http:')

         duration = int_or_none(media.get('duration'))
         view_count = int_or_none(media.get('views'))
 
         duration = int_or_none(media.get('duration'))
         view_count = int_or_none(media.get('views'))
 


@@ -78,13 +112,45 @@ class EaglePlatformIE(InfoExtractor):
         if age_restriction:
             age_limit = 0 if age_restriction == 'allow_all' else 18
 
         if age_restriction:
             age_limit = 0 if age_restriction == 'allow_all' else 18
 

-        m3u8_data = self._download_json(
-            media['sources']['secure_m3u8']['auto'],
-            video_id, 'Downloading m3u8 JSON')
+        secure_m3u8 = self._proto_relative_url(media['sources']['secure_m3u8']['auto'], 'http:')
+
+        formats = []
+
+        m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
+        m3u8_formats = self._extract_m3u8_formats(
+            m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
+            m3u8_id='hls', fatal=False)
+        formats.extend(m3u8_formats)
+
+        m3u8_formats_dict = {}
+        for f in m3u8_formats:
+            if f.get('height') is not None:
+                m3u8_formats_dict[f['height']] = f
+
+        mp4_data = self._download_json(
+            # Secure mp4 URL is constructed according to Player.prototype.mp4 from
+            # http://lentaru.media.eagleplatform.com/player/player.js
+            re.sub(r'm3u8|hlsvod|hls|f4m', 'mp4s', secure_m3u8),
+            video_id, 'Downloading mp4 JSON', fatal=False)
+        if mp4_data:
+            for format_id, format_url in mp4_data.get('data', {}).items():
+                if not isinstance(format_url, compat_str):
+                    continue
+                height = int_or_none(format_id)
+                if height is not None and m3u8_formats_dict.get(height):
+                    f = m3u8_formats_dict[height].copy()
+                    f.update({
+                        'format_id': f['format_id'].replace('hls', 'http'),
+                        'protocol': 'http',
+                    })
+                else:
+                    f = {
+                        'format_id': 'http-%s' % format_id,
+                        'height': int_or_none(format_id),
+                    }
+                f['url'] = format_url
+                formats.append(f)

 
 

-        formats = self._extract_m3u8_formats(
-            m3u8_data['data'][0], video_id,
-            'mp4', entry_protocol='m3u8_native')

         self._sort_formats(formats)
 
         return {
         self._sort_formats(formats)
 
         return {