New upstream release 2017.09.24.

[youtubedl] / youtube_dl / extractor / eagleplatform.py

diff --git a/youtube_dl/extractor/eagleplatform.py b/youtube_dl/extractor/eagleplatform.py
index 12d28d3b9f1e76f84f0f9fa322befd0bfa056f09..42789278e3a04753bcbce06ad62316c458708dc1 100644 (file)
--- a/youtube_dl/extractor/eagleplatform.py
+++ b/youtube_dl/extractor/eagleplatform.py
@@ -4,11 +4,14 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
 import re
 
 from .common import InfoExtractor

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

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

-    url_basename,
+    unsmuggle_url,

 )
 
 
 )
 
 


@@ -29,7 +32,7 @@ class EaglePlatformIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Навальный вышел на свободу',
             'description': 'md5:d97861ac9ae77377f3f20eaf9d04b4f5',
             '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,


@@ -43,20 +46,61 @@ class EaglePlatformIE(InfoExtractor):
             'id': '12820',
             'ext': 'mp4',
             'title': "'O Sole Mio",
             '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',

+    }, {
+        # referrer protected video (https://tvrain.ru/lite/teleshow/kak_vse_nachinalos/namin-418921/)
+        'url': 'eagleplatform:tvrainru.media.eagleplatform.com:582306',
+        'only_matching': True,

     }]
 
     @staticmethod
     def _extract_url(webpage):
     }]
 
     @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')
         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')

+        PLAYER_JS_RE = r'''
+                        <script[^>]+
+                            src=(?P<qjs>["\'])(?:https?:)?//(?P<host>(?:(?!(?P=qjs)).)+\.media\.eagleplatform\.com)/player/player\.js(?P=qjs)
+                        .+?
+                    '''
+        # "Basic usage" embedding (see http://dultonmedia.github.io/eplayer/)
+        mobj = re.search(
+            r'''(?xs)
+                    %s
+                    <div[^>]+
+                        class=(?P<qclass>["\'])eagleplayer(?P=qclass)[^>]+
+                        data-id=["\'](?P<id>\d+)
+            ''' % PLAYER_JS_RE, webpage)
+        if mobj is not None:
+            return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()
+        # Generalization of "Javascript code usage", "Combined usage" and
+        # "Usage without attaching to DOM" embeddings (see
+        # http://dultonmedia.github.io/eplayer/)
+        mobj = re.search(
+            r'''(?xs)
+                    %s
+                    <script>
+                    .+?
+                    new\s+EaglePlayer\(
+                        (?:[^,]+\s*,\s*)?
+                        {
+                            .+?
+                            \bid\s*:\s*["\']?(?P<id>\d+)
+                            .+?
+                        }
+                    \s*\)
+                    .+?
+                    </script>
+            ''' % PLAYER_JS_RE, webpage)
+        if mobj is not None:
+            return 'eagleplatform:%(host)s:%(id)s' % mobj.groupdict()

 
     @staticmethod
     def _handle_error(response):
 
     @staticmethod
     def _handle_error(response):


@@ -64,9 +108,10 @@ class EaglePlatformIE(InfoExtractor):
         if status != 200:
             raise ExtractorError(' '.join(response['errors']), expected=True)
 
         if status != 200:
             raise ExtractorError(' '.join(response['errors']), expected=True)
 

-    def _download_json(self, url_or_request, video_id, note='Downloading JSON metadata'):
+    def _download_json(self, url_or_request, video_id, *args, **kwargs):

         try:
         try:

-            response = super(EaglePlatformIE, self)._download_json(url_or_request, video_id, note)
+            response = super(EaglePlatformIE, self)._download_json(
+                url_or_request, video_id, *args, **kwargs)

         except ExtractorError as ee:
             if isinstance(ee.cause, compat_HTTPError):
                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)
         except ExtractorError as ee:
             if isinstance(ee.cause, compat_HTTPError):
                 response = self._parse_json(ee.cause.read().decode('utf-8'), video_id)


@@ -78,11 +123,24 @@ class EaglePlatformIE(InfoExtractor):
         return self._download_json(url_or_request, video_id, note)['data'][0]
 
     def _real_extract(self, url):
         return self._download_json(url_or_request, video_id, note)['data'][0]
 
     def _real_extract(self, url):

+        url, smuggled_data = unsmuggle_url(url, {})
+

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

+        headers = {}
+        query = {
+            'id': video_id,
+        }
+
+        referrer = smuggled_data.get('referrer')
+        if referrer:
+            headers['Referer'] = referrer
+            query['referrer'] = referrer
+

         player_data = self._download_json(
         player_data = self._download_json(

-            'http://%s/api/player_data?id=%s' % (host, video_id), video_id)
+            'http://%s/api/player_data' % host, video_id,
+            headers=headers, query=query)

 
         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 
 
         media = player_data['data']['playlist']['viewports'][0]['medialist'][0]
 


@@ -103,29 +161,38 @@ class EaglePlatformIE(InfoExtractor):
 
         m3u8_url = self._get_video_url(secure_m3u8, video_id, 'Downloading m3u8 JSON')
         m3u8_formats = self._extract_m3u8_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')
+            m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native',
+            m3u8_id='hls', fatal=False)

         formats.extend(m3u8_formats)
 
         formats.extend(m3u8_formats)
 

-        mp4_url = self._get_video_url(
+        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
             # 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', 'mp4', secure_m3u8),
-            video_id, 'Downloading mp4 JSON')
-        mp4_url_basename = url_basename(mp4_url)
-        for m3u8_format in m3u8_formats:
-            mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url'])
-            if mobj:
-                http_format = m3u8_format.copy()
-                video_url = mp4_url.replace(mp4_url_basename, mobj.group(1))
-                if not self._is_valid_url(video_url, video_id):
+            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
                     continue

-                http_format.update({
-                    'url': video_url,
-                    'format_id': m3u8_format['format_id'].replace('hls', 'http'),
-                    'protocol': 'http',
-                })
-                formats.append(http_format)
+                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)

 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)