Stop libav compatibility links from being considered as genuine.

[youtubedl] / youtube_dl / extractor / leeco.py

diff --git a/youtube_dl/extractor/leeco.py b/youtube_dl/extractor/leeco.py
index 63f581cd9fe44013ce8ec85ce540d04d0df73713..e9cc9aa5983967861b08a2d9ee79297ae3a1726e 100644 (file)
--- a/youtube_dl/extractor/leeco.py
+++ b/youtube_dl/extractor/leeco.py
@@ -20,9 +20,10 @@ from ..utils import (
     int_or_none,
     orderedSet,
     parse_iso8601,
     int_or_none,
     orderedSet,
     parse_iso8601,

-    sanitized_Request,

     str_or_none,
     url_basename,
     str_or_none,
     url_basename,

+    urshift,
+    update_url_query,

 )
 
 
 )
 
 


@@ -74,15 +75,11 @@ class LeIE(InfoExtractor):
         'only_matching': True,
     }]
 
         'only_matching': True,
     }]
 

-    @staticmethod
-    def urshift(val, n):
-        return val >> n if val >= 0 else (val + 0x100000000) >> n
-

     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
     def ror(self, param1, param2):
         _loc3_ = 0
         while _loc3_ < param2:
     # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
     def ror(self, param1, param2):
         _loc3_ = 0
         while _loc3_ < param2:

-            param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
+            param1 = urshift(param1, 1) + ((param1 & 1) << 31)

             _loc3_ += 1
         return param1
 
             _loc3_ += 1
         return param1
 


@@ -93,6 +90,10 @@ class LeIE(InfoExtractor):
         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
         return _loc3_
 
         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
         return _loc3_
 

+    # reversed from http://jstatic.letvcdn.com/sdk/player.js
+    def get_mms_key(self, time):
+        return self.ror(time, 8) ^ 185025305
+

     # see M3U8Encryption class in KLetvPlayer.swf
     @staticmethod
     def decrypt_m3u8(encrypted_data):
     # see M3U8Encryption class in KLetvPlayer.swf
     @staticmethod
     def decrypt_m3u8(encrypted_data):


@@ -113,28 +114,7 @@ class LeIE(InfoExtractor):
 
         return bytes(_loc7_)
 
 
         return bytes(_loc7_)
 

-    def _real_extract(self, url):
-        media_id = self._match_id(url)
-        page = self._download_webpage(url, media_id)
-        params = {
-            'id': media_id,
-            'platid': 1,
-            'splatid': 101,
-            'format': 1,
-            'tkey': self.calc_time_key(int(time.time())),
-            'domain': 'www.le.com'
-        }
-        play_json_req = sanitized_Request(
-            'http://api.le.com/mms/out/video/playJson?' + compat_urllib_parse_urlencode(params)
-        )
-        cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-        if cn_verification_proxy:
-            play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
-
-        play_json = self._download_json(
-            play_json_req,
-            media_id, 'Downloading playJson data')
-
+    def _check_errors(self, play_json):

         # Check for errors
         playstatus = play_json['playstatus']
         if playstatus['status'] == 0:
         # Check for errors
         playstatus = play_json['playstatus']
         if playstatus['status'] == 0:


@@ -145,43 +125,99 @@ class LeIE(InfoExtractor):
                 msg = 'Generic error. flag = %d' % flag
             raise ExtractorError(msg, expected=True)
 
                 msg = 'Generic error. flag = %d' % flag
             raise ExtractorError(msg, expected=True)
 

-        playurl = play_json['playurl']
-
-        formats = ['350', '1000', '1300', '720p', '1080p']
-        dispatch = playurl['dispatch']
+    def _real_extract(self, url):
+        media_id = self._match_id(url)
+        page = self._download_webpage(url, media_id)

 
 

-        urls = []
-        for format_id in formats:
-            if format_id in dispatch:
-                media_url = playurl['domain'][0] + dispatch[format_id][0]
-                media_url += '&' + compat_urllib_parse_urlencode({
-                    'm3v': 1,
+        play_json_h5 = self._download_json(
+            'http://api.le.com/mms/out/video/playJsonH5',
+            media_id, 'Downloading html5 playJson data', query={
+                'id': media_id,
+                'platid': 3,
+                'splatid': 304,
+                'format': 1,
+                'tkey': self.get_mms_key(int(time.time())),
+                'domain': 'www.le.com',
+                'tss': 'no',
+            },
+            headers=self.geo_verification_headers())
+        self._check_errors(play_json_h5)
+
+        play_json_flash = self._download_json(
+            'http://api.le.com/mms/out/video/playJson',
+            media_id, 'Downloading flash playJson data', query={
+                'id': media_id,
+                'platid': 1,
+                'splatid': 101,
+                'format': 1,
+                'tkey': self.calc_time_key(int(time.time())),
+                'domain': 'www.le.com',
+            },
+            headers=self.geo_verification_headers())
+        self._check_errors(play_json_flash)
+
+        def get_h5_urls(media_url, format_id):
+            location = self._download_json(
+                media_url, media_id,
+                'Download JSON metadata for format %s' % format_id, query={

                     'format': 1,
                     'expect': 3,
                     'format': 1,
                     'expect': 3,

-                    'rateid': format_id,
-                })
+                    'tss': 'no',
+                })['location']
+
+            return {
+                'http': update_url_query(location, {'tss': 'no'}),
+                'hls': update_url_query(location, {'tss': 'ios'}),
+            }

 
 

-                nodes_data = self._download_json(
-                    media_url, media_id,
-                    'Download JSON metadata for format %s' % format_id)
+        def get_flash_urls(media_url, format_id):
+            media_url += '&' + compat_urllib_parse_urlencode({
+                'm3v': 1,
+                'format': 1,
+                'expect': 3,
+                'rateid': format_id,
+            })

 
 

-                req = self._request_webpage(
-                    nodes_data['nodelist'][0]['location'], media_id,
-                    note='Downloading m3u8 information for format %s' % format_id)
+            nodes_data = self._download_json(
+                media_url, media_id,
+                'Download JSON metadata for format %s' % format_id)

 
 

-                m3u8_data = self.decrypt_m3u8(req.read())
+            req = self._request_webpage(
+                nodes_data['nodelist'][0]['location'], media_id,
+                note='Downloading m3u8 information for format %s' % format_id)

 
 

-                url_info_dict = {
-                    'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
-                    'ext': determine_ext(dispatch[format_id][1]),
-                    'format_id': format_id,
-                    'protocol': 'm3u8',
-                }
+            m3u8_data = self.decrypt_m3u8(req.read())

 
 

-                if format_id[-1:] == 'p':
-                    url_info_dict['height'] = int_or_none(format_id[:-1])
+            return {
+                'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
+            }

 
 

-                urls.append(url_info_dict)
+        extracted_formats = []
+        formats = []
+        for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)):
+            playurl = play_json['playurl']
+            play_domain = playurl['domain'][0]
+
+            for format_id, format_data in playurl.get('dispatch', []).items():
+                if format_id in extracted_formats:
+                    continue
+                extracted_formats.append(format_id)
+
+                media_url = play_domain + format_data[0]
+                for protocol, format_url in get_urls(media_url, format_id).items():
+                    f = {
+                        'url': format_url,
+                        'ext': determine_ext(format_data[1]),
+                        'format_id': '%s-%s' % (protocol, format_id),
+                        'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
+                        'quality': int_or_none(format_id),
+                    }
+
+                    if format_id[-1:] == 'p':
+                        f['height'] = int_or_none(format_id[:-1])
+
+                    formats.append(f)
+        self._sort_formats(formats, ('height', 'quality', 'format_id'))

 
         publish_time = parse_iso8601(self._html_search_regex(
             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
 
         publish_time = parse_iso8601(self._html_search_regex(
             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),


@@ -190,7 +226,7 @@ class LeIE(InfoExtractor):
 
         return {
             'id': media_id,
 
         return {
             'id': media_id,

-            'formats': urls,
+            'formats': formats,

             'title': playurl['title'],
             'thumbnail': playurl['pic'],
             'description': description,
             'title': playurl['title'],
             'thumbnail': playurl['pic'],
             'description': description,