Merge pull request #7208 from yan12125/letv-fix

[Letv] Fix extraction

youtube_dl/YoutubeDL.py

@@ -37,6 +37,7 @@
     compat_tokenize_tokenize,
     compat_urllib_error,
     compat_urllib_request,
+    compat_urllib_request_DataHandler,
 )
 from .utils import (
     ContentTooShortError,
@@ -1967,8 +1968,9 @@ def _setup_opener(self):
         debuglevel = 1 if self.params.get('debug_printtraffic') else 0
         https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
         ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
+        data_handler = compat_urllib_request_DataHandler()
         opener = compat_urllib_request.build_opener(
-            proxy_handler, https_handler, cookie_processor, ydlh)
+            proxy_handler, https_handler, cookie_processor, ydlh, data_handler)
 
         # Delete the default user-agent header, which would otherwise apply in
         # cases where our custom HTTP handler doesn't come into play

youtube_dl/compat.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
+import binascii
 import collections
+import email
 import getpass
+import io
 import optparse
 import os
 import re
@@ -38,6 +41,11 @@
 except ImportError:  # Python 2
     import urlparse as compat_urlparse
 
+try:
+    import urllib.response as compat_urllib_response
+except ImportError:  # Python 2
+    import urllib as compat_urllib_response
+
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError:  # Python 2
@@ -155,6 +163,40 @@ def compat_urllib_parse_unquote_plus(string, encoding='utf-8', errors='replace')
         string = string.replace('+', ' ')
         return compat_urllib_parse_unquote(string, encoding, errors)
 
+try:
+    from urllib.request import DataHandler as compat_urllib_request_DataHandler
+except ImportError:  # Python < 3.4
+    # Ported from CPython 98774:1733b3bd46db, Lib/urllib/request.py
+    class compat_urllib_request_DataHandler(compat_urllib_request.BaseHandler):
+        def data_open(self, req):
+            # data URLs as specified in RFC 2397.
+            #
+            # ignores POSTed data
+            #
+            # syntax:
+            # dataurl   := "data:" [ mediatype ] [ ";base64" ] "," data
+            # mediatype := [ type "/" subtype ] *( ";" parameter )
+            # data      := *urlchar
+            # parameter := attribute "=" value
+            url = req.get_full_url()
+
+            scheme, data = url.split(":", 1)
+            mediatype, data = data.split(",", 1)
+
+            # even base64 encoded data URLs might be quoted so unquote in any case:
+            data = compat_urllib_parse_unquote_to_bytes(data)
+            if mediatype.endswith(";base64"):
+                data = binascii.a2b_base64(data)
+                mediatype = mediatype[:-7]
+
+            if not mediatype:
+                mediatype = "text/plain;charset=US-ASCII"
+
+            headers = email.message_from_string(
+                "Content-type: %s\nContent-length: %d\n" % (mediatype, len(data)))
+
+            return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
+
 try:
     compat_basestring = basestring  # Python 2
 except NameError:
@@ -489,6 +531,8 @@ def compat_itertools_count(start=0, step=1):
     'compat_urllib_parse_unquote_to_bytes',
     'compat_urllib_parse_urlparse',
     'compat_urllib_request',
+    'compat_urllib_request_DataHandler',
+    'compat_urllib_response',
     'compat_urlparse',
     'compat_urlretrieve',
     'compat_xml_parse_error',

youtube_dl/downloader/hls.py

@@ -30,7 +30,7 @@ def real_download(self, filename, info_dict):
 
         args = [ffpp.executable, '-y']
 
-        if info_dict['http_headers']:
+        if info_dict['http_headers'] and re.match(r'^https?://', url):
             # Trailing \r\n after each HTTP header is important to prevent warning from ffmpeg/avconv:
             # [http @ 00000000003d2fa0] No trailing CRLF found in HTTP header.
             args += [

youtube_dl/extractor/letv.py

@@ -9,13 +9,14 @@
 from ..compat import (
     compat_urllib_parse,
     compat_urllib_request,
-    compat_urlparse,
+    compat_ord,
 )
 from ..utils import (
     determine_ext,
     ExtractorError,
     parse_iso8601,
     int_or_none,
+    encode_data_uri,
 )
 
 
@@ -25,15 +26,16 @@ class LetvIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.letv.com/ptv/vplay/22005890.html',
-        'md5': 'cab23bd68d5a8db9be31c9a222c1e8df',
+        'md5': 'edadcfe5406976f42f9f266057ee5e40',
         'info_dict': {
             'id': '22005890',
             'ext': 'mp4',
             'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
-            'timestamp': 1424747397,
-            'upload_date': '20150224',
             'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
-        }
+        },
+        'params': {
+            'hls_prefer_native': True,
+        },
     }, {
         'url': 'http://www.letv.com/ptv/vplay/1415246.html',
         'info_dict': {
@@ -42,16 +44,22 @@ class LetvIE(InfoExtractor):
             'title': '美人天下01',
             'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
         },
+        'params': {
+            'hls_prefer_native': True,
+        },
     }, {
         'note': 'This video is available only in Mainland China, thus a proxy is needed',
         'url': 'http://www.letv.com/ptv/vplay/1118082.html',
-        'md5': 'f80936fbe20fb2f58648e81386ff7927',
+        'md5': '2424c74948a62e5f31988438979c5ad1',
         'info_dict': {
             'id': '1118082',
             'ext': 'mp4',
             'title': '与龙共舞 完整版',
             'description': 'md5:7506a5eeb1722bb9d4068f85024e3986',
         },
+        'params': {
+            'hls_prefer_native': True,
+        },
         'skip': 'Only available in China',
     }]
 
@@ -74,6 +82,27 @@ def calc_time_key(self, param1):
         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
         return _loc3_
 
+    # see M3U8Encryption class in KLetvPlayer.swf
+    @staticmethod
+    def decrypt_m3u8(encrypted_data):
+        if encrypted_data[:5].decode('utf-8').lower() != 'vc_01':
+            return encrypted_data
+        encrypted_data = encrypted_data[5:]
+
+        _loc4_ = bytearray()
+        while encrypted_data:
+            b = compat_ord(encrypted_data[0])
+            _loc4_.extend([b // 16, b & 0x0f])
+            encrypted_data = encrypted_data[1:]
+        idx = len(_loc4_) - 11
+        _loc4_ = _loc4_[idx:] + _loc4_[:idx]
+        _loc7_ = bytearray()
+        while _loc4_:
+            _loc7_.append(_loc4_[0] * 16 + _loc4_[1])
+            _loc4_ = _loc4_[2:]
+
+        return bytes(_loc7_)
+
     def _real_extract(self, url):
         media_id = self._match_id(url)
         page = self._download_webpage(url, media_id)
@@ -115,23 +144,28 @@ def _real_extract(self, url):
         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({
-                # Mimic what flvxz.com do
+                    'm3v': 1,
-                url_parts = list(compat_urlparse.urlparse(media_url))
+                    'format': 1,
-                qs = dict(compat_urlparse.parse_qs(url_parts[4]))
+                    'expect': 3,
-                qs.update({
+                    'rateid': format_id,
-                    'platid': '14',
-                    'splatid': '1401',
-                    'tss': 'no',
-                    'retry': 1
                 })
-                url_parts[4] = compat_urllib_parse.urlencode(qs)
+
-                media_url = compat_urlparse.urlunparse(url_parts)
+                nodes_data = self._download_json(
+                    media_url, media_id,
+                    'Download JSON metadata for format %s' % format_id)
+
+                req = self._request_webpage(
+                    nodes_data['nodelist'][0]['location'], media_id,
+                    note='Downloading m3u8 information for format %s' % format_id)
+
+                m3u8_data = self.decrypt_m3u8(req.read())
 
                 url_info_dict = {
-                    'url': media_url,
+                    'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
                     'ext': determine_ext(dispatch[format_id][1]),
                     'format_id': format_id,
+                    'protocol': 'm3u8',
                 }
 
                 if format_id[-1:] == 'p':

youtube_dl/utils.py

@@ -3,6 +3,7 @@
 
 from __future__ import unicode_literals
 
+import base64
 import calendar
 import codecs
 import contextlib
@@ -1795,6 +1796,10 @@ def urlhandle_detect_ext(url_handle):
     return mimetype2ext(getheader('Content-Type'))
 
 
+def encode_data_uri(data, mime_type):
+    return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii'))
+
+
 def age_restricted(content_limit, age_limit):
     """ Returns True iff the content should be blocked """