Merge tag 'upstream/2015.02.28'

[youtubedl] / youtube_dl / extractor / theplatform.py

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index a04925633b4bbdea7864bb5f1619e5c846f80b38..feac666f78baff49f4fb312a147acad67d320bc2 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -2,13 +2,21 @@ from __future__ import unicode_literals
 
 import re
 import json
 
 import re
 import json

+import time
+import hmac
+import binascii
+import hashlib
+

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..utils import (
+from ..compat import (

     compat_str,
     compat_str,

+)
+from ..utils import (

     determine_ext,
     ExtractorError,
     xpath_with_ns,
     determine_ext,
     ExtractorError,
     xpath_with_ns,

+    unsmuggle_url,

 )
 
 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})
 )
 
 _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language'})


@@ -16,7 +24,7 @@ _x = lambda p: xpath_with_ns(p, {'smil': 'http://www.w3.org/2005/SMIL21/Language
 
 class ThePlatformIE(InfoExtractor):
     _VALID_URL = r'''(?x)
 
 class ThePlatformIE(InfoExtractor):
     _VALID_URL = r'''(?x)

-        (?:https?://(?:link|player)\.theplatform\.com/[sp]/[^/]+/
+        (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/

            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
          |theplatform:)(?P<id>[^/\?&]+)'''
 
            (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?
          |theplatform:)(?P<id>[^/\?&]+)'''
 


@@ -35,19 +43,49 @@ class ThePlatformIE(InfoExtractor):
             'skip_download': True,
         },
     }
             'skip_download': True,
         },
     }

+
+    @staticmethod
+    def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
+        flags = '10' if include_qs else '00'
+        expiration_date = '%x' % (int(time.time()) + life)
+
+        def str_to_hex(str):
+            return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
+
+        def hex_to_str(hex):
+            return binascii.a2b_hex(hex)
+
+        relative_path = url.split('http://link.theplatform.com/s/')[1].split('?')[0]
+        clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path))
+        checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
+        sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
+        return '%s&sig=%s' % (url, sig)
+

     def _real_extract(self, url):
     def _real_extract(self, url):

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

         mobj = re.match(self._VALID_URL, url)
         mobj = re.match(self._VALID_URL, url)

+        provider_id = mobj.group('provider_id')

         video_id = mobj.group('id')
         video_id = mobj.group('id')

-        if mobj.group('config'):
-            config_url = url+ '&form=json'
+
+        if not provider_id:
+            provider_id = 'dJ5BDC'
+
+        if smuggled_data.get('force_smil_url', False):
+            smil_url = url
+        elif mobj.group('config'):
+            config_url = url + '&form=json'

             config_url = config_url.replace('swf/', 'config/')
             config_url = config_url.replace('onsite/', 'onsite/config/')
             config = self._download_json(config_url, video_id, 'Downloading config')
             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
         else:
             config_url = config_url.replace('swf/', 'config/')
             config_url = config_url.replace('onsite/', 'onsite/config/')
             config = self._download_json(config_url, video_id, 'Downloading config')
             smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4&manifest=f4m'
         else:

-            smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
-                'format=smil&mbr=true'.format(video_id))
+            smil_url = ('http://link.theplatform.com/s/{0}/{1}/meta.smil?'
+                        'format=smil&mbr=true'.format(provider_id, video_id))

 
 

+        sig = smuggled_data.get('sig')
+        if sig:
+            smil_url = self._sign_url(smil_url, sig['key'], sig['secret'])

 
         meta = self._download_xml(smil_url, video_id)
         try:
 
         meta = self._download_xml(smil_url, video_id)
         try:


@@ -60,10 +98,20 @@ class ThePlatformIE(InfoExtractor):
         else:
             raise ExtractorError(error_msg, expected=True)
 
         else:
             raise ExtractorError(error_msg, expected=True)
 

-        info_url = 'http://link.theplatform.com/s/dJ5BDC/{0}?format=preview'.format(video_id)
+        info_url = 'http://link.theplatform.com/s/{0}/{1}?format=preview'.format(provider_id, video_id)

         info_json = self._download_webpage(info_url, video_id)
         info = json.loads(info_json)
 
         info_json = self._download_webpage(info_url, video_id)
         info = json.loads(info_json)
 

+        subtitles = {}
+        captions = info.get('captions')
+        if isinstance(captions, list):
+            for caption in captions:
+                lang, src, mime = caption.get('lang', 'en'), caption.get('src'), caption.get('type')
+                subtitles[lang] = [{
+                    'ext': 'srt' if mime == 'text/srt' else 'ttml',
+                    'url': src,
+                }]
+

         head = meta.find(_x('smil:head'))
         body = meta.find(_x('smil:body'))
 
         head = meta.find(_x('smil:head'))
         body = meta.find(_x('smil:body'))
 


@@ -115,8 +163,9 @@ class ThePlatformIE(InfoExtractor):
         return {
             'id': video_id,
             'title': info['title'],
         return {
             'id': video_id,
             'title': info['title'],

+            'subtitles': subtitles,

             'formats': formats,
             'description': info['description'],
             'thumbnail': info['defaultThumbnailUrl'],
             'formats': formats,
             'description': info['description'],
             'thumbnail': info['defaultThumbnailUrl'],

-            'duration': info['duration']//1000,
+            'duration': info['duration'] // 1000,

         }
         }