Initiate new release.

[youtubedl] / youtube_dl / extractor / theplatform.py

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index d60702325d7d4ffebafcb45cdc9054c8fa4d2af7..feac666f78baff49f4fb312a147acad67d320bc2 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -1,10 +1,22 @@
+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 ..compat import (
+    compat_str,
+)

 from ..utils import (
 from ..utils import (

+    determine_ext,

     ExtractorError,
     xpath_with_ns,
     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'})


@@ -12,93 +24,148 @@ _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]/[^/]+/
-           (?P<config>[^/\?]+/(?:swf|config)/select/)?
+        (?:https?://(?:link|player)\.theplatform\.com/[sp]/(?P<provider_id>[^/]+)/
+           (?P<config>(?:[^/\?]+/(?:swf|config)|onsite)/select/)?

          |theplatform:)(?P<id>[^/\?&]+)'''
 
     _TEST = {
         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/
          |theplatform:)(?P<id>[^/\?&]+)'''
 
     _TEST = {
         # from http://www.metacafe.com/watch/cb-e9I_cZgTgIPd/blackberrys_big_bold_z30/

-        u'url': u'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
-        u'info_dict': {
-            u'id': u'e9I_cZgTgIPd',
-            u'ext': u'flv',
-            u'title': u'Blackberry\'s big, bold Z30',
-            u'description': u'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
-            u'duration': 247,
+        'url': 'http://link.theplatform.com/s/dJ5BDC/e9I_cZgTgIPd/meta.smil?format=smil&Tracking=true&mbr=true',
+        'info_dict': {
+            'id': 'e9I_cZgTgIPd',
+            'ext': 'flv',
+            'title': 'Blackberry\'s big, bold Z30',
+            'description': 'The Z30 is Blackberry\'s biggest, baddest mobile messaging device yet.',
+            'duration': 247,

         },
         },

-        u'params': {
+        'params': {

             # rtmp download
             # rtmp download

-            u'skip_download': True,
+            'skip_download': True,

         },
     }
 
         },
     }
 

-    def _get_info(self, video_id, smil_url):
-        meta = self._download_xml(smil_url, video_id)
+    @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):
+        url, smuggled_data = unsmuggle_url(url, {})
+
+        mobj = re.match(self._VALID_URL, url)
+        provider_id = mobj.group('provider_id')
+        video_id = mobj.group('id')
+
+        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:
+            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:
             error_msg = next(
                 n.attrib['abstract']
                 for n in meta.findall(_x('.//smil:ref'))
         try:
             error_msg = next(
                 n.attrib['abstract']
                 for n in meta.findall(_x('.//smil:ref'))

-                if n.attrib.get('title') == u'Geographic Restriction')
+                if n.attrib.get('title') == 'Geographic Restriction')

         except StopIteration:
             pass
         else:
             raise ExtractorError(error_msg, expected=True)
 
         except StopIteration:
             pass
         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'))
 

-        f4m_node = body.find(_x('smil:seq/smil:video'))
-        if f4m_node is not None:
-            formats = [{
-                'ext': 'flv',
-                # the parameters are from syfy.com, other sites may use others
-                'url': f4m_node.attrib['src'] + '?g=UXWGVKRWHFSP&hdcore=3.0.3',
-            }]
+        f4m_node = body.find(_x('smil:seq//smil:video'))
+        if f4m_node is not None and '.f4m' in f4m_node.attrib['src']:
+            f4m_url = f4m_node.attrib['src']
+            if 'manifest.f4m?' not in f4m_url:
+                f4m_url += '?'
+            # the parameters are from syfy.com, other sites may use others,
+            # they also work for nbc.com
+            f4m_url += '&g=UXWGVKRWHFSP&hdcore=3.0.3'
+            formats = self._extract_f4m_formats(f4m_url, video_id)

         else:
         else:

-            base_url = head.find(_x('smil:meta')).attrib['base']
-            switch = body.find(_x('smil:switch'))

             formats = []
             formats = []

-            for f in switch.findall(_x('smil:video')):
-                attr = f.attrib
-                width = int(attr['width'])
-                height = int(attr['height'])
-                vbr = int(attr['system-bitrate']) // 1000
-                format_id = '%dx%d_%dk' % (width, height, vbr)
-                formats.append({
-                    'format_id': format_id,
-                    'url': base_url,
-                    'play_path': 'mp4:' + attr['src'],
-                    'ext': 'flv',
-                    'width': width,
-                    'height': height,
-                    'vbr': vbr,
-                })
+            switch = body.find(_x('smil:switch'))
+            if switch is not None:
+                base_url = head.find(_x('smil:meta')).attrib['base']
+                for f in switch.findall(_x('smil:video')):
+                    attr = f.attrib
+                    width = int(attr['width'])
+                    height = int(attr['height'])
+                    vbr = int(attr['system-bitrate']) // 1000
+                    format_id = '%dx%d_%dk' % (width, height, vbr)
+                    formats.append({
+                        'format_id': format_id,
+                        'url': base_url,
+                        'play_path': 'mp4:' + attr['src'],
+                        'ext': 'flv',
+                        'width': width,
+                        'height': height,
+                        'vbr': vbr,
+                    })
+            else:
+                switch = body.find(_x('smil:seq//smil:switch'))
+                for f in switch.findall(_x('smil:video')):
+                    attr = f.attrib
+                    vbr = int(attr['system-bitrate']) // 1000
+                    ext = determine_ext(attr['src'])
+                    if ext == 'once':
+                        ext = 'mp4'
+                    formats.append({
+                        'format_id': compat_str(vbr),
+                        'url': attr['src'],
+                        'vbr': vbr,
+                        'ext': ext,
+                    })

             self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': info['title'],
             self._sort_formats(formats)
 
         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,

         }
         }

-        
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        if mobj.group('config'):
-            config_url = url+ '&form=json'
-            config_url = config_url.replace('swf/', 'config/')
-            config_json = self._download_webpage(config_url, video_id, u'Downloading config')
-            config = json.loads(config_json)
-            smil_url = config['releaseUrl'] + '&format=SMIL&formats=MPEG4'
-        else:
-            smil_url = ('http://link.theplatform.com/s/dJ5BDC/{0}/meta.smil?'
-                'format=smil&mbr=true'.format(video_id))
-        return self._get_info(video_id, smil_url)