Update upstream source from tag 'upstream/2019.06.08'

[youtubedl] / youtube_dl / extractor / infoq.py

diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py
index 016af2084b3301bccf6f6191df66b2eec84968a4..18249cf9b4a29fabdcc7cf4b41061671e7ea33e8 100644 (file)
--- a/youtube_dl/extractor/infoq.py
+++ b/youtube_dl/extractor/infoq.py
@@ -2,17 +2,19 @@
 
 from __future__ import unicode_literals
 
 
 from __future__ import unicode_literals
 

-import base64
-
-from .common import InfoExtractor

 from ..compat import (
 from ..compat import (

+    compat_b64decode,

     compat_urllib_parse_unquote,
     compat_urllib_parse_unquote,

-    compat_parse_qs,
+    compat_urlparse,
+)
+from ..utils import (
+    determine_ext,
+    update_url_query,

 )
 )

-from ..utils import determine_ext
+from .bokecc import BokeCCBaseIE

 
 
 
 

-class InfoQIE(InfoExtractor):
+class InfoQIE(BokeCCBaseIE):

     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{
     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{


@@ -36,29 +38,21 @@ class InfoQIE(InfoExtractor):
             'ext': 'flv',
             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
         },
             'ext': 'flv',
             'description': 'md5:308d981fb28fa42f49f9568322c683ff',
         },

+    }, {
+        'url': 'https://www.infoq.com/presentations/Simple-Made-Easy',
+        'md5': '0e34642d4d9ef44bf86f66f6399672db',
+        'info_dict': {
+            'id': 'Simple-Made-Easy',
+            'title': 'Simple Made Easy',
+            'ext': 'mp3',
+            'description': 'md5:3e0e213a8bbd074796ef89ea35ada25b',
+        },
+        'params': {
+            'format': 'bestaudio',
+        },

     }]
 
     }]
 

-    def _extract_bokecc_videos(self, webpage, video_id):
-        # TODO: bokecc.com is a Chinese video cloud platform
-        # It should have an independent extractor but I don't have other
-        # examples using bokecc
-        player_params_str = self._html_search_regex(
-            r'<script[^>]+src="http://p\.bokecc\.com/player\?([^"]+)',
-            webpage, 'player params', default=None)
-
-        player_params = compat_parse_qs(player_params_str)
-
-        info_xml = self._download_xml(
-            'http://p.bokecc.com/servlet/playinfo?uid=%s&vid=%s&m=1' % (
-                player_params['siteid'][0], player_params['vid'][0]), video_id)
-
-        return [{
-            'format_id': 'bokecc',
-            'url': quality.find('./copy').attrib['playurl'],
-            'preference': int(quality.attrib['value']),
-        } for quality in info_xml.findall('./video/quality')]
-
-    def _extract_rtmp_videos(self, webpage):
+    def _extract_rtmp_video(self, webpage):

         # The server URL is hardcoded
         video_url = 'rtmpe://video.infoq.com/cfx/st/'
 
         # The server URL is hardcoded
         video_url = 'rtmpe://video.infoq.com/cfx/st/'
 


@@ -66,30 +60,54 @@ class InfoQIE(InfoExtractor):
         encoded_id = self._search_regex(
             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
 
         encoded_id = self._search_regex(
             r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)
 

-        real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
+        real_id = compat_urllib_parse_unquote(compat_b64decode(encoded_id).decode('utf-8'))

         playpath = 'mp4:' + real_id
 
         return [{
         playpath = 'mp4:' + real_id
 
         return [{

-            'format_id': 'rtmp',
+            'format_id': 'rtmp_video',

             'url': video_url,
             'ext': determine_ext(playpath),
             'play_path': playpath,
         }]
 
             'url': video_url,
             'ext': determine_ext(playpath),
             'play_path': playpath,
         }]
 

-    def _extract_http_videos(self, webpage):
+    def _extract_cf_auth(self, webpage):
+        policy = self._search_regex(r'InfoQConstants\.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
+        signature = self._search_regex(r'InfoQConstants\.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
+        key_pair_id = self._search_regex(r'InfoQConstants\.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
+        return {
+            'Policy': policy,
+            'Signature': signature,
+            'Key-Pair-Id': key_pair_id,
+        }
+
+    def _extract_http_video(self, webpage):

         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
         http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')

+        http_video_url = update_url_query(http_video_url, self._extract_cf_auth(webpage))
+        return [{
+            'format_id': 'http_video',
+            'url': http_video_url,
+        }]
+
+    def _extract_http_audio(self, webpage, video_id):
+        fields = self._hidden_inputs(webpage)
+        http_audio_url = fields.get('filename')
+        if not http_audio_url:
+            return []

 
 

-        policy = self._search_regex(r'InfoQConstants.scp\s*=\s*\'([^\']+)\'', webpage, 'policy')
-        signature = self._search_regex(r'InfoQConstants.scs\s*=\s*\'([^\']+)\'', webpage, 'signature')
-        key_pair_id = self._search_regex(r'InfoQConstants.sck\s*=\s*\'([^\']+)\'', webpage, 'key-pair-id')
+        # base URL is found in the Location header in the response returned by
+        # GET https://www.infoq.com/mp3download.action?filename=... when logged in.
+        http_audio_url = compat_urlparse.urljoin('http://res.infoq.com/downloads/mp3downloads/', http_audio_url)
+        http_audio_url = update_url_query(http_audio_url, self._extract_cf_auth(webpage))
+
+        # audio file seem to be missing some times even if there is a download link
+        # so probe URL to make sure
+        if not self._is_valid_url(http_audio_url, video_id):
+            return []

 
         return [{
 
         return [{

-            'format_id': 'http',
-            'url': http_video_url,
-            'http_headers': {
-                'Cookie': 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
-                    policy, signature, key_pair_id),
-            },
+            'format_id': 'http_audio',
+            'url': http_audio_url,
+            'vcodec': 'none',

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


@@ -101,9 +119,12 @@ class InfoQIE(InfoExtractor):
 
         if '/cn/' in url:
             # for China videos, HTTP video URL exists but always fails with 403
 
         if '/cn/' in url:
             # for China videos, HTTP video URL exists but always fails with 403

-            formats = self._extract_bokecc_videos(webpage, video_id)
+            formats = self._extract_bokecc_formats(webpage, video_id)

         else:
         else:

-            formats = self._extract_rtmp_videos(webpage) + self._extract_http_videos(webpage)
+            formats = (
+                self._extract_rtmp_video(webpage)
+                + self._extract_http_video(webpage)
+                + self._extract_http_audio(webpage, video_id))

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