Prepare to upload.

[youtubedl] / youtube_dl / extractor / infoq.py

diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py
index 71cfd12c56549d0be540c9daee6a2732959039de..18249cf9b4a29fabdcc7cf4b41061671e7ea33e8 100644 (file)
--- a/youtube_dl/extractor/infoq.py
+++ b/youtube_dl/extractor/infoq.py
@@ -1,22 +1,27 @@
-from __future__ import unicode_literals
+# coding: utf-8

 
 

-import base64
+from __future__ import unicode_literals

 
 

-from .common import InfoExtractor

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

+    compat_b64decode,

     compat_urllib_parse_unquote,
     compat_urlparse,
 )
     compat_urllib_parse_unquote,
     compat_urlparse,
 )

+from ..utils import (
+    determine_ext,
+    update_url_query,
+)
+from .bokecc import BokeCCBaseIE

 
 
 
 

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

     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{
         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
         'info_dict': {
     _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
 
     _TESTS = [{
         'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
         'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
         'info_dict': {

-            'id': '12-jan-pythonthings',
+            'id': 'A-Few-of-My-Favorite-Python-Things',

             'ext': 'mp4',
             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
             'title': 'A Few of My Favorite [Python] Things',
             'ext': 'mp4',
             'description': 'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
             'title': 'A Few of My Favorite [Python] Things',


@@ -24,40 +29,103 @@ class InfoQIE(InfoExtractor):
     }, {
         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
         'only_matching': True,
     }, {
         'url': 'http://www.infoq.com/fr/presentations/changez-avis-sur-javascript',
         'only_matching': True,

+    }, {
+        'url': 'http://www.infoq.com/cn/presentations/openstack-continued-delivery',
+        'md5': '4918d0cca1497f2244572caf626687ef',
+        'info_dict': {
+            'id': 'openstack-continued-delivery',
+            'title': 'OpenStack持续交付之路',
+            '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 _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
-        video_description = self._html_search_meta('description', webpage, 'description')
-
+    def _extract_rtmp_video(self, webpage):

         # The server URL is hardcoded
         video_url = 'rtmpe://video.infoq.com/cfx/st/'
 
         # Extract video URL
         encoded_id = self._search_regex(
         # The server URL is hardcoded
         video_url = 'rtmpe://video.infoq.com/cfx/st/'
 
         # Extract video URL
         encoded_id = self._search_regex(

-            r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id')
-        real_id = compat_urllib_parse_unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
-        playpath = 'mp4:' + real_id
+            r"jsclassref\s*=\s*'([^']*)'", webpage, 'encoded id', default=None)

 
 

-        video_filename = playpath.split('/')[-1]
-        video_id, extension = video_filename.split('.')
-
-        http_base = self._search_regex(
-            r'EXPRESSINSTALL_SWF\s*=\s*[^"]*"((?:https?:)?//[^/"]+/)', webpage,
-            'HTTP base URL')
+        real_id = compat_urllib_parse_unquote(compat_b64decode(encoded_id).decode('utf-8'))
+        playpath = 'mp4:' + real_id

 
 

-        formats = [{
-            'format_id': 'rtmp',
+        return [{
+            'format_id': 'rtmp_video',

             'url': video_url,
             'url': video_url,

-            'ext': extension,
+            'ext': determine_ext(playpath),

             'play_path': playpath,
             'play_path': playpath,

-        }, {
-            'format_id': 'http',
-            'url': compat_urlparse.urljoin(url, http_base) + real_id,

         }]
         }]

+
+    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 = 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 []
+
+        # 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 [{
+            'format_id': 'http_audio',
+            'url': http_audio_url,
+            'vcodec': 'none',
+        }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
+        video_description = self._html_search_meta('description', webpage, 'description')
+
+        if '/cn/' in url:
+            # for China videos, HTTP video URL exists but always fails with 403
+            formats = self._extract_bokecc_formats(webpage, video_id)
+        else:
+            formats = (
+                self._extract_rtmp_video(webpage)
+                + self._extract_http_video(webpage)
+                + self._extract_http_audio(webpage, video_id))
+

         self._sort_formats(formats)
 
         return {
         self._sort_formats(formats)
 
         return {