X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Finfoq.py;h=cca0b8a9323c0d2412c65610a3acb3ef2943ba6f;hb=4875ff68476ff7de9733c80effb652fc6ab07ea0;hp=65f0b76cb0e2a413e82005405b012b4a93fae107;hpb=7560096db5ce0002b3cf3f3bcbbb53ef05e6fb13;p=youtube-dl

diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py
index 65f0b76cb..cca0b8a93 100644
--- a/youtube_dl/extractor/infoq.py
+++ b/youtube_dl/extractor/infoq.py
@@ -1,58 +1,92 @@
+# coding: utf-8
+
 from __future__ import unicode_literals
 
 import base64
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse,
-)
-
-
-class InfoQIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P<id>[^/]+)$'
-
-    _TEST = {
-        u'name': u'InfoQ',
-        u'url': u'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
-        u'md5': u'fcaa3d995e04080dcb9465d86b5eef62',
-        u'info_dict': {
-            u'id': u'12-jan-pythonthings',
-            u'ext': u'mp4',
-            u'description': u'Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.',
-            u'title': u'A Few of My Favorite [Python] Things',
-        },
-    }
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+from ..compat import compat_urllib_parse_unquote
+from ..utils import determine_ext
+from .bokecc import BokeCCBaseIE
 
-        webpage = self._download_webpage(url, video_id)
 
-        self.report_extraction(video_id)
+class InfoQIE(BokeCCBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'
 
-        video_title = self._html_search_regex(r'<title>(.*?)</title>', webpage, 'title')
-        video_description = self._html_search_meta('description', webpage, 'description')
+    _TESTS = [{
+        'url': 'http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things',
+        'md5': 'b5ca0e0a8c1fed93b0e65e48e462f9a2',
+        'info_dict': {
+            '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',
+        },
+    }, {
+        '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',
+        },
+    }]
 
+    def _extract_rtmp_videos(self, webpage):
         # The server URL is hardcoded
         video_url = 'rtmpe://video.infoq.com/cfx/st/'
 
         # Extract video URL
-        encoded_id = self._search_regex(r"jsclassref ?= ?'([^']*)'", webpage, 'encoded id')
-        real_id = compat_urllib_parse.unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
+        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'))
         playpath = 'mp4:' + real_id
 
-        video_filename = playpath.split('/')[-1]
-        video_id, extension = video_filename.split('.')
+        return [{
+            'format_id': 'rtmp',
+            'url': video_url,
+            'ext': determine_ext(playpath),
+            'play_path': playpath,
+        }]
+
+    def _extract_http_videos(self, webpage):
+        http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
+
+        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 [{
+            '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),
+            },
+        }]
+
+    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_videos(webpage) + self._extract_http_videos(webpage)
+
+        self._sort_formats(formats)
+
+        return {
             'id': video_id,
             'title': video_title,
             'description': video_description,
-            'formats': [{
-                'url': video_url,
-                'ext': extension,
-                'play_path': playpath,
-            }],
-        }]
+            'formats': formats,
+        }