Merge tag 'upstream/2017.03.07'

[youtubedl] / youtube_dl / extractor / infoq.py

diff --git a/youtube_dl/extractor/infoq.py b/youtube_dl/extractor/infoq.py
index 9056742821a2fd99fadd5b1271df7dc6f5127749..9fb71e8effe107c6e182c3537cd634b7ac21e9bb 100644 (file)
--- a/youtube_dl/extractor/infoq.py
+++ b/youtube_dl/extractor/infoq.py
@@ -1,50 +1,135 @@
-import base64
-import re
+# coding: utf-8
+
+from __future__ import unicode_literals

 
 

-from .common import InfoExtractor
-from ..utils import (
-    compat_urllib_parse,
+import base64

 
 

-    ExtractorError,
+from ..compat import (
+    compat_urllib_parse_unquote,
+    compat_urlparse,

 )
 )

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

 
 
 
 

-class InfoQIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?infoq\.com/[^/]+/[^/]+$'
+class InfoQIE(BokeCCBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?infoq\.com/(?:[^/]+/)+(?P<id>[^/]+)'

 
 

-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+    _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',
+        },
+    }, {
+        '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',
+        },
+    }]

 
 

-        webpage = self._download_webpage(url, video_id=url)
-        self.report_extraction(url)
+    def _extract_rtmp_video(self, webpage):
+        # The server URL is hardcoded
+        video_url = 'rtmpe://video.infoq.com/cfx/st/'

 
         # Extract video URL
 
         # Extract video URL

-        mobj = re.search(r"jsclassref ?= ?'([^']*)'", webpage)
-        if mobj is None:
-            raise ExtractorError(u'Unable to extract video url')
-        real_id = compat_urllib_parse.unquote(base64.b64decode(mobj.group(1).encode('ascii')).decode('utf-8'))
-        video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id
+        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
+
+        return [{
+            'format_id': 'rtmp_video',
+            'url': video_url,
+            'ext': determine_ext(playpath),
+            'play_path': playpath,
+        }]
+
+    def _extract_cookies(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 'CloudFront-Policy=%s; CloudFront-Signature=%s; CloudFront-Key-Pair-Id=%s' % (
+            policy, signature, key_pair_id)
+
+    def _extract_http_video(self, webpage):
+        http_video_url = self._search_regex(r'P\.s\s*=\s*\'([^\']+)\'', webpage, 'video URL')
+        return [{
+            'format_id': 'http_video',
+            'url': http_video_url,
+            'http_headers': {
+                'Cookie': self._extract_cookies(webpage)
+            },
+        }]

 
 

-        # Extract title
-        video_title = self._search_regex(r'contentTitle = "(.*?)";',
-            webpage, u'title')
+    def _extract_http_audio(self, webpage, video_id):
+        fields = self._hidden_inputs(webpage)
+        http_audio_url = fields['filename']
+        if http_audio_url is None:
+            return []

 
 

-        # Extract description
-        video_description = self._html_search_regex(r'<meta name="description" content="(.*)"(?:\s*/)?>',
-            webpage, u'description', fatal=False)
+        cookies_header = {'Cookie': self._extract_cookies(webpage)}

 
 

-        video_filename = video_url.split('/')[-1]
-        video_id, extension = video_filename.split('.')
+        # 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)

 
 

-        info = {
+        # 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, headers=cookies_header):
+            return []
+
+        return [{
+            'format_id': 'http_audio',
+            'url': http_audio_url,
+            'vcodec': 'none',
+            'http_headers': cookies_header,
+        }]
+
+    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 {

             'id': video_id,
             'id': video_id,

-            'url': video_url,
-            'uploader': None,
-            'upload_date': None,

             'title': video_title,
             'title': video_title,

-            'ext': extension, # Extension is always(?) mp4, but seems to be flv
-            'thumbnail': None,

             'description': video_description,
             'description': video_description,

+            'formats': formats,

         }
         }

-
-        return [info]
\ No newline at end of file