Fix extraction from youtube.

[youtubedl] / youtube_dl / extractor / shared.py

diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py
index e7e5f653eb2117936f568195e508f9e7778b1085..931a0f70e0e0924689dfb6b71c374e6fa2c14457 100644 (file)
--- a/youtube_dl/extractor/shared.py
+++ b/youtube_dl/extractor/shared.py
@@ -1,21 +1,47 @@
 from __future__ import unicode_literals
 
-import base64
-
 from .common import InfoExtractor
+from ..compat import compat_b64decode
 from ..utils import (
     ExtractorError,
     int_or_none,
-    sanitized_Request,
+    url_or_none,
     urlencode_postdata,
 )
 
 
-class SharedIE(InfoExtractor):
-    IE_DESC = 'shared.sx and vivo.sx'
-    _VALID_URL = r'https?://(?:shared|vivo)\.sx/(?P<id>[\da-z]{10})'
+class SharedBaseIE(InfoExtractor):
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage, urlh = self._download_webpage_handle(url, video_id)
+
+        if self._FILE_NOT_FOUND in webpage:
+            raise ExtractorError(
+                'Video %s does not exist' % video_id, expected=True)
+
+        video_url = self._extract_video_url(webpage, video_id, url)
+
+        title = compat_b64decode(self._html_search_meta(
+            'full:title', webpage, 'title')).decode('utf-8')
+        filesize = int_or_none(self._html_search_meta(
+            'full:size', webpage, 'file size', fatal=False))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'filesize': filesize,
+            'title': title,
+        }
+
 
-    _TESTS = [{
+class SharedIE(SharedBaseIE):
+    IE_DESC = 'shared.sx'
+    _VALID_URL = r'https?://shared\.sx/(?P<id>[\da-z]{10})'
+    _FILE_NOT_FOUND = '>File does not exist<'
+
+    _TEST = {
         'url': 'http://shared.sx/0060718775',
         'md5': '106fefed92a8a2adb8c98e6a0652f49b',
         'info_dict': {
@@ -24,7 +50,32 @@ class SharedIE(InfoExtractor):
             'title': 'Bmp4',
             'filesize': 1720110,
         },
-    }, {
+    }
+
+    def _extract_video_url(self, webpage, video_id, url):
+        download_form = self._hidden_inputs(webpage)
+
+        video_page = self._download_webpage(
+            url, video_id, 'Downloading video page',
+            data=urlencode_postdata(download_form),
+            headers={
+                'Content-Type': 'application/x-www-form-urlencoded',
+                'Referer': url,
+            })
+
+        video_url = self._html_search_regex(
+            r'data-url=(["\'])(?P<url>(?:(?!\1).)+)\1',
+            video_page, 'video URL', group='url')
+
+        return video_url
+
+
+class VivoIE(SharedBaseIE):
+    IE_DESC = 'vivo.sx'
+    _VALID_URL = r'https?://vivo\.sx/(?P<id>[\da-z]{10})'
+    _FILE_NOT_FOUND = '>The file you have requested does not exists or has been removed'
+
+    _TEST = {
         'url': 'http://vivo.sx/d7ddda0e78',
         'md5': '15b3af41be0b4fe01f4df075c2678b2c',
         'info_dict': {
@@ -33,38 +84,19 @@ class SharedIE(InfoExtractor):
             'title': 'Chicken',
             'filesize': 528031,
         },
-    }]
+    }
 
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+    def _extract_video_url(self, webpage, video_id, *args):
+        def decode_url(encoded_url):
+            return compat_b64decode(encoded_url).decode('utf-8')
 
-        if '>File does not exist<' in webpage:
-            raise ExtractorError(
-                'Video %s does not exist' % video_id, expected=True)
-
-        download_form = self._hidden_inputs(webpage)
-        request = sanitized_Request(
-            url, urlencode_postdata(download_form))
-        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
-
-        video_page = self._download_webpage(
-            request, video_id, 'Downloading video page')
-
-        video_url = self._html_search_regex(
-            r'data-url="([^"]+)"', video_page, 'video URL')
-        title = base64.b64decode(self._html_search_meta(
-            'full:title', webpage, 'title').encode('utf-8')).decode('utf-8')
-        filesize = int_or_none(self._html_search_meta(
-            'full:size', webpage, 'file size', fatal=False))
-        thumbnail = self._html_search_regex(
-            r'data-poster="([^"]+)"', video_page, 'thumbnail', default=None)
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'ext': 'mp4',
-            'filesize': filesize,
-            'title': title,
-            'thumbnail': thumbnail,
-        }
+        stream_url = url_or_none(decode_url(self._search_regex(
+            r'data-stream\s*=\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage,
+            'stream url', default=None, group='url')))
+        if stream_url:
+            return stream_url
+        return self._parse_json(
+            self._search_regex(
+                r'InitializeStream\s*\(\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
+                webpage, 'stream', group='url'),
+            video_id, transform_source=decode_url)[0]