New upstream version 2018.04.25

[youtubedl] / youtube_dl / extractor / veehd.py

diff --git a/youtube_dl/extractor/veehd.py b/youtube_dl/extractor/veehd.py
index 3cf8c853d2e466e00228d7eb3cb0f33d664beb9b..0d8d832cc0890a8cb99c8388b1b9fc1964396d23 100644 (file)
--- a/youtube_dl/extractor/veehd.py
+++ b/youtube_dl/extractor/veehd.py
@@ -1,55 +1,117 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
 from .common import InfoExtractor
-from ..utils import (
+from ..compat import (
+    compat_urllib_parse_unquote,
     compat_urlparse,
-    get_element_by_id,
+)
+from ..utils import (
+    ExtractorError,
     clean_html,
+    get_element_by_id,
 )
 
+
 class VeeHDIE(InfoExtractor):
     _VALID_URL = r'https?://veehd\.com/video/(?P<id>\d+)'
 
-    _TEST = {
-        u'url': u'http://veehd.com/video/4686958',
-        u'file': u'4686958.mp4',
-        u'info_dict': {
-            u'title': u'Time Lapse View from Space ( ISS)',
-            u'uploader_id': u'spotted',
-            u'description': u'md5:f0094c4cf3a72e22bc4e4239ef767ad7',
+    # Seems VeeHD videos have multiple copies on several servers, all of
+    # whom have different MD5 checksums, so omit md5 field in all tests
+    _TESTS = [{
+        'url': 'http://veehd.com/video/4639434_Solar-Sinter',
+        'info_dict': {
+            'id': '4639434',
+            'ext': 'mp4',
+            'title': 'Solar Sinter',
+            'uploader_id': 'VideoEyes',
+            'description': 'md5:46a840e8692ddbaffb5f81d9885cb457',
         },
-    }
+        'skip': 'Video deleted',
+    }, {
+        'url': 'http://veehd.com/video/4905758_Elysian-Fields-Channeling',
+        'info_dict': {
+            'id': '4905758',
+            'ext': 'mp4',
+            'title': 'Elysian Fields - Channeling',
+            'description': 'md5:360e4e95fdab58aefbea0f2a19e5604b',
+            'uploader_id': 'spotted',
+        }
+    }, {
+        'url': 'http://veehd.com/video/2046729_2012-2009-DivX-Trailer',
+        'info_dict': {
+            'id': '2046729',
+            'ext': 'avi',
+            'title': '2012 (2009) DivX Trailer',
+            'description': 'md5:75435ee95255e6a9838ac6f6f3a2396b',
+            'uploader_id': 'Movie_Trailers',
+        }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
+        # VeeHD seems to send garbage on the first request.
+        # See https://github.com/rg3/youtube-dl/issues/2102
+        self._download_webpage(url, video_id, 'Requesting webpage')
         webpage = self._download_webpage(url, video_id)
-        player_path = self._search_regex(r'\$\("#playeriframe"\).attr\({src : "(.+?)"',
-            webpage, u'player path')
+
+        if 'This video has been removed<' in webpage:
+            raise ExtractorError('Video %s has been removed' % video_id, expected=True)
+
+        player_path = self._search_regex(
+            r'\$\("#playeriframe"\).attr\({src : "(.+?)"',
+            webpage, 'player path')
         player_url = compat_urlparse.urljoin(url, player_path)
-        player_page = self._download_webpage(player_url, video_id,
-            u'Downloading player page')
-        config_json = self._search_regex(r'value=\'config=({.+?})\'',
-            player_page, u'config json')
-        config = json.loads(config_json)
 
-        video_url = compat_urlparse.unquote(config['clip']['url'])
+        self._download_webpage(player_url, video_id, 'Requesting player page')
+        player_page = self._download_webpage(
+            player_url, video_id, 'Downloading player page')
+
+        video_url = None
+
+        config_json = self._search_regex(
+            r'value=\'config=({.+?})\'', player_page, 'config json', default=None)
+
+        if config_json:
+            config = json.loads(config_json)
+            video_url = compat_urllib_parse_unquote(config['clip']['url'])
+
+        if not video_url:
+            video_url = self._html_search_regex(
+                r'<embed[^>]+type="video/divx"[^>]+src="([^"]+)"',
+                player_page, 'video url', default=None)
+
+        if not video_url:
+            iframe_src = self._search_regex(
+                r'<iframe[^>]+src="/?([^"]+)"', player_page, 'iframe url')
+            iframe_url = 'http://veehd.com/%s' % iframe_src
+
+            self._download_webpage(iframe_url, video_id, 'Requesting iframe page')
+            iframe_page = self._download_webpage(
+                iframe_url, video_id, 'Downloading iframe page')
+
+            video_url = self._search_regex(
+                r"file\s*:\s*'([^']+)'", iframe_page, 'video url')
+
         title = clean_html(get_element_by_id('videoName', webpage).rpartition('|')[0])
-        uploader_id = self._html_search_regex(r'<a href="/profile/\d+">(.+?)</a>',
-            webpage, u'uploader')
-        thumbnail = self._search_regex(r'<img id="veehdpreview" src="(.+?)"',
-            webpage, u'thumbnail')
-        description = self._html_search_regex(r'<td class="infodropdown".*?<div>(.*?)<ul',
-            webpage, u'description', flags=re.DOTALL)
+        uploader_id = self._html_search_regex(
+            r'<a href="/profile/\d+">(.+?)</a>',
+            webpage, 'uploader')
+        thumbnail = self._search_regex(
+            r'<img id="veehdpreview" src="(.+?)"',
+            webpage, 'thumbnail')
+        description = self._html_search_regex(
+            r'<td class="infodropdown".*?<div>(.*?)<ul',
+            webpage, 'description', flags=re.DOTALL)
 
         return {
             '_type': 'video',
             'id': video_id,
             'title': title,
             'url': video_url,
-            'ext': 'mp4',
             'uploader_id': uploader_id,
             'thumbnail': thumbnail,
             'description': description,