Merge remote-tracking branch 'akretz/youtube_fix'

diff --git a/youtube_dl/extractor/behindkink.py b/youtube_dl/extractor/behindkink.py
index 31fdc0dcc0614babf4ff3b48186566904cfcc57a..1bdc25812b6afb4cf133007f2d12b89fd56b353f 100644 (file)
--- a/youtube_dl/extractor/behindkink.py
+++ b/youtube_dl/extractor/behindkink.py
@@ -10,15 +10,15 @@ from ..utils import url_basename
 class BehindKinkIE(InfoExtractor):
     _VALID_URL = r'http://(?:www\.)?behindkink\.com/(?P<year>[0-9]{4})/(?P<month>[0-9]{2})/(?P<day>[0-9]{2})/(?P<id>[^/#?_]+)'
     _TEST = {
-        'url': 'http://www.behindkink.com/2014/08/14/ab1576-performers-voice-finally-heard-the-bill-is-killed/',
-        'md5': '41ad01222b8442089a55528fec43ec01',
+        'url': 'http://www.behindkink.com/2014/12/05/what-are-you-passionate-about-marley-blaze/',
+        'md5': '507b57d8fdcd75a41a9a7bdb7989c762',
         'info_dict': {
-            'id': '36370',
+            'id': '37127',
             'ext': 'mp4',
-            'title': 'AB1576 - PERFORMERS VOICE FINALLY HEARD - THE BILL IS KILLED!',
-            'description': 'The adult industry voice was finally heard as Assembly Bill 1576 remained\xa0 in suspense today at the Senate Appropriations Hearing. AB1576 was, among other industry damaging issues, a condom mandate...',
-            'upload_date': '20140814',
-            'thumbnail': 'http://www.behindkink.com/wp-content/uploads/2014/08/36370_AB1576_Win.jpg',
+            'title': 'What are you passionate about – Marley Blaze',
+            'description': 'md5:aee8e9611b4ff70186f752975d9b94b4',
+            'upload_date': '20141205',
+            'thumbnail': 'http://www.behindkink.com/wp-content/uploads/2014/12/blaze-1.jpg',
             'age_limit': 18,
         }
     }
@@ -26,26 +26,19 @@ class BehindKinkIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         display_id = mobj.group('id')
-        year = mobj.group('year')
-        month = mobj.group('month')
-        day = mobj.group('day')
-        upload_date = year + month + day
 
         webpage = self._download_webpage(url, display_id)
 
         video_url = self._search_regex(
-            r"'file':\s*'([^']+)'",
-            webpage, 'URL base')
-
-        video_id = url_basename(video_url)
-        video_id = video_id.split('_')[0]
+            r'<source src="([^"]+)"', webpage, 'video URL')
+        video_id = url_basename(video_url).split('_')[0]
+        upload_date = mobj.group('year') + mobj.group('month') + mobj.group('day')
 
         return {
             'id': video_id,
+            'display_id': display_id,
             'url': video_url,
-            'ext': 'mp4',
             'title': self._og_search_title(webpage),
-            'display_id': display_id,
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': self._og_search_description(webpage),
             'upload_date': upload_date,

diff --git a/youtube_dl/extractor/ntv.py b/youtube_dl/extractor/ntv.py
index 13c8d79cd8ac6346dbe4e9810bc8bf0b20825dcc..ee740cd9c0fe71a48b79aee00c40ea610e81ea99 100644 (file)
--- a/youtube_dl/extractor/ntv.py
+++ b/youtube_dl/extractor/ntv.py
@@ -130,7 +130,7 @@ class NTVIE(InfoExtractor):
                 'rtmp_conn': 'B:1',
                 'player_url': 'http://www.ntv.ru/swf/vps1.swf?update=20131128',
                 'page_url': 'http://www.ntv.ru',
-                'flash_ver': 'LNX 11,2,202,341',
+                'flash_version': 'LNX 11,2,202,341',
                 'rtmp_live': True,
                 'ext': 'flv',
                 'filesize': int(size.text),

diff --git a/youtube_dl/extractor/tvplay.py b/youtube_dl/extractor/tvplay.py
index eb94737546fb725e992249d356588417d90ad81a..0157392cce30984f62024f1c9588cfed5b257cd2 100644 (file)
--- a/youtube_dl/extractor/tvplay.py
+++ b/youtube_dl/extractor/tvplay.py
@@ -182,8 +182,8 @@ class TVPlayIE(InfoExtractor):
             'http://playapi.mtgx.tv/v1/videos/%s' % video_id, video_id, 'Downloading video JSON')
 
         if video['is_geo_blocked']:
-            raise ExtractorError(
-                'This content is not available in your country due to copyright reasons', expected=True)
+            self.report_warning(
+                'This content might not be available in your country due to copyright reasons')
 
         streams = self._download_json(
             'http://playapi.mtgx.tv/v1/videos/stream/%s' % video_id, video_id, 'Downloading streams JSON')

diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index faac98ce08dabd077d9bf1f8bd1b10b9aea83207..f8d5cdd5d6210477f5d286facfa52921dc5b4186 100644 (file)
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -14,23 +14,24 @@ from .common import InfoExtractor, SearchInfoExtractor
 from .subtitles import SubtitlesInfoExtractor
 from ..jsinterp import JSInterpreter
 from ..swfinterp import SWFInterpreter
-from ..utils import (
+from ..compat import (
     compat_chr,
     compat_parse_qs,
     compat_urllib_parse,
     compat_urllib_request,
     compat_urlparse,
     compat_str,
-
+)
+from ..utils import (
     clean_html,
-    get_element_by_id,
-    get_element_by_attribute,
     ExtractorError,
+    get_element_by_attribute,
+    get_element_by_id,
     int_or_none,
     OnDemandPagedList,
+    orderedSet,
     unescapeHTML,
     unified_strdate,
-    orderedSet,
     uppercase_escape,
 )
 
@@ -432,6 +433,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
             'expected_warnings': [
                 'DASH manifest missing',
             ]
+        },
+        # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
+        {
+            'url': 'lqQg6PlCWgI',
+            'info_dict': {
+                'id': 'lqQg6PlCWgI',
+                'ext': 'mp4',
+                
+            }
         }
     ]
 

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index a95d2c942f35dbbe1cd8402698b398b3a887d45a..75f9594e6c8abcecfe128af1be2e73575d6eecc1 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -712,8 +712,10 @@ def date_from_str(date_str):
     Return a datetime object from a string in the format YYYYMMDD or
     (now|today)[+-][0-9](day|week|month|year)(s)?"""
     today = datetime.date.today()
-    if date_str == 'now'or date_str == 'today':
+    if date_str in ('now', 'today'):
         return today
+    if date_str == 'yesterday':
+        return today - datetime.timedelta(days=1)
     match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str)
     if match is not None:
         sign = match.group('sign')