release 2014.11.21.1

Update jsinterp.py

youtube_dl/YoutubeDL.py

@@ -624,7 +624,7 @@ class YoutubeDL(object):
 
             return self.process_ie_result(
                 new_result, download=download, extra_info=extra_info)
-        elif result_type == 'playlist':
+        elif result_type == 'playlist' or result_type == 'multi_video':
             # We process each entry in the playlist
             playlist = ie_result.get('title', None) or ie_result.get('id', None)
             self.to_screen('[download] Downloading playlist: %s' % playlist)
@@ -679,6 +679,9 @@ class YoutubeDL(object):
             ie_result['entries'] = playlist_results
             return ie_result
         elif result_type == 'compat_list':
+            self.report_warning(
+                'Extractor %s returned a compat_list result. '
+                'It needs to be updated.' % ie_result.get('extractor'))
             def _fixup(r):
                 self.add_extra_info(r,
                     {

youtube_dl/compat.py

@@ -306,7 +306,7 @@ def workaround_optparse_bug9161():
     og = optparse.OptionGroup(op, 'foo')
     try:
         og.add_option('-t')
-    except TypeError as te:
+    except TypeError:
         real_add_option = optparse.OptionGroup.add_option
 
         def _compat_add_option(self, *args, **kwargs):

youtube_dl/extractor/__init__.py

@@ -115,6 +115,7 @@ from .fktv import (
     FKTVPosteckeIE,
 )
 from .flickr import FlickrIE
+from .folketinget import FolketingetIE
 from .fourtube import FourTubeIE
 from .franceculture import FranceCultureIE
 from .franceinter import FranceInterIE

youtube_dl/extractor/arte.py

@@ -5,7 +5,6 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    ExtractorError,
     find_xpath_attr,
     unified_strdate,
     get_element_by_id,

youtube_dl/extractor/bliptv.py

@@ -166,9 +166,17 @@ class BlipTVIE(SubtitlesInfoExtractor):
 
 
 class BlipTVUserIE(InfoExtractor):
-    _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
+    _VALID_URL = r'(?:(?:https?://(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
     _PAGE_SIZE = 12
     IE_NAME = 'blip.tv:user'
+    _TEST = {
+        'url': 'http://blip.tv/actone',
+        'info_dict': {
+            'id': 'actone',
+            'title': 'Act One: The Series',
+        },
+        'playlist_count': 5,
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -179,6 +187,7 @@ class BlipTVUserIE(InfoExtractor):
         page = self._download_webpage(url, username, 'Downloading user page')
         mobj = re.search(r'data-users-id="([^"]+)"', page)
         page_base = page_base % mobj.group(1)
+        title = self._og_search_title(page)
 
         # Download video ids using BlipTV Ajax calls. Result size per
         # query is limited (currently to 12 videos) so we need to query
@@ -215,4 +224,5 @@ class BlipTVUserIE(InfoExtractor):
 
         urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
-        return [self.playlist_result(url_entries, playlist_title=username)]
+        return self.playlist_result(
+            url_entries, playlist_title=title, playlist_id=username)

youtube_dl/extractor/brightcove.py

@@ -111,6 +111,8 @@ class BrightcoveIE(InfoExtractor):
                             lambda m: m.group(1) + '/>', object_str)
         # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
         object_str = object_str.replace('<--', '<!--')
+        # remove namespace to simplify extraction
+        object_str = re.sub(r'(<object[^>]*)(xmlns=".*?")', r'\1', object_str)
         object_str = fix_xml_ampersands(object_str)
 
         object_doc = xml.etree.ElementTree.fromstring(object_str.encode('utf-8'))
@@ -219,7 +221,7 @@ class BrightcoveIE(InfoExtractor):
         webpage = self._download_webpage(req, video_id)
 
         error_msg = self._html_search_regex(
-            r"<h1>We're sorry.</h1>\s*<p>(.*?)</p>", webpage,
+            r"<h1>We're sorry.</h1>([\s\n]*<p>.*?</p>)+", webpage,
             'error message', default=None)
         if error_msg is not None:
             raise ExtractorError(

youtube_dl/extractor/comedycentral.py

@@ -2,7 +2,6 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
 from .mtv import MTVServicesInfoExtractor
 from ..utils import (
     compat_str,
@@ -110,9 +109,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url, re.VERBOSE)
+        mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
 
         if mobj.group('shortname'):
             if mobj.group('shortname') in ('tds', 'thedailyshow'):

youtube_dl/extractor/common.py

@@ -43,7 +43,11 @@ class InfoExtractor(object):
     information possibly downloading the video to the file system, among
     other possible outcomes.
 
-    The dictionaries must include the following fields:
+    The type field determines the the type of the result.
+    By far the most common value (and the default if _type is missing) is
+    "video", which indicates a single video.
+
+    For a video, the dictionaries must include the following fields:
 
     id:             Video identifier.
     title:          Video title, unescaped.
@@ -151,6 +155,38 @@ class InfoExtractor(object):
 
     Unless mentioned otherwise, None is equivalent to absence of information.
 
+
+    _type "playlist" indicates multiple videos.
+    There must be a key "entries", which is a list or a PagedList object, each
+    element of which is a valid dictionary under this specfication.
+
+    Additionally, playlists can have "title" and "id" attributes with the same
+    semantics as videos (see above).
+
+
+    _type "multi_video" indicates that there are multiple videos that
+    form a single show, for examples multiple acts of an opera or TV episode.
+    It must have an entries key like a playlist and contain all the keys
+    required for a video at the same time.
+
+
+    _type "url" indicates that the video must be extracted from another
+    location, possibly by a different extractor. Its only required key is:
+    "url" - the next URL to extract.
+
+    Additionally, it may have properties believed to be identical to the
+    resolved entity, for example "title" if the title of the referred video is
+    known ahead of time.
+
+
+    _type "url_transparent" entities have the same specification as "url", but
+    indicate that the given additional information is more precise than the one
+    associated with the resolved URL.
+    This is useful when a site employs a video service that hosts the video and
+    its technical metadata, but that video service does not embed a useful
+    title, description etc.
+
+
     Subclasses of this one should re-define the _real_initialize() and
     _real_extract() methods and define a _VALID_URL regexp.
     Probably, they should also be added to the list of extractors.

youtube_dl/extractor/folketinget.py

@@ -0,0 +1,75 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_parse_qs
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+    xpath_text,
+)
+
+
+class FolketingetIE(InfoExtractor):
+    IE_DESC = 'Folketinget (ft.dk; Danish parliament)'
+    _VALID_URL = r'https?://(?:www\.)?ft\.dk/webtv/video/[^?#]*?\.(?P<id>[0-9]+)\.aspx'
+    _TEST = {
+        'url': 'http://www.ft.dk/webtv/video/20141/eru/td.1165642.aspx?as=1#player',
+        'info_dict': {
+            'id': '1165642',
+            'ext': 'mp4',
+            'title': 'Åbent samråd i Erhvervsudvalget',
+            'description': 'Åbent samråd med erhvervs- og vækstministeren om regeringens politik på teleområdet',
+            'view_count': int,
+            'width': 768,
+            'height': 432,
+            'tbr': 928000,
+            'timestamp': 1416493800,
+            'upload_date': '20141120',
+            'duration': 3960,
+        },
+        'params': {
+            'skip_download': 'rtmpdump required',
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._og_search_title(webpage)
+        description = self._html_search_regex(
+            r'(?s)<div class="video-item-agenda"[^>]*>(.*?)<',
+            webpage, 'description', fatal=False)
+
+        player_params = compat_parse_qs(self._search_regex(
+            r'<embed src="http://ft\.arkena\.tv/flash/ftplayer\.swf\?([^"]+)"',
+            webpage, 'player params'))
+        xml_url = player_params['xml'][0]
+        doc = self._download_xml(xml_url, video_id)
+
+        timestamp = parse_iso8601(xpath_text(doc, './/date'))
+        duration = parse_duration(xpath_text(doc, './/duration'))
+        width = int_or_none(xpath_text(doc, './/width'))
+        height = int_or_none(xpath_text(doc, './/height'))
+        view_count = int_or_none(xpath_text(doc, './/views'))
+
+        formats = [{
+            'format_id': n.attrib['bitrate'],
+            'url': xpath_text(n, './url', fatal=True),
+            'tbr': int_or_none(n.attrib['bitrate']),
+        } for n in doc.findall('.//streams/stream')]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'description': description,
+            'timestamp': timestamp,
+            'width': width,
+            'height': height,
+            'duration': duration,
+            'view_count': view_count,
+        }

youtube_dl/extractor/generic.py

@@ -979,7 +979,7 @@ class GenericIE(InfoExtractor):
                 found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage))
         if not found:
             # HTML5 video
-            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src="([^"]+)"', webpage)
+            found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']', webpage)
         if not found:
             found = re.search(
                 r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'

youtube_dl/extractor/goldenmoustache.py

@@ -1,9 +1,7 @@
 from __future__ import unicode_literals
 
-import re
 from .common import InfoExtractor
 from ..utils import (
-    parse_duration,
     int_or_none,
 )
 

youtube_dl/extractor/mtv.py

@@ -145,7 +145,8 @@ class MTVServicesInfoExtractor(InfoExtractor):
         idoc = self._download_xml(
             feed_url + '?' + data, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
-        return [self._get_video_info(item) for item in idoc.findall('.//item')]
+        return self.playlist_result(
+            [self._get_video_info(item) for item in idoc.findall('.//item')])
 
     def _real_extract(self, url):
         title = url_basename(url)

youtube_dl/extractor/stanfordoc.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -9,24 +11,23 @@ from ..utils import (
 
 
 class StanfordOpenClassroomIE(InfoExtractor):
-    IE_NAME = u'stanfordoc'
+    IE_NAME = 'stanfordoc'
-    IE_DESC = u'Stanford Open ClassRoom'
+    IE_DESC = 'Stanford Open ClassRoom'
-    _VALID_URL = r'^(?:https?://)?openclassroom\.stanford\.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
+    _VALID_URL = r'https?://openclassroom\.stanford\.edu(?P<path>/?|(/MainFolder/(?:HomePage|CoursePage|VideoPage)\.php([?]course=(?P<course>[^&]+)(&video=(?P<video>[^&]+))?(&.*)?)?))$'
     _TEST = {
-        u'url': u'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100',
+        'url': 'http://openclassroom.stanford.edu/MainFolder/VideoPage.php?course=PracticalUnix&video=intro-environment&speed=100',
-        u'file': u'PracticalUnix_intro-environment.mp4',
+        'md5': '544a9468546059d4e80d76265b0443b8',
-        u'md5': u'544a9468546059d4e80d76265b0443b8',
+        'info_dict': {
-        u'info_dict': {
+            'id': 'PracticalUnix_intro-environment',
-            u"title": u"Intro Environment"
+            'ext': 'mp4',
+            'title': 'Intro Environment',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
 
-        if mobj.group('course') and mobj.group('video'): # A specific video
+        if mobj.group('course') and mobj.group('video'):  # A specific video
             course = mobj.group('course')
             video = mobj.group('video')
             info = {
@@ -35,7 +36,6 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 'upload_date': None,
             }
 
-            self.report_extraction(info['id'])
             baseUrl = 'http://openclassroom.stanford.edu/MainFolder/courses/' + course + '/videos/'
             xmlUrl = baseUrl + video + '.xml'
             mdoc = self._download_xml(xmlUrl, info['id'])
@@ -43,63 +43,49 @@ class StanfordOpenClassroomIE(InfoExtractor):
                 info['title'] = mdoc.findall('./title')[0].text
                 info['url'] = baseUrl + mdoc.findall('./videoFile')[0].text
             except IndexError:
-                raise ExtractorError(u'Invalid metadata XML file')
+                raise ExtractorError('Invalid metadata XML file')
-            info['ext'] = info['url'].rpartition('.')[2]
+            return info
-            return [info]
+        elif mobj.group('course'):  # A course page
-        elif mobj.group('course'): # A course page
             course = mobj.group('course')
             info = {
                 'id': course,
-                'type': 'playlist',
+                '_type': 'playlist',
                 'uploader': None,
                 'upload_date': None,
             }
 
-            coursepage = self._download_webpage(url, info['id'],
+            coursepage = self._download_webpage(
-                                        note='Downloading course info page',
+                url, info['id'],
-                                        errnote='Unable to download course info page')
+                note='Downloading course info page',
+                errnote='Unable to download course info page')
 
-            info['title'] = self._html_search_regex('<h1>([^<]+)</h1>', coursepage, 'title', default=info['id'])
+            info['title'] = self._html_search_regex(
+                r'<h1>([^<]+)</h1>', coursepage, 'title', default=info['id'])
 
-            info['description'] = self._html_search_regex('<description>([^<]+)</description>',
+            info['description'] = self._html_search_regex(
-                coursepage, u'description', fatal=False)
+                r'(?s)<description>([^<]+)</description>',
+                coursepage, 'description', fatal=False)
 
             links = orderedSet(re.findall('<a href="(VideoPage.php\?[^"]+)">', coursepage))
-            info['list'] = [
+            info['entries'] = [self.url_result(
-                {
+                'http://openclassroom.stanford.edu/MainFolder/%s' % unescapeHTML(l)
-                    'type': 'reference',
+            ) for l in links]
-                    'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(vpage),
+            return info
-                }
+        else:  # Root page
-                    for vpage in links]
-            results = []
-            for entry in info['list']:
-                assert entry['type'] == 'reference'
-                results += self.extract(entry['url'])
-            return results
-        else: # Root page
             info = {
                 'id': 'Stanford OpenClassroom',
-                'type': 'playlist',
+                '_type': 'playlist',
                 'uploader': None,
                 'upload_date': None,
             }
+            info['title'] = info['id']
 
             rootURL = 'http://openclassroom.stanford.edu/MainFolder/HomePage.php'
             rootpage = self._download_webpage(rootURL, info['id'],
-                errnote=u'Unable to download course info page')
+                errnote='Unable to download course info page')
-
-            info['title'] = info['id']
 
             links = orderedSet(re.findall('<a href="(CoursePage.php\?[^"]+)">', rootpage))
-            info['list'] = [
+            info['entries'] = [self.url_result(
-                {
+                'http://openclassroom.stanford.edu/MainFolder/%s' % unescapeHTML(l)
-                    'type': 'reference',
+            ) for l in links]
-                    'url': 'http://openclassroom.stanford.edu/MainFolder/' + unescapeHTML(cpage),
+            return info
-                }
-                    for cpage in links]
-
-            results = []
-            for entry in info['list']:
-                assert entry['type'] == 'reference'
-                results += self.extract(entry['url'])
-            return results

youtube_dl/extractor/sztvhu.py

@@ -1,27 +1,24 @@
 # -*- coding: utf-8 -*-
-
+from __future__ import unicode_literals
-import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 
 class SztvHuIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:(?:www\.)?sztv\.hu|www\.tvszombathely\.hu)/(?:[^/]+)/.+-(?P<id>[0-9]+)'
+    _VALID_URL = r'http://(?:(?:www\.)?sztv\.hu|www\.tvszombathely\.hu)/(?:[^/]+)/.+-(?P<id>[0-9]+)'
     _TEST = {
-        u'url': u'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909',
+        'url': 'http://sztv.hu/hirek/cserkeszek-nepszerusitettek-a-kornyezettudatos-eletmodot-a-savaria-teren-20130909',
-        u'file': u'20130909.mp4',
+        'md5': 'a6df607b11fb07d0e9f2ad94613375cb',
-        u'md5': u'a6df607b11fb07d0e9f2ad94613375cb',
+        'info_dict': {
-        u'info_dict': {
+            'id': '20130909',
-            u"title": u"Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren",
+            'ext': 'mp4',
-            u"description": u'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...',
+            'title': 'Cserkészek népszerűsítették a környezettudatos életmódot a Savaria téren',
+            'description': 'A zöld nap játékos ismeretterjesztő programjait a Magyar Cserkész Szövetség szervezte, akik az ország nyolc városában adják át tudásukat az érdeklődőknek. A PET...',
         },
-        u'skip': u'Service temporarily disabled as of 2013-11-20'
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         video_file = self._search_regex(
             r'file: "...:(.*?)",', webpage, 'video file')
@@ -39,7 +36,6 @@ class SztvHuIE(InfoExtractor):
             'id': video_id,
             'url': video_url,
             'title': title,
-            'ext': determine_ext(video_url),
             'description': description,
             'thumbnail': thumbnail,
         }

youtube_dl/extractor/vh1.py

@@ -121,4 +121,7 @@ class VH1IE(MTVIE):
         idoc = self._download_xml(
             doc_url, video_id,
             'Downloading info', transform_source=fix_xml_ampersands)
-        return [self._get_video_info(item) for item in idoc.findall('.//item')]
+        return self.playlist_result(
+            [self._get_video_info(item) for item in idoc.findall('.//item')],
+            playlist_id=video_id,
+        )

youtube_dl/jsinterp.py

@@ -61,7 +61,7 @@ class JSInterpreter(object):
             pass
 
         m = re.match(
-            r'^(?P<var>[a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
+            r'^(?P<var>[$a-zA-Z0-9_]+)\.(?P<member>[^(]+)(?:\(+(?P<args>[^()]*)\))?$',
             expr)
         if m:
             variable = m.group('var')

youtube_dl/version.py

@@ -1,2 +1,2 @@
 
-__version__ = '2014.11.20.1'
+__version__ = '2014.11.21.1'