release 2015.02.10

For now, we just wrap some common errors. More may follow. We do not want to catch actual programming errors in the extractors, such as 1 // 0.

docs/supportedsites.md

@@ -392,6 +392,7 @@
  - **StreamCZ**
  - **StreetVoice**
  - **SunPorno**
+ - **SVTPlay**
  - **SWRMediathek**
  - **Syfy**
  - **SztvHu**

youtube_dl/YoutubeDL.py

@@ -1546,7 +1546,6 @@ class YoutubeDL(object):
             line(f, idlen) for f in formats
             if f.get('preference') is None or f['preference'] >= -1000]
         if len(formats) > 1:
-            formats_s[0] += (' ' if self._format_note(formats[0]) else '') + '(worst)'
             formats_s[-1] += (' ' if self._format_note(formats[-1]) else '') + '(best)'
 
         header_line = line({

youtube_dl/extractor/__init__.py

@@ -428,6 +428,7 @@ from .streamcloud import StreamcloudIE
 from .streamcz import StreamCZIE
 from .streetvoice import StreetVoiceIE
 from .sunporno import SunPornoIE
+from .svtplay import SVTPlayIE
 from .swrmediathek import SWRMediathekIE
 from .syfy import SyfyIE
 from .sztvhu import SztvHuIE

youtube_dl/extractor/bandcamp.py

@@ -72,26 +72,29 @@ class BandcampIE(InfoExtractor):
 
         download_link = m_download.group(1)
         video_id = self._search_regex(
-            r'var TralbumData = {.*?id: (?P<id>\d+),?$',
-            webpage, 'video id', flags=re.MULTILINE | re.DOTALL)
+            r'(?ms)var TralbumData = {.*?id: (?P<id>\d+),?$',
+            webpage, 'video id')
 
         download_webpage = self._download_webpage(download_link, video_id, 'Downloading free downloads page')
         # We get the dictionary of the track from some javascript code
-        info = re.search(r'items: (.*?),$', download_webpage, re.MULTILINE).group(1)
-        info = json.loads(info)[0]
+        all_info = self._parse_json(self._search_regex(
+            r'(?sm)items: (.*?),$', download_webpage, 'items'), video_id)
+        info = All_info[0]
         # We pick mp3-320 for now, until format selection can be easily implemented.
         mp3_info = info['downloads']['mp3-320']
         # If we try to use this url it says the link has expired
         initial_url = mp3_info['url']
-        re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
-        m_url = re.match(re_url, initial_url)
+        m_url = re.match(
+            r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$',
+            initial_url)
         # We build the url we will use to get the final track url
         # This url is build in Bandcamp in the script download_bunde_*.js
         request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
         final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
         # If we could correctly generate the .rand field the url would be
         # in the "download_url" key
-        final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
+        final_url = self._search_regex(
+            r'"retry_url":"(.*?)"', final_url_webpage, 'final video URL')
 
         return {
             'id': video_id,

youtube_dl/extractor/common.py

@@ -264,8 +264,15 @@ class InfoExtractor(object):
 
     def extract(self, url):
         """Extracts URL information and returns it in list of dicts."""
-        self.initialize()
-        return self._real_extract(url)
+        try:
+            self.initialize()
+            return self._real_extract(url)
+        except ExtractorError:
+            raise
+        except compat_http_client.IncompleteRead as e:
+            raise ExtractorError('A network error has occured.', cause=e, expected=True)
+        except (KeyError,) as e:
+            raise ExtractorError('An extractor error has occured.', cause=e)
 
     def set_downloader(self, downloader):
         """Sets the downloader for this IE."""

youtube_dl/extractor/generic.py

@@ -524,6 +524,19 @@ class GenericIE(InfoExtractor):
                 'upload_date': '20150126',
             },
             'add_ie': ['Viddler'],
+        },
+        # jwplayer YouTube
+        {
+            'url': 'http://media.nationalarchives.gov.uk/index.php/webinar-using-discovery-national-archives-online-catalogue/',
+            'info_dict': {
+                'id': 'Mrj4DVp2zeA',
+                'ext': 'mp4',
+                'upload_date': '20150204',
+                'uploader': 'The National Archives UK',
+                'description': 'md5:a236581cd2449dd2df4f93412f3f01c6',
+                'uploader_id': 'NationalArchives08',
+                'title': 'Webinar: Using Discovery, The National Archives’ online catalogue',
+            },
         }
     ]
 
@@ -1034,7 +1047,12 @@ class GenericIE(InfoExtractor):
 
         # Look for embedded sbs.com.au player
         mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)sbs\.com\.au/ondemand/video/single/.+?)\1',
+            r'''(?x)
+            (?:
+                <meta\s+property="og:video"\s+content=|
+                <iframe[^>]+?src=
+            )
+            (["\'])(?P<url>https?://(?:www\.)?sbs\.com\.au/ondemand/video/.+?)\1''',
             webpage)
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'SBS')
@@ -1065,6 +1083,8 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Livestream')
 
         def check_video(vurl):
+            if YoutubeIE.suitable(vurl):
+                return True
             vpath = compat_urlparse.urlparse(vurl).path
             vext = determine_ext(vpath)
             return '.' in vpath and vext not in ('swf', 'png', 'jpg', 'srt', 'sbv', 'sub', 'vtt', 'ttml')
@@ -1082,7 +1102,8 @@ class GenericIE(InfoExtractor):
                     JWPlayerOptions|
                     jwplayer\s*\(\s*["'][^'"]+["']\s*\)\s*\.setup
                 )
-                .*?file\s*:\s*["\'](.*?)["\']''', webpage))
+                .*?
+                ['"]?file['"]?\s*:\s*["\'](.*?)["\']''', webpage))
         if not found:
             # Broaden the search a little bit
             found = filter_video(re.findall(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage))

youtube_dl/extractor/rtlnow.py

@@ -91,6 +91,15 @@ class RTLnowIE(InfoExtractor):
             },
         },
         {
+            'url': 'http://rtl-now.rtl.de/der-bachelor/folge-4.php?film_id=188729&player=1&season=5',
+            'info_dict': {
+                'id': '188729',
+                'ext': 'flv',
+                'upload_date': '20150204',
+                'description': 'md5:5e1ce23095e61a79c166d134b683cecc',
+                'title': 'Der Bachelor - Folge 4',
+            }
+        }, {
             'url': 'http://www.n-tvnow.de/deluxe-alles-was-spass-macht/thema-ua-luxushotel-fuer-vierbeiner.php?container_id=153819&player=1&season=0',
             'only_matching': True,
         },
@@ -134,9 +143,18 @@ class RTLnowIE(InfoExtractor):
                     'player_url': video_page_url + 'includes/vodplayer.swf',
                 }
             else:
-                fmt = {
-                    'url': filename.text,
-                }
+                mobj = re.search(r'.*/(?P<hoster>[^/]+)/videos/(?P<play_path>.+)\.f4m', filename.text)
+                if mobj:
+                    fmt = {
+                        'url': 'rtmpe://fmspay-fra2.rtl.de/' + mobj.group('hoster'),
+                        'play_path': 'mp4:' + mobj.group('play_path'),
+                        'page_url': url,
+                        'player_url': video_page_url + 'includes/vodplayer.swf',
+                    }
+                else:
+                    fmt = {
+                        'url': filename.text,
+                    }
             fmt.update({
                 'width': int_or_none(filename.get('width')),
                 'height': int_or_none(filename.get('height')),

youtube_dl/extractor/svtplay.py

@@ -0,0 +1,56 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+)
+
+
+class SVTPlayIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?svtplay\.se/video/(?P<id>[0-9]+)'
+    _TEST = {
+        'url': 'http://www.svtplay.se/video/2609989/sm-veckan/sm-veckan-rally-final-sasong-1-sm-veckan-rally-final',
+        'md5': 'f4a184968bc9c802a9b41316657aaa80',
+        'info_dict': {
+            'id': '2609989',
+            'ext': 'mp4',
+            'title': 'SM veckan vinter, Örebro - Rally, final',
+            'duration': 4500,
+            'thumbnail': 're:^https?://.*[\.-]jpg$',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        info = self._download_json(
+            'http://www.svtplay.se/video/%s?output=json' % video_id, video_id)
+
+        title = info['context']['title']
+        thumbnail = info['context'].get('thumbnailImage')
+
+        video_info = info['video']
+        formats = []
+        for vr in video_info['videoReferences']:
+            vurl = vr['url']
+            if determine_ext(vurl) == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    vurl, video_id,
+                    ext='mp4', entry_protocol='m3u8_native',
+                    m3u8_id=vr.get('playerType')))
+            else:
+                formats.append({
+                    'format_id': vr.get('playerType'),
+                    'url': vurl,
+                })
+        self._sort_formats(formats)
+
+        duration = video_info.get('materialLength')
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': thumbnail,
+            'duration': duration,
+        }

youtube_dl/extractor/trilulilu.py

@@ -1,40 +1,55 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
-import json
+import re
 
 from .common import InfoExtractor
+from ..utils import ExtractorError
 
 
 class TriluliluIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?trilulilu\.ro/video-[^/]+/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?trilulilu\.ro/(?:video-[^/]+/)?(?P<id>[^/#\?]+)'
     _TEST = {
         'url': 'http://www.trilulilu.ro/video-animatie/big-buck-bunny-1',
+        'md5': 'c1450a00da251e2769b74b9005601cac',
         'info_dict': {
-            'id': 'big-buck-bunny-1',
+            'id': 'ae2899e124140b',
             'ext': 'mp4',
             'title': 'Big Buck Bunny',
             'description': ':) pentru copilul din noi',
         },
-        # Server ignores Range headers (--test)
-        'params': {
-            'skip_download': True
-        }
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
 
+        if re.search(r'Fişierul nu este disponibil pentru vizionare în ţara dumneavoastră', webpage):
+            raise ExtractorError(
+                'This video is not available in your country.', expected=True)
+        elif re.search('Fişierul poate fi accesat doar de către prietenii lui', webpage):
+            raise ExtractorError('This video is private.', expected=True)
+
+        flashvars_str = self._search_regex(
+            r'block_flash_vars\s*=\s*(\{[^\}]+\})', webpage, 'flashvars', fatal=False, default=None)
+
+        if flashvars_str:
+            flashvars = self._parse_json(flashvars_str, display_id)
+        else:
+            raise ExtractorError(
+                'This page does not contain videos', expected=True)
+
+        if flashvars['isMP3'] == 'true':
+            raise ExtractorError(
+                'Audio downloads are currently not supported', expected=True)
+
+        video_id = flashvars['hash']
         title = self._og_search_title(webpage)
         thumbnail = self._og_search_thumbnail(webpage)
-        description = self._og_search_description(webpage)
-
-        log_str = self._search_regex(
-            r'block_flash_vars[ ]=[ ]({[^}]+})', webpage, 'log info')
-        log = json.loads(log_str)
+        description = self._og_search_description(webpage, default=None)
 
         format_url = ('http://fs%(server)s.trilulilu.ro/%(hash)s/'
-                      'video-formats2' % log)
+                      'video-formats2' % flashvars)
         format_doc = self._download_xml(
             format_url, video_id,
             note='Downloading formats',
@@ -44,10 +59,10 @@ class TriluliluIE(InfoExtractor):
             'http://fs%(server)s.trilulilu.ro/stream.php?type=video'
             '&source=site&hash=%(hash)s&username=%(userid)s&'
             'key=ministhebest&format=%%s&sig=&exp=' %
-            log)
+            flashvars)
         formats = [
             {
-                'format': fnode.text,
+                'format_id': fnode.text.partition('-')[2],
                 'url': video_url_template % fnode.text,
                 'ext': fnode.text.partition('-')[0]
             }
@@ -56,8 +71,8 @@ class TriluliluIE(InfoExtractor):
         ]
 
         return {
-            '_type': 'video',
             'id': video_id,
+            'display_id': display_id,
             'formats': formats,
             'title': title,
             'description': description,

youtube_dl/extractor/youtube.py

@@ -780,8 +780,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
                     fo for fo in formats
                     if fo['format_id'] == format_id)
             except StopIteration:
-                f.update(self._formats.get(format_id, {}).items())
-                formats.append(f)
+                full_info = self._formats.get(format_id, {}).copy()
+                full_info.update(f)
+                formats.append(full_info)
             else:
                 existing_format.update(f)
         return formats

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.02.09'
+__version__ = '2015.02.10'