release 2015.02.03.1

720p @ 1000KB/s looks way better than 1080p @ 500KB/s

docs/supportedsites.md

@@ -511,6 +511,7 @@
  - **Wistia**
  - **WorldStarHipHop**
  - **wrzuta.pl**
+ - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
  - **XHamster**

test/helper.py

@@ -103,6 +103,16 @@ def expect_info_dict(self, got_dict, expected_dict):
             self.assertTrue(
                 match_rex.match(got),
                 'field %s (value: %r) should match %r' % (info_field, got, match_str))
+        elif isinstance(expected, compat_str) and expected.startswith('startswith:'):
+            got = got_dict.get(info_field)
+            start_str = expected[len('startswith:'):]
+            self.assertTrue(
+                isinstance(got, compat_str),
+                'Expected a %s object, but got %s for field %s' % (
+                    compat_str.__name__, type(got).__name__, info_field))
+            self.assertTrue(
+                got.startswith(start_str),
+                'field %s (value: %r) should start with %r' % (info_field, got, start_str))
         elif isinstance(expected, type):
             got = got_dict.get(info_field)
             self.assertTrue(isinstance(got, expected),

test/test_utils.py

@@ -156,6 +156,9 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(
             unified_strdate('11/26/2014 11:30:00 AM PST', day_first=False),
             '20141126')
+        self.assertEqual(
+            unified_strdate('2/2/2015 6:47:40 PM', day_first=False),
+            '20150202')
 
     def test_find_xpath_attr(self):
         testxml = '''<root>

youtube_dl/YoutubeDL.py

@@ -964,9 +964,11 @@ class YoutubeDL(object):
             thumbnails.sort(key=lambda t: (
                 t.get('preference'), t.get('width'), t.get('height'),
                 t.get('id'), t.get('url')))
-            for t in thumbnails:
+            for i, t in enumerate(thumbnails):
                 if 'width' in t and 'height' in t:
                     t['resolution'] = '%dx%d' % (t['width'], t['height'])
+                if t.get('id') is None:
+                    t['id'] = '%d' % i
 
         if thumbnails and 'thumbnail' not in info_dict:
             info_dict['thumbnail'] = thumbnails[-1]['url']

youtube_dl/extractor/__init__.py

@@ -554,6 +554,7 @@ from .wimp import WimpIE
 from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
 from .wrzuta import WrzutaIE
+from .wsj import WSJIE
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
 from .xhamster import XHamsterIE

youtube_dl/extractor/aftonbladet.py

@@ -1,8 +1,6 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
@@ -21,9 +19,7 @@ class AftonbladetIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.search(self._VALID_URL, url)
-
-        video_id = mobj.group('video_id')
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
         # find internal video meta data

youtube_dl/extractor/common.py

@@ -145,6 +145,7 @@ class InfoExtractor(object):
     thumbnail:      Full URL to a video thumbnail image.
     description:    Full video description.
     uploader:       Full name of the video uploader.
+    creator:        The main artist who created the video.
     timestamp:      UNIX timestamp of the moment the video became available.
     upload_date:    Video upload date (YYYYMMDD).
                     If not explicitly set, calculated from timestamp.
@@ -704,11 +705,11 @@ class InfoExtractor(object):
                 preference,
                 f.get('language_preference') if f.get('language_preference') is not None else -1,
                 f.get('quality') if f.get('quality') is not None else -1,
-                f.get('height') if f.get('height') is not None else -1,
-                f.get('width') if f.get('width') is not None else -1,
-                ext_preference,
                 f.get('tbr') if f.get('tbr') is not None else -1,
                 f.get('vbr') if f.get('vbr') is not None else -1,
+                ext_preference,
+                f.get('height') if f.get('height') is not None else -1,
+                f.get('width') if f.get('width') is not None else -1,
                 f.get('abr') if f.get('abr') is not None else -1,
                 audio_ext_preference,
                 f.get('fps') if f.get('fps') is not None else -1,

youtube_dl/extractor/franceculture.py

@@ -1,77 +1,69 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import json
-import re
-
 from .common import InfoExtractor
 from ..compat import (
-    compat_parse_qs,
     compat_urlparse,
 )
+from ..utils import (
+    determine_ext,
+    int_or_none,
+)
 
 
 class FranceCultureIE(InfoExtractor):
-    _VALID_URL = r'(?P<baseurl>http://(?:www\.)?franceculture\.fr/)player/reecouter\?play=(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/player/reecouter\?play=(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.franceculture.fr/player/reecouter?play=4795174',
         'info_dict': {
             'id': '4795174',
             'ext': 'mp3',
             'title': 'Rendez-vous au pays des geeks',
+            'alt_title': 'Carnet nomade | 13-14',
             'vcodec': 'none',
-            'uploader': 'Colette Fellous',
             'upload_date': '20140301',
-            'duration': 3601,
             'thumbnail': r're:^http://www\.franceculture\.fr/.*/images/player/Carnet-nomade\.jpg$',
-            'description': 'Avec :Jean-Baptiste Péretié pour son documentaire sur Arte "La revanche des « geeks », une enquête menée aux Etats-Unis dans la S ...',
+            'description': 'startswith:Avec :Jean-Baptiste Péretié pour son documentaire sur Arte "La revanche des « geeks », une enquête menée aux Etats',
+            'timestamp': 1393700400,
         }
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        baseurl = mobj.group('baseurl')
-
+        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
-        params_code = self._search_regex(
-            r"<param name='movie' value='/sites/all/modules/rf/rf_player/swf/loader.swf\?([^']+)' />",
-            webpage, 'parameter code')
-        params = compat_parse_qs(params_code)
-        video_url = compat_urlparse.urljoin(baseurl, params['urlAOD'][0])
+
+        video_path = self._search_regex(
+            r'<a id="player".*?href="([^"]+)"', webpage, 'video path')
+        video_url = compat_urlparse.urljoin(url, video_path)
+        timestamp = int_or_none(self._search_regex(
+            r'<a id="player".*?data-date="([0-9]+)"',
+            webpage, 'upload date', fatal=False))
+        thumbnail = self._search_regex(
+            r'<a id="player".*?>\s+<img src="([^"]+)"',
+            webpage, 'thumbnail', fatal=False)
 
         title = self._html_search_regex(
-            r'<h1 class="title[^"]+">(.+?)</h1>', webpage, 'title')
+            r'<span class="title-diffusion">(.*?)</span>', webpage, 'title')
+        alt_title = self._html_search_regex(
+            r'<span class="title">(.*?)</span>',
+            webpage, 'alt_title', fatal=False)
+        description = self._html_search_regex(
+            r'<span class="description">(.*?)</span>',
+            webpage, 'description', fatal=False)
+
         uploader = self._html_search_regex(
             r'(?s)<div id="emission".*?<span class="author">(.*?)</span>',
-            webpage, 'uploader', fatal=False)
-        thumbnail_part = self._html_search_regex(
-            r'(?s)<div id="emission".*?<img src="([^"]+)"', webpage,
-            'thumbnail', fatal=False)
-        if thumbnail_part is None:
-            thumbnail = None
-        else:
-            thumbnail = compat_urlparse.urljoin(baseurl, thumbnail_part)
-        description = self._html_search_regex(
-            r'(?s)<p class="desc">(.*?)</p>', webpage, 'description')
-
-        info = json.loads(params['infoData'][0])[0]
-        duration = info.get('media_length')
-        upload_date_candidate = info.get('media_section5')
-        upload_date = (
-            upload_date_candidate
-            if (upload_date_candidate is not None and
-                re.match(r'[0-9]{8}$', upload_date_candidate))
-            else None)
+            webpage, 'uploader', default=None)
+        vcodec = 'none' if determine_ext(video_url.lower()) == 'mp3' else None
 
         return {
             'id': video_id,
             'url': video_url,
-            'vcodec': 'none' if video_url.lower().endswith('.mp3') else None,
-            'duration': duration,
+            'vcodec': vcodec,
             'uploader': uploader,
-            'upload_date': upload_date,
+            'timestamp': timestamp,
             'title': title,
+            'alt_title': alt_title,
             'thumbnail': thumbnail,
             'description': description,
         }

youtube_dl/extractor/wsj.py

@@ -0,0 +1,89 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
+
+
+class WSJIE(InfoExtractor):
+    _VALID_URL = r'https?://video-api\.wsj\.com/api-video/player/iframe\.html\?guid=(?P<id>[a-zA-Z0-9-]+)'
+    IE_DESC = 'Wall Street Journal'
+    _TEST = {
+        'url': 'http://video-api.wsj.com/api-video/player/iframe.html?guid=1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
+        'md5': '9747d7a6ebc2f4df64b981e1dde9efa9',
+        'info_dict': {
+            'id': '1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
+            'ext': 'mp4',
+            'upload_date': '20150202',
+            'uploader_id': 'bbright',
+            'creator': 'bbright',
+            'categories': list,  # a long list
+            'duration': 90,
+            'title': 'Bills Coach Rex Ryan Updates His Old Jets Tattoo',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        bitrates = [128, 174, 264, 320, 464, 664, 1264]
+        api_url = (
+            'http://video-api.wsj.com/api-video/find_all_videos.asp?'
+            'type=guid&count=1&query=%s&'
+            'fields=hls,adZone,thumbnailList,guid,state,secondsUntilStartTime,'
+            'author,description,name,linkURL,videoStillURL,duration,videoURL,'
+            'adCategory,catastrophic,linkShortURL,doctypeID,youtubeID,'
+            'titletag,rssURL,wsj-section,wsj-subsection,allthingsd-section,'
+            'allthingsd-subsection,sm-section,sm-subsection,provider,'
+            'formattedCreationDate,keywords,keywordsOmniture,column,editor,'
+            'emailURL,emailPartnerID,showName,omnitureProgramName,'
+            'omnitureVideoFormat,linkRelativeURL,touchCastID,'
+            'omniturePublishDate,%s') % (
+                video_id, ','.join('video%dkMP4Url' % br for br in bitrates))
+        info = self._download_json(api_url, video_id)['items'][0]
+
+        # Thumbnails are conveniently in the correct format already
+        thumbnails = info.get('thumbnailList')
+        creator = info.get('author')
+        uploader_id = info.get('editor')
+        categories = info.get('keywords')
+        duration = int_or_none(info.get('duration'))
+        upload_date = unified_strdate(
+            info.get('formattedCreationDate'), day_first=False)
+        title = info.get('name', info.get('titletag'))
+
+        formats = [{
+            'format_id': 'f4m',
+            'format_note': 'f4m (meta URL)',
+            'url': info['videoURL'],
+        }]
+        if info.get('hls'):
+            formats.extend(self._extract_m3u8_formats(
+                info['hls'], video_id, ext='mp4',
+                preference=0, entry_protocol='m3u8_native'))
+        for br in bitrates:
+            field = 'video%dkMP4Url' % br
+            if info.get(field):
+                formats.append({
+                    'format_id': 'mp4-%d' % br,
+                    'container': 'mp4',
+                    'tbr': br,
+                    'url': info[field],
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'creator': creator,
+            'uploader_id': uploader_id,
+            'duration': duration,
+            'upload_date': upload_date,
+            'title': title,
+            'formats': formats,
+            'categories': categories,
+        }

youtube_dl/utils.py

@@ -701,7 +701,7 @@ def unified_strdate(date_str, day_first=True):
     # %z (UTC offset) is only supported in python>=3.2
     date_str = re.sub(r' ?(\+|-)[0-9]{2}:?[0-9]{2}$', '', date_str)
     # Remove AM/PM + timezone
-    date_str = re.sub(r'(?i)\s*(?:AM|PM)\s+[A-Z]+', '', date_str)
+    date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str)
 
     format_expressions = [
         '%d %B %Y',

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.02.03'
+__version__ = '2015.02.03.1'