Merge remote-tracking branch 'upstream/master'

devscripts/gh-pages/update-sites.py

@@ -0,0 +1,33 @@
+#!/usr/bin/env python3
+
+import sys
+import os
+import textwrap
+
+# We must be able to import youtube_dl
+sys.path.append(os.path.dirname(os.path.dirname(os.path.dirname(os.path.abspath(__file__)))))
+
+import youtube_dl
+
+def main():
+    with open('supportedsites.html.in', 'r', encoding='utf-8') as tmplf:
+        template = tmplf.read()
+
+    ie_htmls = []
+    for ie in sorted(youtube_dl.gen_extractors(), key=lambda i: i.IE_NAME):
+        ie_html = '<b>{}</b>'.format(ie.IE_NAME)
+        try:
+            ie_html += ': {}'.format(ie.IE_DESC)
+        except AttributeError:
+            pass
+        if ie.working() == False:
+            ie_html += ' (Currently broken)'
+        ie_htmls.append('<li>{}</li>'.format(ie_html))
+
+    template = template.replace('@SITES@', textwrap.indent('\n'.join(ie_htmls), '\t'))
+
+    with open('supportedsites.html', 'w', encoding='utf-8') as sitesf:
+        sitesf.write(template)
+
+if __name__ == '__main__':
+    main()

devscripts/release.sh

@@ -85,6 +85,7 @@ ROOT=$(pwd)
     "$ROOT/devscripts/gh-pages/sign-versions.py" < "$ROOT/updates_key.pem"
     "$ROOT/devscripts/gh-pages/generate-download.py"
     "$ROOT/devscripts/gh-pages/update-copyright.py"
+    "$ROOT/devscripts/gh-pages/update-sites.py"
     git add *.html *.html.in update
     git commit -m "release $version"
     git show HEAD

youtube_dl/extractor/__init__.py

@@ -89,6 +89,7 @@
 from .unistra import UnistraIE
 from .ustream import UstreamIE
 from .vbox7 import Vbox7IE
+from .veehd import VeeHDIE
 from .veoh import VeohIE
 from .vevo import VevoIE
 from .videofyme import VideofyMeIE

youtube_dl/extractor/dailymotion.py

@@ -55,7 +55,8 @@ def _real_extract(self, url):
         embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
         embed_page = self._download_webpage(embed_url, video_id,
                                             u'Downloading embed page')
-        info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
+        info = self._search_regex(r'var info = ({.*?}),$', embed_page,
+            'video info', flags=re.MULTILINE)
         info = json.loads(info)
 
         # TODO: support choosing qualities

youtube_dl/extractor/metacafe.py

@@ -122,7 +122,7 @@ def _real_extract(self, url):
         video_title = self._html_search_regex(r'(?im)<title>(.*) - Video</title>', webpage, u'title')
         description = self._og_search_description(webpage)
         video_uploader = self._html_search_regex(
-                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("channel","([^"]+)"\);',
+                r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
                 webpage, u'uploader nickname', fatal=False)
 
         return {

youtube_dl/extractor/orf.py

@@ -14,19 +14,6 @@
 class ORFIE(InfoExtractor):
     _VALID_URL = r'https?://tvthek.orf.at/(programs/.+?/episodes|topics/.+?)/(?P<id>\d+)'
 
-    _TEST = {
-        u'url': u'http://tvthek.orf.at/programs/1171769-Wetter-ZIB/episodes/6557323-Wetter',
-        u'file': u'6566957.flv',
-        u'info_dict': {
-            u'title': u'Wetter',
-            u'description': u'Christa Kummer, Marcus Wadsak und Kollegen  präsentieren abwechselnd ihre täglichen Wetterprognosen für Österreich.\r \r Mehr Wetter unter wetter.ORF.at',
-        },
-        u'params': {
-            # It uses rtmp
-            u'skip_download': True,
-        }
-    }
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         playlist_id = mobj.group('id')

youtube_dl/extractor/veehd.py

@@ -0,0 +1,56 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    get_element_by_id,
+    clean_html,
+)
+
+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',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        player_path = self._search_regex(r'\$\("#playeriframe"\).attr\({src : "(.+?)"',
+            webpage, u'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'])
+        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)
+
+        return {
+            '_type': 'video',
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'ext': 'mp4',
+            'uploader_id': uploader_id,
+            'thumbnail': thumbnail,
+            'description': description,
+        }

youtube_dl/extractor/vimeo.py

@@ -44,6 +44,16 @@ class VimeoIE(InfoExtractor):
                 u'title': u'Andy Allan - Putting the Carto into OpenStreetMap Cartography',
             },
         },
+        {
+            u'url': u'http://player.vimeo.com/video/54469442',
+            u'file': u'54469442.mp4',
+            u'md5': u'619b811a4417aa4abe78dc653becf511',
+            u'note': u'Videos that embed the url in the player page',
+            u'info_dict': {
+                u'title': u'Kathy Sierra: Building the minimum Badass User, Business of Software',
+                u'uploader': u'The BLN & Business of Software',
+            },
+        },
     ]
 
     def _login(self):
@@ -112,7 +122,8 @@ def _real_extract(self, url, new_video=True):
 
         # Extract the config JSON
         try:
-            config = webpage.split(' = {config:')[1].split(',assets:')[0]
+            config = self._search_regex([r' = {config:({.+?}),assets:', r'c=({.+?);'],
+                webpage, u'info section', flags=re.DOTALL)
             config = json.loads(config)
         except:
             if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage):
@@ -132,12 +143,22 @@ def _real_extract(self, url, new_video=True):
         video_uploader_id = config["video"]["owner"]["url"].split('/')[-1] if config["video"]["owner"]["url"] else None
 
         # Extract video thumbnail
-        video_thumbnail = config["video"]["thumbnail"]
+        video_thumbnail = config["video"].get("thumbnail")
+        if video_thumbnail is None:
+            _, video_thumbnail = sorted((int(width), t_url) for (width, t_url) in config["video"]["thumbs"].items())[-1]
 
         # Extract video description
-        video_description = get_element_by_attribute("itemprop", "description", webpage)
-        if video_description: video_description = clean_html(video_description)
-        else: video_description = u''
+        video_description = None
+        try:
+            video_description = get_element_by_attribute("itemprop", "description", webpage)
+            if video_description: video_description = clean_html(video_description)
+        except AssertionError as err:
+            # On some pages like (http://player.vimeo.com/video/54469442) the
+            # html tags are not closed, python 2.6 cannot handle it
+            if err.args[0] == 'we should not get here!':
+                pass
+            else:
+                raise
 
         # Extract upload date
         video_upload_date = None
@@ -154,14 +175,15 @@ def _real_extract(self, url, new_video=True):
         # TODO bind to format param
         codecs = [('h264', 'mp4'), ('vp8', 'flv'), ('vp6', 'flv')]
         files = { 'hd': [], 'sd': [], 'other': []}
+        config_files = config["video"].get("files") or config["request"].get("files")
         for codec_name, codec_extension in codecs:
-            if codec_name in config["video"]["files"]:
-                if 'hd' in config["video"]["files"][codec_name]:
+            if codec_name in config_files:
+                if 'hd' in config_files[codec_name]:
                     files['hd'].append((codec_name, codec_extension, 'hd'))
-                elif 'sd' in config["video"]["files"][codec_name]:
+                elif 'sd' in config_files[codec_name]:
                     files['sd'].append((codec_name, codec_extension, 'sd'))
                 else:
-                    files['other'].append((codec_name, codec_extension, config["video"]["files"][codec_name][0]))
+                    files['other'].append((codec_name, codec_extension, config_files[codec_name][0]))
 
         for quality in ('hd', 'sd', 'other'):
             if len(files[quality]) > 0:
@@ -173,8 +195,12 @@ def _real_extract(self, url, new_video=True):
         else:
             raise ExtractorError(u'No known codec found')
 
-        video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
-                    %(video_id, sig, timestamp, video_quality, video_codec.upper())
+        video_url = None
+        if isinstance(config_files[video_codec], dict):
+            video_url = config_files[video_codec][video_quality].get("url")
+        if video_url is None:
+            video_url = "http://player.vimeo.com/play_redirect?clip_id=%s&sig=%s&time=%s&quality=%s&codecs=%s&type=moogaloop_local&embed_location=" \
+                        %(video_id, sig, timestamp, video_quality, video_codec.upper())
 
         return [{
             'id':       video_id,