Merge branch 'master' into subtitles_rework

6 files changed, 143 insertions, 10 deletions

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index fbe0b8cb7..26cf24935 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -29,6 +29,10 @@ from .escapist import EscapistIE
 from .exfm import ExfmIE
 from .facebook import FacebookIE
 from .flickr import FlickrIE
+from .francetv import (
+    PluzzIE,
+    FranceTvInfoIE,
+)
 from .freesound import FreesoundIE
 from .funnyordie import FunnyOrDieIE
 from .gamespot import GameSpotIE
@@ -75,6 +79,7 @@ from .roxwel import RoxwelIE
 from .rtlnow import RTLnowIE
 from .sina import SinaIE
 from .slashdot import SlashdotIE
+from .slideshare import SlideshareIE
 from .sohu import SohuIE
 from .soundcloud import SoundcloudIE, SoundcloudSetIE
 from .spiegel import SpiegelIE
diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py
index 50832217a..e7f4fa9fd 100644
--- a/youtube_dl/extractor/canalc2.py
+++ b/youtube_dl/extractor/canalc2.py
@@ -5,7 +5,7 @@ from .common import InfoExtractor
 
 
 class Canalc2IE(InfoExtractor):
-    _IE_NAME = 'canalc2.tv'
+    IE_NAME = 'canalc2.tv'
     _VALID_URL = r'http://.*?\.canalc2\.tv/video\.asp\?idVideo=(\d+)&voir=oui'
 
     _TEST = {
diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py
new file mode 100644
index 000000000..f2b12c884
--- /dev/null
+++ b/youtube_dl/extractor/francetv.py
@@ -0,0 +1,77 @@
+# encoding: utf-8
+import re
+import xml.etree.ElementTree
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+)
+
+
+class FranceTVBaseInfoExtractor(InfoExtractor):
+    def _extract_video(self, video_id):
+        xml_desc = self._download_webpage(
+            'http://www.francetvinfo.fr/appftv/webservices/video/'
+            'getInfosOeuvre.php?id-diffusion='
+            + video_id, video_id, 'Downloading XML config')
+        info = xml.etree.ElementTree.fromstring(xml_desc.encode('utf-8'))
+
+        manifest_url = info.find('videos/video/url').text
+        video_url = manifest_url.replace('manifest.f4m', 'index_2_av.m3u8')
+        video_url = video_url.replace('/z/', '/i/')
+        thumbnail_path = info.find('image').text
+
+        return {'id': video_id,
+                'ext': 'mp4',
+                'url': video_url,
+                'title': info.find('titre').text,
+                'thumbnail': compat_urlparse.urljoin('http://pluzz.francetv.fr', thumbnail_path),
+                'description': info.find('synopsis').text,
+                }
+
+
+class PluzzIE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'pluzz.francetv.fr'
+    _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'
+
+    _TEST = {
+        u'url': u'http://pluzz.francetv.fr/videos/allo_rufo_saison5_,88439064.html',
+        u'file': u'88439064.mp4',
+        u'info_dict': {
+            u'title': u'Allô Rufo',
+            u'description': u'md5:d909f1ebdf963814b65772aea250400e',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        title = re.match(self._VALID_URL, url).group(1)
+        webpage = self._download_webpage(url, title)
+        video_id = self._search_regex(
+            r'data-diffusion="(\d+)"', webpage, 'ID')
+        return self._extract_video(video_id)
+
+
+class FranceTvInfoIE(FranceTVBaseInfoExtractor):
+    IE_NAME = u'francetvinfo.fr'
+    _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+).html'
+
+    _TEST = {
+        u'url': u'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
+        u'file': u'84981923.mp4',
+        u'info_dict': {
+            u'title': u'Soir 3',
+        },
+        u'params': {
+            u'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        page_title = mobj.group('title')
+        webpage = self._download_webpage(url, page_title)
+        video_id = self._search_regex(r'id-video=(\d+?)"', webpage, u'video id')
+        return self._extract_video(video_id)
diff --git a/youtube_dl/extractor/slideshare.py b/youtube_dl/extractor/slideshare.py
new file mode 100644
index 000000000..afc3001b5
--- /dev/null
+++ b/youtube_dl/extractor/slideshare.py
@@ -0,0 +1,47 @@
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urlparse,
+    ExtractorError,
+)
+
+
+class SlideshareIE(InfoExtractor):
+    _VALID_URL = r'https?://www\.slideshare\.net/[^/]+?/(?P<title>.+?)($|\?)'
+
+    _TEST = {
+        u'url': u'http://www.slideshare.net/Dataversity/keynote-presentation-managing-scale-and-complexity',
+        u'file': u'25665706.mp4',
+        u'info_dict': {
+            u'title': u'Managing Scale and Complexity',
+            u'description': u'This was a keynote presentation at the NoSQL Now! 2013 Conference & Expo (http://www.nosqlnow.com). This presentation was given by Adrian Cockcroft from Netflix',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        page_title = mobj.group('title')
+        webpage = self._download_webpage(url, page_title)
+        slideshare_obj = self._search_regex(
+            r'var slideshare_object =  ({.*?}); var user_info =',
+            webpage, u'slideshare object')
+        info = json.loads(slideshare_obj)
+        if info['slideshow']['type'] != u'video':
+            raise ExtractorError(u'Webpage type is "%s": only video extraction is supported for Slideshare' % info['slideshow']['type'], expected=True)
+
+        doc = info['doc']
+        bucket = info['jsplayer']['video_bucket']
+        ext = info['jsplayer']['video_extension']
+        video_url = compat_urlparse.urljoin(bucket, doc + '-SD.' + ext)
+
+        return {
+            '_type': 'video',
+            'id': info['slideshow']['id'],
+            'title': info['slideshow']['title'],
+            'ext': ext,
+            'url': video_url,
+            'thumbnail': info['slideshow']['pin_image_url'],
+            'description': self._og_search_description(webpage),
+        }
diff --git a/youtube_dl/extractor/sohu.py b/youtube_dl/extractor/sohu.py
index 77bb0a8dc..2b9bf0cb7 100644
--- a/youtube_dl/extractor/sohu.py
+++ b/youtube_dl/extractor/sohu.py
@@ -8,7 +8,7 @@ from ..utils import ExtractorError
 
 
 class SohuIE(InfoExtractor):
-    _VALID_URL = r'https?://tv\.sohu\.com/\d+?/n(?P<id>\d+)\.shtml.*?'
+    _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
 
     _TEST = {
         u'url': u'http://tv.sohu.com/20130724/n382479172.shtml#super',
@@ -21,8 +21,11 @@ class SohuIE(InfoExtractor):
 
     def _real_extract(self, url):
 
-        def _fetch_data(vid_id):
-            base_data_url = u'http://hot.vrs.sohu.com/vrs_flash.action?vid='
+        def _fetch_data(vid_id, mytv=False):
+            if mytv:
+                base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
+            else:
+                base_data_url = u'http://hot.vrs.sohu.com/vrs_flash.action?vid='
             data_url = base_data_url + str(vid_id)
             data_json = self._download_webpage(
                 data_url, video_id,
@@ -31,15 +34,16 @@ class SohuIE(InfoExtractor):
 
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        mytv = mobj.group('mytv') is not None
 
         webpage = self._download_webpage(url, video_id)
         raw_title = self._html_search_regex(r'(?s)<title>(.+?)</title>',
                                             webpage, u'video title')
         title = raw_title.partition('-')[0].strip()
 
-        vid = self._html_search_regex(r'var vid="(\d+)"', webpage,
+        vid = self._html_search_regex(r'var vid ?= ?["\'](\d+)["\']', webpage,
                                       u'video path')
-        data = _fetch_data(vid)
+        data = _fetch_data(vid, mytv)
 
         QUALITIES = ('ori', 'super', 'high', 'nor')
         vid_ids = [data['data'][q + 'Vid']
@@ -51,7 +55,7 @@ class SohuIE(InfoExtractor):
         # For now, we just pick the highest available quality
         vid_id = vid_ids[-1]
 
-        format_data = data if vid == vid_id else _fetch_data(vid_id)
+        format_data = data if vid == vid_id else _fetch_data(vid_id, mytv)
         part_count = format_data['data']['totalBlocks']
         allot = format_data['allot']
         prot = format_data['prot']
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 11611f10d..e71cd62ec 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -209,7 +209,7 @@ class YoutubeIE(YoutubeSubtitlesIE, YoutubeBaseInfoExtractor):
                          |youtu\.be/                                          # just youtu.be/xxxx
                          )
                      )?                                                       # all until now is optional -> you can pass the naked ID
-                     ([0-9A-Za-z_-]+)                                         # here is it! the YouTube video ID
+                     ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
                      (?(1).+)?                                                # if we found the ID, everything can follow
                      $"""
     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
@@ -484,7 +484,7 @@ class YoutubeIE(YoutubeSubtitlesIE, YoutubeBaseInfoExtractor):
         elif len(s) == 86:
             return s[5:34] + s[0] + s[35:38] + s[3] + s[39:45] + s[38] + s[46:53] + s[73] + s[54:73] + s[85] + s[74:85] + s[53]
         elif len(s) == 85:
-            return s[83:34:-1] + s[0] + s[33:27:-1] + s[3] + s[26:19:-1] + s[34] + s[18:3:-1] + s[27]
+            return s[40] + s[82:43:-1] + s[22] + s[42:40:-1] + s[83] + s[39:22:-1] + s[0] + s[21:2:-1]
         elif len(s) == 84:
             return s[81:36:-1] + s[0] + s[35:2:-1]
         elif len(s) == 83:
@@ -583,7 +583,7 @@ class YoutubeIE(YoutubeSubtitlesIE, YoutubeBaseInfoExtractor):
         manifest = self._download_webpage(manifest_url, video_id, u'Downloading formats manifest')
         formats_urls = _get_urls(manifest)
         for format_url in formats_urls:
-            itag = self._search_regex(r'itag%3D(\d+?)/', format_url, 'itag')
+            itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
             url_map[itag] = format_url
         return url_map