Merge tag 'upstream/2014.01.17.2'

[youtubedl] / youtube_dl / extractor / arte.py

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index d39b489518f7bc699854d9ddeb7cd4fa357a6c2b..7cf3785ac687a4fae01a53589af26fb11054463d 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -1,7 +1,8 @@
 # encoding: utf-8
 # encoding: utf-8

+from __future__ import unicode_literals
+

 import re
 import json
 import re
 import json

-import xml.etree.ElementTree

 
 from .common import InfoExtractor
 from ..utils import (
 
 from .common import InfoExtractor
 from ..utils import (


@@ -10,6 +11,8 @@ from ..utils import (
     unified_strdate,
     determine_ext,
     get_element_by_id,
     unified_strdate,
     determine_ext,
     get_element_by_id,

+    compat_str,
+    get_element_by_attribute,

 )
 
 # There are different sources of video in arte.tv, the extraction process 
 )
 
 # There are different sources of video in arte.tv, the extraction process 


@@ -17,11 +20,11 @@ from ..utils import (
 # add tests.
 
 class ArteTvIE(InfoExtractor):
 # add tests.
 
 class ArteTvIE(InfoExtractor):

-    _VIDEOS_URL = r'(?:http://)?videos.arte.tv/(?P<lang>fr|de)/.*-(?P<id>.*?).html'
-    _LIVEWEB_URL = r'(?:http://)?liveweb.arte.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'
+    _VIDEOS_URL = r'(?:http://)?videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
+    _LIVEWEB_URL = r'(?:http://)?liveweb\.arte\.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'

     _LIVE_URL = r'index-[0-9]+\.html$'
 
     _LIVE_URL = r'index-[0-9]+\.html$'
 

-    IE_NAME = u'arte.tv'
+    IE_NAME = 'arte.tv'

 
     @classmethod
     def suitable(cls, url):
 
     @classmethod
     def suitable(cls, url):


@@ -36,7 +39,7 @@ class ArteTvIE(InfoExtractor):
     #         r'src="(.*?/videothek_js.*?\.js)',
     #         0,
     #         [
     #         r'src="(.*?/videothek_js.*?\.js)',
     #         0,
     #         [

-    #             (1, 'url', u'Invalid URL: %s' % url)
+    #             (1, 'url', 'Invalid URL: %s' % url)

     #         ]
     #     )
     #     http_host = url.split('/')[2]
     #         ]
     #     )
     #     http_host = url.split('/')[2]


@@ -48,12 +51,12 @@ class ArteTvIE(InfoExtractor):
     #             '(rtmp://.*?)\'',
     #         re.DOTALL,
     #         [
     #             '(rtmp://.*?)\'',
     #         re.DOTALL,
     #         [

-    #             (1, 'path',   u'could not extract video path: %s' % url),
-    #             (2, 'player', u'could not extract video player: %s' % url),
-    #             (3, 'url',    u'could not extract video url: %s' % url)
+    #             (1, 'path',   'could not extract video path: %s' % url),
+    #             (2, 'player', 'could not extract video player: %s' % url),
+    #             (3, 'url',    'could not extract video url: %s' % url)

     #         ]
     #     )
     #         ]
     #     )

-    #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
+    #     video_url = '%s/%s' % (info.get('url'), info.get('path'))

 
     def _real_extract(self, url):
         mobj = re.match(self._VIDEOS_URL, url)
 
     def _real_extract(self, url):
         mobj = re.match(self._VIDEOS_URL, url)


@@ -68,7 +71,7 @@ class ArteTvIE(InfoExtractor):
             lang = mobj.group('lang')
             return self._extract_liveweb(url, name, lang)
 
             lang = mobj.group('lang')
             return self._extract_liveweb(url, name, lang)
 

-        if re.search(self._LIVE_URL, video_id) is not None:
+        if re.search(self._LIVE_URL, url) is not None:

             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
             # return
             raise ExtractorError(u'Arte live streams are not yet supported, sorry')
             # self.extractLiveStream(url)
             # return


@@ -77,8 +80,7 @@ class ArteTvIE(InfoExtractor):
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
         """Extract from videos.arte.tv"""
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')

-        ref_xml = self._download_webpage(ref_xml_url, video_id, note=u'Downloading metadata')
-        ref_xml_doc = xml.etree.ElementTree.fromstring(ref_xml)
+        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')

         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
         config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')
         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
         config_xml = self._download_webpage(config_xml_url, video_id, note=u'Downloading configuration')


@@ -107,14 +109,13 @@ class ArteTvIE(InfoExtractor):
     def _extract_liveweb(self, url, name, lang):
         """Extract form http://liveweb.arte.tv/"""
         webpage = self._download_webpage(url, name)
     def _extract_liveweb(self, url, name, lang):
         """Extract form http://liveweb.arte.tv/"""
         webpage = self._download_webpage(url, name)

-        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id')
-        config_xml = self._download_webpage('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
-                                            video_id, u'Downloading information')
-        config_doc = xml.etree.ElementTree.fromstring(config_xml.encode('utf-8'))
+        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, 'event id')
+        config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
+                                            video_id, 'Downloading information')

         event_doc = config_doc.find('event')
         url_node = event_doc.find('video').find('urlHd')
         if url_node is None:
         event_doc = config_doc.find('event')
         url_node = event_doc.find('video').find('urlHd')
         if url_node is None:

-            url_node = video_doc.find('urlSd')
+            url_node = event_doc.find('urlSd')

 
         return {'id': video_id,
                 'title': event_doc.find('name%s' % lang.capitalize()).text,
 
         return {'id': video_id,
                 'title': event_doc.find('name%s' % lang.capitalize()).text,


@@ -125,7 +126,7 @@ class ArteTvIE(InfoExtractor):
 
 
 class ArteTVPlus7IE(InfoExtractor):
 
 
 class ArteTVPlus7IE(InfoExtractor):

-    IE_NAME = u'arte.tv:+7'
+    IE_NAME = 'arte.tv:+7'

     _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
 
     @classmethod
     _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
 
     @classmethod


@@ -144,7 +145,9 @@ class ArteTVPlus7IE(InfoExtractor):
 
     def _extract_from_webpage(self, webpage, video_id, lang):
         json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
 
     def _extract_from_webpage(self, webpage, video_id, lang):
         json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')

+        return self._extract_from_json_url(json_url, video_id, lang)

 
 

+    def _extract_from_json_url(self, json_url, video_id, lang):

         json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
         self.report_extraction(video_id)
         info = json.loads(json_info)
         json_info = self._download_webpage(json_url, video_id, 'Downloading info json')
         self.report_extraction(video_id)
         info = json.loads(json_info)


@@ -158,7 +161,9 @@ class ArteTVPlus7IE(InfoExtractor):
             'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
         }
 
             'thumbnail': player_info.get('programImage') or player_info.get('VTU', {}).get('IUR'),
         }
 

-        formats = player_info['VSR'].values()
+        all_formats = player_info['VSR'].values()
+        # Some formats use the m3u8 protocol
+        all_formats = list(filter(lambda f: f.get('videoFormat') != 'M3U8', all_formats))

         def _match_lang(f):
             if f.get('versionCode') is None:
                 return True
         def _match_lang(f):
             if f.get('versionCode') is None:
                 return True


@@ -170,35 +175,50 @@ class ArteTVPlus7IE(InfoExtractor):
             regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
             return any(re.match(r, f['versionCode']) for r in regexes)
         # Some formats may not be in the same language as the url
             regexes = [r'VO?%s' % l, r'VO?.-ST%s' % l]
             return any(re.match(r, f['versionCode']) for r in regexes)
         # Some formats may not be in the same language as the url

-        formats = filter(_match_lang, formats)
-        # Some formats use the m3u8 protocol
-        formats = filter(lambda f: f.get('videoFormat') != 'M3U8', formats)
-        # We order the formats by quality
+        formats = filter(_match_lang, all_formats)

         formats = list(formats) # in python3 filter returns an iterator
         formats = list(formats) # in python3 filter returns an iterator

+        if not formats:
+            # Some videos are only available in the 'Originalversion'
+            # they aren't tagged as being in French or German
+            if all(f['versionCode'] == 'VO' for f in all_formats):
+                formats = all_formats
+            else:
+                raise ExtractorError(u'The formats list is empty')
+

         if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
         if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:

-            sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
+            def sort_key(f):
+                return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])

         else:
         else:

-            sort_key = lambda f: int(f.get('height',-1))
+            def sort_key(f):
+                return (
+                    # Sort first by quality
+                    int(f.get('height',-1)),
+                    int(f.get('bitrate',-1)),
+                    # The original version with subtitles has lower relevance
+                    re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None,
+                    # The version with sourds/mal subtitles has also lower relevance
+                    re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None,
+                )

         formats = sorted(formats, key=sort_key)
         formats = sorted(formats, key=sort_key)

-        # Prefer videos without subtitles in the same language
-        formats = sorted(formats, key=lambda f: re.match(r'VO(F|A)-STM\1', f.get('versionCode', '')) is None)
-        # Pick the best quality

         def _format(format_info):
         def _format(format_info):

-            quality = format_info['quality']
-            m_quality = re.match(r'\w*? - (\d*)p', quality)
-            if m_quality is not None:
-                quality = m_quality.group(1)
+            quality = ''
+            height = format_info.get('height')
+            if height is not None:
+                quality = compat_str(height)
+            bitrate = format_info.get('bitrate')
+            if bitrate is not None:
+                quality += '-%d' % bitrate

             if format_info.get('versionCode') is not None:
             if format_info.get('versionCode') is not None:

-                format_id = u'%s-%s' % (quality, format_info['versionCode'])
+                format_id = '%s-%s' % (quality, format_info['versionCode'])

             else:
                 format_id = quality
             info = {
                 'format_id': format_id,
                 'format_note': format_info.get('versionLibelle'),
                 'width': format_info.get('width'),
             else:
                 format_id = quality
             info = {
                 'format_id': format_id,
                 'format_note': format_info.get('versionLibelle'),
                 'width': format_info.get('width'),

-                'height': format_info.get('height'),
+                'height': height,

             }
             }

-            if format_info['mediaType'] == u'rtmp':
+            if format_info['mediaType'] == 'rtmp':

                 info['url'] = format_info['streamer']
                 info['play_path'] = 'mp4:' + format_info['url']
                 info['ext'] = 'flv'
                 info['url'] = format_info['streamer']
                 info['play_path'] = 'mp4:' + format_info['url']
                 info['ext'] = 'flv'


@@ -213,27 +233,27 @@ class ArteTVPlus7IE(InfoExtractor):
 
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
 
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):

-    IE_NAME = u'arte.tv:creative'
+    IE_NAME = 'arte.tv:creative'

     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
 
     _TEST = {
     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
 
     _TEST = {

-        u'url': u'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
-        u'file': u'050489-002.mp4',
-        u'info_dict': {
-            u'title': u'Agentur Amateur / Agence Amateur #2 : Corporate Design',
+        'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        'file': '050489-002.mp4',
+        'info_dict': {
+            'title': 'Agentur Amateur / Agence Amateur #2 : Corporate Design',

         },
     }
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):
         },
     }
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):

-    IE_NAME = u'arte.tv:future'
+    IE_NAME = 'arte.tv:future'

     _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
 
     _TEST = {
     _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
 
     _TEST = {

-        u'url': u'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
-        u'file': u'050940-003.mp4',
-        u'info_dict': {
-            u'title': u'Les champignons au secours de la planète',
+        'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
+        'file': '050940-003.mp4',
+        'info_dict': {
+            'title': 'Les champignons au secours de la planète',

         },
     }
 
         },
     }
 


@@ -242,3 +262,21 @@ class ArteTVFutureIE(ArteTVPlus7IE):
         webpage = self._download_webpage(url, anchor_id)
         row = get_element_by_id(anchor_id, webpage)
         return self._extract_from_webpage(row, anchor_id, lang)
         webpage = self._download_webpage(url, anchor_id)
         row = get_element_by_id(anchor_id, webpage)
         return self._extract_from_webpage(row, anchor_id, lang)

+
+
+class ArteTVDDCIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:ddc'
+    _VALID_URL = r'http?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
+
+    def _real_extract(self, url):
+        video_id, lang = self._extract_url_info(url)
+        if lang == 'folge':
+            lang = 'de'
+        elif lang == 'emission':
+            lang = 'fr'
+        webpage = self._download_webpage(url, video_id)
+        scriptElement = get_element_by_attribute('class', 'visu_video_block', webpage)
+        script_url = self._html_search_regex(r'src="(.*?)"', scriptElement, 'script url')
+        javascriptPlayerGenerator = self._download_webpage(script_url, video_id, 'Download javascript player generator')
+        json_url = self._search_regex(r"json_url=(.*)&rendering_place.*", javascriptPlayerGenerator, 'json url')
+        return self._extract_from_json_url(json_url, video_id, lang)