debian/changelog: Finalize release.

[youtubedl] / youtube_dl / extractor / arte.py

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index 9254fbfe0de5cb9138cb50deeb4719f94c18f92e..9591bad8a66254e90247a204b43b10b6db4f6406 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -1,6 +1,7 @@
 # encoding: utf-8
 # encoding: utf-8

+from __future__ import unicode_literals
+

 import re
 import re

-import json

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


@@ -17,115 +18,49 @@ from ..utils import (
 # is different for each one. The videos usually expire in 7 days, so we can't
 # add tests.
 
 # is different for each one. The videos usually expire in 7 days, so we can't
 # 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>.+)'
-    _LIVE_URL = r'index-[0-9]+\.html$'
-
-    IE_NAME = u'arte.tv'

 
 

-    @classmethod
-    def suitable(cls, url):
-        return any(re.match(regex, url) for regex in (cls._VIDEOS_URL, cls._LIVEWEB_URL))
-
-    # TODO implement Live Stream
-    # from ..utils import compat_urllib_parse
-    # def extractLiveStream(self, url):
-    #     video_lang = url.split('/')[-4]
-    #     info = self.grep_webpage(
-    #         url,
-    #         r'src="(.*?/videothek_js.*?\.js)',
-    #         0,
-    #         [
-    #             (1, 'url', u'Invalid URL: %s' % url)
-    #         ]
-    #     )
-    #     http_host = url.split('/')[2]
-    #     next_url = 'http://%s%s' % (http_host, compat_urllib_parse.unquote(info.get('url')))
-    #     info = self.grep_webpage(
-    #         next_url,
-    #         r'(s_artestras_scst_geoFRDE_' + video_lang + '.*?)\'.*?' +
-    #             '(http://.*?\.swf).*?' +
-    #             '(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)
-    #         ]
-    #     )
-    #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
+class ArteTvIE(InfoExtractor):
+    _VALID_URL = r'http://videos\.arte\.tv/(?P<lang>fr|de)/.*-(?P<id>.*?)\.html'
+    IE_NAME = 'arte.tv'

 
     def _real_extract(self, url):
 
     def _real_extract(self, url):

-        mobj = re.match(self._VIDEOS_URL, url)
-        if mobj is not None:
-            id = mobj.group('id')
-            lang = mobj.group('lang')
-            return self._extract_video(url, id, lang)
-
-        mobj = re.match(self._LIVEWEB_URL, url)
-        if mobj is not None:
-            name = mobj.group('name')
-            lang = mobj.group('lang')
-            return self._extract_liveweb(url, name, lang)
-
-        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
-
-    def _extract_video(self, url, video_id, lang):
-        """Extract from videos.arte.tv"""
+        mobj = re.match(self._VALID_URL, url)
+        lang = mobj.group('lang')
+        video_id = mobj.group('id')
+

         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')
         ref_xml_url = url.replace('/videos/', '/do_delegate/videos/')
         ref_xml_url = ref_xml_url.replace('.html', ',view,asPlayerXml.xml')

-        ref_xml_doc = self._download_xml(ref_xml_url, video_id, note=u'Downloading metadata')
+        ref_xml_doc = self._download_xml(
+            ref_xml_url, video_id, note='Downloading metadata')

         config_node = find_xpath_attr(ref_xml_doc, './/video', 'lang', lang)
         config_xml_url = config_node.attrib['ref']
         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')
-
-        video_urls = list(re.finditer(r'<url quality="(?P<quality>.*?)">(?P<url>.*?)</url>', config_xml))
-        def _key(m):
-            quality = m.group('quality')
-            if quality == 'hd':
-                return 2
-            else:
-                return 1
-        # We pick the best quality
-        video_urls = sorted(video_urls, key=_key)
-        video_url = list(video_urls)[-1].group('url')
-        
-        title = self._html_search_regex(r'<name>(.*?)</name>', config_xml, 'title')
-        thumbnail = self._html_search_regex(r'<firstThumbnailUrl>(.*?)</firstThumbnailUrl>',
-                                            config_xml, 'thumbnail')
-        return {'id': video_id,
-                'title': title,
-                'thumbnail': thumbnail,
-                'url': video_url,
-                'ext': 'flv',
-                }
-
-    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_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
-                                            video_id, u'Downloading information')
-        event_doc = config_doc.find('event')
-        url_node = event_doc.find('video').find('urlHd')
-        if url_node is None:
-            url_node = event_doc.find('urlSd')
-
-        return {'id': video_id,
-                'title': event_doc.find('name%s' % lang.capitalize()).text,
-                'url': url_node.text.replace('MP4', 'mp4'),
-                'ext': 'flv',
-                'thumbnail': self._og_search_thumbnail(webpage),
-                }
+        config = self._download_xml(
+            config_xml_url, video_id, note='Downloading configuration')
+
+        formats = [{
+            'forma_id': q.attrib['quality'],
+            # The playpath starts at 'mp4:', if we don't manually
+            # split the url, rtmpdump will incorrectly parse them
+            'url': q.text.split('mp4:', 1)[0],
+            'play_path': 'mp4:' + q.text.split('mp4:', 1)[1],
+            'ext': 'flv',
+            'quality': 2 if q.attrib['quality'] == 'hd' else 1,
+        } for q in config.findall('./urls/url')]
+        self._sort_formats(formats)
+
+        title = config.find('.//name').text
+        thumbnail = config.find('.//firstThumbnailUrl').text
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }

 
 
 class ArteTVPlus7IE(InfoExtractor):
 
 
 class ArteTVPlus7IE(InfoExtractor):

-    IE_NAME = u'arte.tv:+7'
-    _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
+    IE_NAME = 'arte.tv:+7'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'

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


@@ -142,13 +77,12 @@ class ArteTVPlus7IE(InfoExtractor):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):
         return self._extract_from_webpage(webpage, video_id, lang)
 
     def _extract_from_webpage(self, webpage, video_id, lang):

-        json_url = self._html_search_regex(r'arte_vp_url="(.*?)"', webpage, 'json url')
+        json_url = self._html_search_regex(
+            r'arte_vp_url="(.*?)"', webpage, 'json vp url')

         return self._extract_from_json_url(json_url, video_id, lang)
 
     def _extract_from_json_url(self, json_url, video_id, lang):
         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)
+        info = self._download_json(json_url, video_id)

         player_info = info['videoJsonPlayer']
 
         info_dict = {
         player_info = info['videoJsonPlayer']
 
         info_dict = {


@@ -170,6 +104,8 @@ class ArteTVPlus7IE(InfoExtractor):
                 l = 'F'
             elif lang == 'de':
                 l = 'A'
                 l = 'F'
             elif lang == 'de':
                 l = 'A'

+            else:
+                l = lang

             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


@@ -178,7 +114,7 @@ class ArteTVPlus7IE(InfoExtractor):
         if not formats:
             # Some videos are only available in the 'Originalversion'
             # they aren't tagged as being in French or German
         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):
+            if all(f['versionCode'] == 'VO' or f['versionCode'] == 'VA' for f in all_formats):

                 formats = all_formats
             else:
                 raise ExtractorError(u'The formats list is empty')
                 formats = all_formats
             else:
                 raise ExtractorError(u'The formats list is empty')


@@ -188,14 +124,19 @@ class ArteTVPlus7IE(InfoExtractor):
                 return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
         else:
             def sort_key(f):
                 return ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
         else:
             def sort_key(f):

+                versionCode = f.get('versionCode')
+                if versionCode is None:
+                    versionCode = ''

                 return (
                     # Sort first by quality
                 return (
                     # Sort first by quality

-                    int(f.get('height',-1)),
-                    int(f.get('bitrate',-1)),
+                    int(f.get('height', -1)),
+                    int(f.get('bitrate', -1)),

                     # The original version with subtitles has lower relevance
                     # The original version with subtitles has lower relevance

-                    re.match(r'VO-ST(F|A)', f.get('versionCode', '')) is None,
+                    re.match(r'VO-ST(F|A)', versionCode) is None,

                     # The version with sourds/mal subtitles has also lower relevance
                     # The version with sourds/mal subtitles has also lower relevance

-                    re.match(r'VO?(F|A)-STM\1', f.get('versionCode', '')) is None,
+                    re.match(r'VO?(F|A)-STM\1', versionCode) is None,
+                    # Prefer http downloads over m3u8
+                    0 if f['url'].endswith('m3u8') else 1,

                 )
         formats = sorted(formats, key=sort_key)
         def _format(format_info):
                 )
         formats = sorted(formats, key=sort_key)
         def _format(format_info):


@@ -207,7 +148,7 @@ class ArteTVPlus7IE(InfoExtractor):
             if bitrate is not None:
                 quality += '-%d' % bitrate
             if format_info.get('versionCode') is not None:
             if bitrate is not None:
                 quality += '-%d' % bitrate
             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 = {
             else:
                 format_id = quality
             info = {


@@ -216,7 +157,7 @@ class ArteTVPlus7IE(InfoExtractor):
                 'width': format_info.get('width'),
                 'height': height,
             }
                 'width': format_info.get('width'),
                 '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'


@@ -231,27 +172,30 @@ 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',
+        'info_dict': {
+            'id': '050489-002',
+            'ext': 'mp4',
+            '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',
+        'info_dict': {
+            'id': '5201',
+            'ext': 'mp4',
+            'title': 'Les champignons au secours de la planète',
+            'upload_date': '20131101',

         },
     }
 
         },
     }
 


@@ -263,8 +207,8 @@ class ArteTVFutureIE(ArteTVPlus7IE):
 
 
 class ArteTVDDCIE(ArteTVPlus7IE):
 
 
 class ArteTVDDCIE(ArteTVPlus7IE):

-    IE_NAME = u'arte.tv:ddc'
-    _VALID_URL = r'http?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
+    IE_NAME = 'arte.tv:ddc'
+    _VALID_URL = r'https?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'

 
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
 
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)


@@ -278,3 +222,39 @@ class ArteTVDDCIE(ArteTVPlus7IE):
         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)
         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)

+
+
+class ArteTVConcertIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:concert'
+    _VALID_URL = r'https?://concert\.arte\.tv/(?P<lang>de|fr)/(?P<id>.+)'
+
+    _TEST = {
+        'url': 'http://concert.arte.tv/de/notwist-im-pariser-konzertclub-divan-du-monde',
+        'md5': '9ea035b7bd69696b67aa2ccaaa218161',
+        'info_dict': {
+            'id': '186',
+            'ext': 'mp4',
+            'title': 'The Notwist im Pariser Konzertclub "Divan du Monde"',
+            'upload_date': '20140128',
+            'description': 'md5:486eb08f991552ade77439fe6d82c305',
+        },
+    }
+
+
+class ArteTVEmbedIE(ArteTVPlus7IE):
+    IE_NAME = 'arte.tv:embed'
+    _VALID_URL = r'''(?x)
+        http://www\.arte\.tv
+        /playerv2/embed\.php\?json_url=
+        (?P<json_url>
+            http://arte\.tv/papi/tvguide/videos/stream/player/
+            (?P<lang>[^/]+)/(?P<id>[^/]+)[^&]*
+        )
+    '''
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        lang = mobj.group('lang')
+        json_url = mobj.group('json_url')
+        return self._extract_from_json_url(json_url, video_id, lang)