Merge pull request #1620 from jaimeMF/console_script

[youtube-dl] / youtube_dl / extractor / arte.py

diff --git a/youtube_dl/extractor/arte.py b/youtube_dl/extractor/arte.py
index d296b6d63343a0b9d4f5b365cbec4b16f0a1df46..d39b489518f7bc699854d9ddeb7cd4fa357a6c2b 100644 (file)
--- a/youtube_dl/extractor/arte.py
+++ b/youtube_dl/extractor/arte.py
@@ -1,3 +1,4 @@
+# encoding: utf-8
 import re
 import json
 import xml.etree.ElementTree
@@ -8,6 +9,7 @@ from ..utils import (
     find_xpath_attr,
     unified_strdate,
     determine_ext,
+    get_element_by_id,
 )
 
 # There are different sources of video in arte.tv, the extraction process 
@@ -126,14 +128,21 @@ 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>.*?)(\?.*)?'
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+    @classmethod
+    def _extract_url_info(cls, url):
+        mobj = re.match(cls._VALID_URL, url)
         lang = mobj.group('lang')
         # This is not a real id, it can be for example AJT for the news
         # http://www.arte.tv/guide/fr/emissions/AJT/arte-journal
         video_id = mobj.group('id')
+        return video_id, lang
 
+    def _real_extract(self, url):
+        video_id, lang = self._extract_url_info(url)
         webpage = self._download_webpage(url, video_id)
+        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_info = self._download_webpage(json_url, video_id, 'Downloading info json')
@@ -165,12 +174,27 @@ class ArteTVPlus7IE(InfoExtractor):
         # Some formats use the m3u8 protocol
         formats = filter(lambda f: f.get('videoFormat') != 'M3U8', formats)
         # We order the formats by quality
-        formats = sorted(formats, key=lambda f: int(f.get('height',-1)))
+        formats = list(formats) # in python3 filter returns an iterator
+        if re.match(r'[A-Z]Q', formats[0]['quality']) is not None:
+            sort_key = lambda f: ['HQ', 'MQ', 'EQ', 'SQ'].index(f['quality'])
+        else:
+            sort_key = lambda f: int(f.get('height',-1))
+        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):
+            quality = format_info['quality']
+            m_quality = re.match(r'\w*? - (\d*)p', quality)
+            if m_quality is not None:
+                quality = m_quality.group(1)
+            if format_info.get('versionCode') is not None:
+                format_id = u'%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'),
                 'height': format_info.get('height'),
             }
@@ -183,8 +207,6 @@ class ArteTVPlus7IE(InfoExtractor):
                 info['ext'] = determine_ext(info['url'])
             return info
         info_dict['formats'] = [_format(f) for f in formats]
-        # TODO: Remove when #980 has been merged 
-        info_dict.update(info_dict['formats'][-1])
 
         return info_dict
 
@@ -198,7 +220,25 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
         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 #2 - Corporate Design',
+            u'title': u'Agentur Amateur / Agence Amateur #2 : Corporate Design',
         },
     }
 
+
+class ArteTVFutureIE(ArteTVPlus7IE):
+    IE_NAME = u'arte.tv:future'
+    _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',
+        },
+    }
+
+    def _real_extract(self, url):
+        anchor_id, lang = self._extract_url_info(url)
+        webpage = self._download_webpage(url, anchor_id)
+        row = get_element_by_id(anchor_id, webpage)
+        return self._extract_from_webpage(row, anchor_id, lang)