fixes

- parse video xml with lxml instead of regexpes
- add a prefix to IDs to know what kind of object it is

modules/arte/backend.py

@@ -46,8 +46,21 @@ class ArteBackend(BaseBackend, ICapVideo, ICapCollection):
     def create_default_browser(self):
         return self.create_browser(lang=self.config['lang'].get(), quality=self.config['quality'].get())
 
+    def split_id(self, _id):
+        try:
+            site, _id = _id.split('.', 1)
+        except ValueError:
+            site = 'videos'
+
+        return site, _id
+
     def get_video(self, _id):
         with self.browser:
+            site, _id = self.split_id(_id)
+
+            if site == 'live':
+                return self.browser.get_live_video(_id)
+            else:
                 return self.browser.get_video(_id)
 
     def search_videos(self, pattern, sortby=ICapVideo.SEARCH_RELEVANCE, nsfw=False, max_results=None):
@@ -58,10 +71,12 @@ class ArteBackend(BaseBackend, ICapVideo, ICapCollection):
         if fields != ['thumbnail']:
             # if we don't want only the thumbnail, we probably want also every fields
             with self.browser:
+                site, _id = self.split_id(video.id)
+
                 if isinstance(video,ArteVideo):
-                    video = self.browser.get_video(ArteVideo.id2url(video.id), video)
+                    video = self.browser.get_video(_id, video)
                 if isinstance(video,ArteLiveVideo):
-                    video = self.browser.get_live_video(ArteLiveVideo.id2url(video.id), video)
+                    video = self.browser.get_live_video(_id, video)
         if 'thumbnail' in fields and video and video.thumbnail:
             with self.browser:
                 video.thumbnail.data = self.browser.readurl(video.thumbnail.url)
@@ -84,7 +99,7 @@ class ArteBackend(BaseBackend, ICapVideo, ICapCollection):
                             yield categorie
                 if collection.path_level == 2:
                     if collection.split_path[0] == u'live':
-                        for video in self.browser.live_videos(ArteLiveCollection.id2url(collection.basename)):
+                        for video in self.browser.live_videos(ArteLiveCollection.id2url(collection.basename, self.browser.lang)):
                             yield video
 
     def validate_collection(self, objs, collection):

modules/arte/browser.py

@@ -22,7 +22,7 @@ from weboob.tools.browser import BaseBrowser
 from weboob.tools.browser.decorators import id2url
 
 from .pages import IndexPage, VideoPage, ArteLivePage, ArteLiveCategorieVideoPage, ArteLiveVideoPage
-from .video import ArteVideo
+from .video import ArteVideo, ArteLiveVideo
 
 
 __all__ = ['ArteBrowser']
@@ -42,22 +42,23 @@ class ArteBrowser(BaseBrowser):
     SEARCH_LANG = {'fr': 'recherche', 'de': 'suche', 'en': 'search'}
 
     def __init__(self, lang, quality, *args, **kwargs):
-        BaseBrowser.__init__(self, *args, **kwargs)
         self.lang = lang
         self.quality = quality
+        BaseBrowser.__init__(self, *args, **kwargs)
 
     @id2url(ArteVideo.id2url)
     def get_video(self, url, video=None):
         self.location(url)
         return self.page.get_video(video, self.lang, self.quality)
 
+    @id2url(ArteLiveVideo.id2url)
     def get_live_video(self, url, video=None):
         self.location(url)
         assert self.is_on_page(ArteLiveVideoPage)
-        return self.page.get_video(url, video, self.lang, self.quality)
+        return self.page.get_video(video, self.lang, self.quality)
 
     def home(self):
-        self.location('http://videos.arte.tv/fr/videos/toutesLesVideos')
+        self.location('http://videos.arte.tv/%s/videos/toutesLesVideos' % self.lang)
 
     def search_videos(self, pattern):
         self.location(self.buildurl('/%s/do_search/videos/%s' % (self.lang, self.SEARCH_LANG[self.lang]), q=pattern.encode('utf-8')))

modules/arte/collection.py

@@ -23,5 +23,5 @@ __all__ = ['ArteLiveCollection']
 
 class ArteLiveCollection(Collection):
     @classmethod
-    def id2url(cls, _id):
+    def id2url(cls, _id, lang):
-        return 'http://liveweb.arte.tv/fr/cat/%s/' % _id
+        return 'http://liveweb.arte.tv/%s/cat/%s/' % (lang, _id)

modules/arte/pages.py

@@ -33,24 +33,19 @@ from .collection import ArteLiveCollection
 __all__ = ['IndexPage', 'VideoPage', 'ArteLivePage', 'ArteLiveCategorieVideoPage', 'ArteLiveVideoPage']
 
 class ArteLiveVideoPage(BasePage):
-    def get_video(self, url, video=None, lang='fr', quality='hd'):
+    def get_video(self, video=None, lang='fr', quality='hd'):
         if not video:
             video = ArteVideo(self.group_dict['id'])
 
-        HD = re.compile("(?<=<urlHd>)(.*)(?=</urlHd>)", re.DOTALL)
-        SD = re.compile("(?<=<urlSd>)(.*)(?=</urlSd>)", re.DOTALL)
-
-        page = self.browser.readurl(url)
         urls = {}
-        try:
+        for url in self.document.xpath('//video')[0].getchildren():
-            urls['hd'] = u'%s' %HD.search(page).group(0).split('?')[0]
+            if url.tag.startswith('url'):
-        except AttributeError:
+                urls[url.tag[-2:]] = url.text
-            urls['hd'] = None
+
-        try:
+        if quality in urls:
-            urls['sd'] = u'%s' %SD.search(page).group(0).split('?')[0]
-        except AttributeError:
-            urls['sd'] = None
             video.url = urls[quality]
+        else:
+            video.url = urls.popitem()[1]
         return video
 
 class ArteLiveCategorieVideoPage(BasePage):

modules/arte/video.py

@@ -31,6 +31,9 @@ class ArteVideo(BaseVideo):
 
 
 class ArteLiveVideo(BaseVideo):
+    def __init__(self, _id, *args, **kwargs):
+        BaseVideo.__init__(self, 'live.%s' % _id, *args, **kwargs)
+
     @classmethod
     def id2url(cls, _id):
         return 'http://arte.vo.llnwd.net/o21/liveweb/events/event-%s.xml' % _id