[plugin.video.mediathek] 0.8.2 (#1106)

5 files changed, 65 insertions, 122 deletions

diff --git a/plugin.video.mediathek/addon.xml b/plugin.video.mediathek/addon.xml
index da032f8..abd64e4 100644
--- a/plugin.video.mediathek/addon.xml
+++ b/plugin.video.mediathek/addon.xml
@@ -2,7 +2,7 @@
 <addon
   id="plugin.video.mediathek"
   name="Mediathek"
-  version="0.8.1"
+  version="0.8.2"
   provider-name="Raptor 2101">
   <requires>
    <import addon="xbmc.python" version="2.25.0"/>
diff --git a/plugin.video.mediathek/changelog.txt b/plugin.video.mediathek/changelog.txt
index 77edc82..4d5387b 100644
--- a/plugin.video.mediathek/changelog.txt
+++ b/plugin.video.mediathek/changelog.txt
@@ -1,3 +1,4 @@
+0.8.2 - CHG: Reimplement ORF
 0.8.1 - CHG: Recover ARTE Search and Browsing Functionality
 0.8.0 - CHG: Version pump to seperate pre-krypton version and krypton version
 0.7.5 - CHG: reimplement ZDF Mediathak (important: ZDF enforces TLSv1 with SNI enabled, what is only supported by python 2.7.9.)
diff --git a/plugin.video.mediathek/mediathek/arte.py b/plugin.video.mediathek/mediathek/arte.py
index ac3fec6..4c61c4e 100644
--- a/plugin.video.mediathek/mediathek/arte.py
+++ b/plugin.video.mediathek/mediathek/arte.py
@@ -89,7 +89,7 @@ class ARTEMediathek(Mediathek):
     link = self.serachLink%searchText;
     pageContent = self.loadPage(link).decode('UTF-8');
     content = self.searchContent.search(pageContent).group(1);
-    content = BeautifulSoup(content);
+    content = BeautifulSoup(content,"html.parser");
     jsonContent = json.loads(content.prettify(formatter=None));
     linkCount = len(jsonContent["programs"]);
     for jsonObject in jsonContent["programs"]:
@@ -138,7 +138,7 @@ class ARTEMediathek(Mediathek):
     for name,regex in self.categories.iteritems():
       match = regex.search(pageContent);
       if(match is not None):
-        content = BeautifulSoup(match.group(1));
+        content = BeautifulSoup(match.group(1),"html.parser");
         jsonContent = json.loads(content.prettify(formatter=None))
         if(isinstance(jsonContent,list)):
           self.buildJsonLink(name,jsonContent)
@@ -154,7 +154,7 @@ class ARTEMediathek(Mediathek):
       match = regex.search(htmlPage);
       if(match is not None):
         someMatch = True;
-        content = BeautifulSoup(match.group(1));
+        content = BeautifulSoup(match.group(1),"html.parser");
         self.gui.log(content.prettify(formatter=None));
         jsonContent = json.loads(content.prettify(formatter=None))
         self.extractVideoLinksFromJson(jsonContent)
@@ -167,14 +167,14 @@ class ARTEMediathek(Mediathek):
 
   def showCluster(self):
     pageContent = self.loadPage(self.basePage).decode('UTF-8');
-    content = BeautifulSoup(self.regex_cluster.search(pageContent).group(1));
+    content = BeautifulSoup(self.regex_cluster.search(pageContent).group(1),"html.parser");
     jsonContent = json.loads(content.prettify(formatter=None))
     for menuItem in jsonContent:
       self.buildMenuEntry(menuItem);
 
   def showCategories(self):
     pageContent = self.loadPage(self.basePage).decode('UTF-8');
-    content = BeautifulSoup(self.regex_categories.search(pageContent).group(1));
+    content = BeautifulSoup(self.regex_categories.search(pageContent).group(1),"html.parser");
     jsonContent = json.loads(content.prettify(formatter=None))
     for jsonObject in jsonContent:
       jsonCategorie = jsonObject["category"]
diff --git a/plugin.video.mediathek/mediathek/orf.py b/plugin.video.mediathek/mediathek/orf.py
index 1146abf..ef7df1b 100644
--- a/plugin.video.mediathek/mediathek/orf.py
+++ b/plugin.video.mediathek/mediathek/orf.py
@@ -15,10 +15,9 @@
 #
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
-import re,time,urllib
-from xml.dom import Node;
-from xml.dom import minidom;
+import re,json,urllib;
 from mediathek import *
+from bs4 import BeautifulSoup;
 
 class ORFMediathek(Mediathek):
   def __init__(self, simpleXbmcGui):
@@ -27,44 +26,19 @@ class ORFMediathek(Mediathek):
     self.gui = simpleXbmcGui;
     self.menuTree = [];
     self.menuTree.append(TreeNode("0","Startseite","http://tvthek.orf.at/",True));
+    self.menuTree.append(TreeNode("1","Sendungen","http://tvthek.orf.at/profiles/a-z",True));
 
-    menuPage = self.loadPage(self.rootLink+"/programs");
-    findMenuLink = re.compile("<li><a href=\"(/programs/.*?)\" title=\".*?\">(.*?)</a></li>");
-    findCategorie = re.compile("<h4>(.*?)</h4>\\s*?<ul>((\\s*?%s\\s*?)+)</ul>"%findMenuLink.pattern)
-    categories = [];
 
-    for categorieMatch in findCategorie.finditer(menuPage):
-      title = categorieMatch.group(1);
-      items = [];
-      for menuMatch in findMenuLink.finditer(categorieMatch.group(2)):
-        items.append(TreeNode("1.%d.%d"%(len(categories),len(items)), menuMatch.group(2),"%s%s"%(self.rootLink,menuMatch.group(1)),True));
-      categories.append(TreeNode("1.%d"%len(categories), title,"",False,items));
+    self.searchLink = "http://tvthek.orf.at/search?q=%s"
 
-    self.menuTree.append(TreeNode("1","Sendungen","",False,categories));
 
-    videoLinkPage = "/programs/.*"
-    imageLink = "http://tvthek.orf.at/assets/.*?.jpeg"
 
 
-    self.regex_extractVideoPageLink = re.compile(videoLinkPage+"?\"");
-    self.regex_extractImageLink = re.compile(imageLink);
-    self.regex_extractTitle = re.compile("<strong>.*<span");
-    self.regex_extractVideoLink = re.compile("/programs/.*.asx");
-    self.regex_extractVideoObject = re.compile("<a href=\""+videoLinkPage+"\" title=\".*\">\\s*<span class=\"spcr\">\\s*<img src=\""+imageLink+"\" title=\".*\" alt=\".*\" />\\s*<span class=\".*\"></span>\\s*<strong>.*<span class=\"nowrap duration\">.*</span></strong>\\s*<span class=\"desc\">.*</span>\\s*</span>\\s*</a>");
+    self.regex_extractProfileSites = re.compile("<a class=\"item_inner clearfix\"\s*?href=\"(http://tvthek.orf.at/profile/.*?/\d+)\".*src=\"(http://api-tvthek.orf.at/uploads/media/profiles/.*?_profiles_list.jpeg)\"(.|\s)*?<h4 class=\"item_title\">(.*?)</h4>");
+    self.regex_extractTopicSites = re.compile("<a href=\"(http://tvthek.orf.at/topic/.*?/\d+)\"\s*?title=\"(.*?)\"\s*?class=\"more");
+    self.regex_extractVideoPages = re.compile("<a href=\"(http://tvthek.orf.at/.*?/\d+)\"");
+    self.regex_extractJson = re.compile("data-jsb=\"({&quot;videoplayer_id&quot;.*})\">");
 
-    self.regex_extractSearchObject = re.compile("<li class=\"clearfix\">\\s*<a href=\".*\" title=\".*\" class=\".*\"><img src=\".*\" alt=\".*\" /><span class=\"btn_play\">.*</span></a>\\s*<p>.*</p>\\s*<h4><a href=\".*\" title=\".*\">.*</a></h4>\\s*<p><a href=\".*\" title=\".*\"></a></p>\\s*</li>");
-
-    self.regex_extractProgrammLink = re.compile("/programs/.*?\"");
-    self.regex_extractProgrammTitle = re.compile("title=\".*?\"");
-    self.regex_extractProgrammPicture = re.compile("/binaries/asset/segments/\\d*/image1");
-
-    self.regex_extractFlashVars = re.compile("ORF.flashXML = '.*?'");
-    self.regex_extractHiddenDate = re.compile("\d{4}-\d{2}-\d{2}");
-    self.regex_extractXML = re.compile("%3C.*%3E");
-    self.regex_extractReferingSites = re.compile("<li><a href=\"/programs/\d+.*?/episodes/\d+.*?\"");
-
-    self.replace_html = re.compile("<.*?>");
-    self.searchLink = "http://tvthek.orf.at/search?q="
   @classmethod
   def name(self):
     return "ORF";
@@ -72,91 +46,59 @@ class ORFMediathek(Mediathek):
   def isSearchable(self):
     return True;
 
-  def createVideoLink(self,title,image,videoPageLink,elementCount):
-    videoPage = self.loadPage(self.rootLink+videoPageLink);
-
-    videoLink = self.regex_extractVideoLink.search(videoPage);
-    if(videoLink == None):
-      return;
-
-    simpleLink = SimpleLink(self.rootLink+videoLink.group(), 0);
-    videoLink = {0:simpleLink};
-    counter = 0
-    playlist = self.loadPage(simpleLink.basePath);
-    for line in playlist:
-      counter+=1;
-
-    if(counter == 1):
-      self.gui.buildVideoLink(DisplayObject(title,"",image,"",videoLink, True, time.gmtime()),self,elementCount);
-    else:
-      self.gui.buildVideoLink(DisplayObject(title,"",image,"",videoLink, "PlayList", time.gmtime()),self,elementCount);
-
   def searchVideo(self, searchText):
-    link = self.searchLink = "http://tvthek.orf.at/search?q="+searchText;
-    mainPage = self.loadPage(link);
-    result = self.regex_extractSearchObject.findall(mainPage);
-    for searchObject in result:
-      videoLink = self.regex_extractProgrammLink.search(searchObject).group().replace("\"","");
-      title = self.regex_extractProgrammTitle.search(searchObject).group().replace("title=\"","").replace("\"","");
-      title = title.decode("UTF-8");
-      pictureLink = self.regex_extractProgrammPicture.search(searchObject).group();
-
-      print videoLink;
+    self.buildPageMenu(self.searchLink%urllib.quote(searchText.encode('UTF-8')),0);
 
-      self.createVideoLink(title,pictureLink,videoLink, len(result));
-
-  def extractLinksFromFlashXml(self, flashXml, date, elementCount):
-    print flashXml.toprettyxml().encode('UTF-8');
-    playlistNode = flashXml.getElementsByTagName("Playlist")[0];
-    linkNode=flashXml.getElementsByTagName("AsxUrl")[0];
-    link=linkNode.firstChild.data;
-    asxLink = SimpleLink(self.rootLink+link,0);
-    videoLink = {0:asxLink};
-    for videoItem in playlistNode.getElementsByTagName("Items")[0].childNodes:
-      if(videoItem.nodeType == Node.ELEMENT_NODE):
-        titleNode=videoItem.getElementsByTagName("Title")[0];
-
-        descriptionNode=videoItem.getElementsByTagName("Description")[0];
-        title=titleNode.firstChild.data;
-
-        stringArray = link.split("mp4:");
+  def extractVideoLinks(self,videoPageLinks,elementCount):
+    for videoPageLink in videoPageLinks:
 
+      videoPage = self.loadPage(videoPageLink.group(1));
+      jsonContent = self.regex_extractJson.search(videoPage);
+      if(jsonContent == None):
+        return;
+      jsonContent = jsonContent.group(1);
+      jsonContent = BeautifulSoup(jsonContent,"html.parser");
+      jsonContent = json.loads(jsonContent.prettify(formatter=None).encode('UTF-8'));
+      jsonContent = jsonContent["selected_video"];
+      title = jsonContent["title"];
+      pictureLink = jsonContent["preview_image_url"];
+
+      videoLinks={};
+
+      for source in jsonContent["sources"]:
+        if(source["protocol"] == "http"):
+          quality = source["quality"];
+          url = source["src"];
+          if(quality == "Q1A"):
+            videoLinks[0] = SimpleLink(url, -1);
+          if(quality == "Q4A"):
+            videoLinks[1] = SimpleLink(url, -1);
+          if(quality == "Q6A"):
+            videoLinks[2] = SimpleLink(url, -1);
+          if(quality == "Q8C"):
+            videoLinks[3] = SimpleLink(url, -1);
+      if("title_separator" in jsonContent):
+        titleSeperator = jsonContent["title_separator"];
+        titleArray = title.split(titleSeperator);
         try:
-          description=descriptionNode.firstChild.data;
-        except:
-          description="";
-        self.gui.buildVideoLink(DisplayObject(title,"","",description,videoLink, True, date),self,elementCount);
-  def extractFlashLinks(self, flashVars,videoPageLinks,elementCount):
-    for flashVar in flashVars:
-      encodedXML = self.regex_extractXML.search(flashVar).group();
-      dateString = self.regex_extractHiddenDate.search(flashVar).group();
-      date = time.strptime(dateString,"%Y-%m-%d");      
-      parsedXML = minidom.parseString(urllib.unquote(encodedXML));  
-      self.extractLinksFromFlashXml(parsedXML, date,elementCount);
-    for videoPageLink in videoPageLinks:
-      videoPageLink = self.rootLink+videoPageLink.replace("<li><a href=\"","").replace("\"","");
-      print videoPageLink;
-      videoPage = self.loadPage(videoPageLink);
-      flashVars = self.regex_extractFlashVars.findall(videoPage);
-      for flashVar in flashVars:
-        encodedXML = self.regex_extractXML.search(flashVar).group();
-        dateString = self.regex_extractHiddenDate.search(flashVar).group();
-        date = time.strptime(dateString,"%Y-%m-%d");
-        parsedXML = minidom.parseString(urllib.unquote(encodedXML));  
-        self.extractLinksFromFlashXml(parsedXML,date,elementCount);
+          title = titleArray[0];
+          subTitle = titleArray[1];
+        except IndexError:
+          subTitle = "";
+        self.gui.buildVideoLink(DisplayObject(title,subTitle,pictureLink,"",videoLinks, True, None),self,elementCount);
+      else:
+        self.gui.buildVideoLink(DisplayObject(title,None,pictureLink,"",videoLinks, True, None),self,elementCount);
+
   def buildPageMenu(self, link, initCount):
     mainPage = self.loadPage(link);
-    videoPageLinks = self.regex_extractReferingSites.findall(mainPage);
-    flashVars = self.regex_extractFlashVars.findall(mainPage);
-    links = self.regex_extractVideoObject.findall(mainPage);
-    elementCount = initCount + len(links)+len(flashVars)+len(videoPageLinks);
-    self.extractFlashLinks(flashVars,videoPageLinks,elementCount);
-    for linkObject in links:
-
-      videoLink = self.regex_extractVideoPageLink.search(linkObject).group().replace("\"","");
-      image = self.regex_extractImageLink.search(linkObject).group();
-      title = self.regex_extractTitle.search(linkObject).group().decode('UTF8');
-      title = self.replace_html.sub("", title);
-      title = title.replace(" <span","");
-      self.createVideoLink(title,image,videoLink, elementCount);
+
+    for topic in self.regex_extractTopicSites.finditer(mainPage):
+      self.gui.buildVideoLink(DisplayObject(topic.group(2),None,None,"",topic.group(1), False, None),self,0);
+      initCount=initCount+1;
+    for profile in self.regex_extractProfileSites.finditer(mainPage):
+      self.gui.buildVideoLink(DisplayObject(profile.group(4),None,profile.group(2),"",profile.group(1), False, None),self,0);
+      initCount=initCount+1;
+    videoPageLinks = list(self.regex_extractVideoPages.finditer(mainPage));
+
+    self.extractVideoLinks(videoPageLinks,len(videoPageLinks)+initCount);
 
diff --git a/plugin.video.mediathek/simplexbmc.py b/plugin.video.mediathek/simplexbmc.py
index e55d0d8..1529b83 100644
--- a/plugin.video.mediathek/simplexbmc.py
+++ b/plugin.video.mediathek/simplexbmc.py
@@ -99,7 +99,7 @@ class SimpleXbmcGui(object):
     return metaData;
 
   def transformHtmlCodes(self, content):
-    return BeautifulSoup(content).prettify(formatter=None);
+    return BeautifulSoup(content,"html.parser").prettify(formatter=None);
 
   def buildMenuLink(self,menuObject,mediathek,objectCount):
     title = menuObject.name;