spiegel works / golem not

reader/main.py

@@ -14,7 +14,9 @@ logging.getLogger('flask_ask').setLevel(logging.DEBUG)
 @ask.intent('searchon', mapping={'site': 'Site'}, default={'site': 'golem'})
 def search_on(site):
     try:
+
         session.attributes["siteName"] = site
+        print(session.attributes["siteName"])
     except: 
         print("error")
 
@@ -36,6 +38,8 @@ def search_for(searchTerm):
 
     if site == "golem":
         obj = site2.Golem()
+    elif site == "spiegel":
+        obj = site2.Spiegel()
     elif site is None:
         session.attributes["searchTerm"] = searchTerm
         session.attributes["lastCall"] = "searchfor"
@@ -48,7 +52,7 @@ def search_for(searchTerm):
     response = "Für welchen der folgenden Artikel interessieren Sie sich?"
 
     if len(articles) > 0:
-        for i in range(0, max(5, len(articles))):
+        for i in range(0, min(5, len(articles))):
             response += articles[i] 
     else:
         return question("Dazu konnte nichts gefunden werden. Möchten Sie nach etwas anderem Suchen?")
@@ -59,34 +63,58 @@ def search_for(searchTerm):
 
 @ask.intent('News', mapping={'site': 'Site'}, default={'site': ''})
 def news(site):
-
+    try:
+        site = site.lower()
+        session.attributes["siteName"] = site
+    except: 
+        print("error")
+    print(site)
     if site == "golem":
         obj = site2.Golem()
+    elif site == "spiegel":
+        obj = site2.Spiegel()
     elif site == '':
         session.attributes["lastCall"] = "news"
         return question("Auf welcher Seite wollen Sie hiernach Suchen?")
     else:
         return statement("error")
 
-    news = obj.get_news()
+    news, links = obj.get_news()
+    print(news)
+    session.attributes["lastSearch"] = links
 
     response = ""
-    for i in range(0, 5):
+    for i in range(0, min(5, len(news))):
         response += news[i] + ". "
 
     session.attributes["lastCall"] = "news"
-    return statement(response)
+    return question(response)
 
 @ask.intent('SearchTwo', mapping={'number': 'Nummer'}, default={'number': 1})
 def search_answer(number):
-    print(number)
-    obj = site2.Golem()
+    try:
+        site = session.attributes["siteName"]
+    except:
+        site = None
 
-    art = obj.read_headlines(session.attributes["lastSearch"][int(number)-1])
+    if site == "golem":
+        obj = site2.Golem()
+    elif site == "spiegel":
+        obj = site2.Spiegel()
+
+    links = session.attributes["lastSearch"]
+
+    newLinks = []
+    for link in links:
+        if "http" not in link:
+            newLinks.append(obj.baseURL + link)
+    links = newLinks
+
+    art = obj.read_headlines(links[int(number)-1])
     response = ""
     for element in art:
         response +=  element + "   "
-
+    print(links)
     session.attributes["lastCall"] = "search2"
     return statement(response)
 

reader/siteobj.py

@@ -12,18 +12,11 @@ class Site:
     xPath["searchArticle"] = ""
     xPath["searchLinks"] = ""
     xPath["newsArticle"] = ""
+    xPath["newsLinks"] = ""
     xPath["readHeadlineTitle"] = ""
     xPath["readHeadlineText"] = ""
     xPath["readArticleText"] = ""
     
-    header_values = {
-            'Connection:' : 'Keep-alive',
-            'name' : 'Michael Foord',
-            'location' : 'Northampton',
-            'language' : 'German',
-            'User-Agent': 'Mozilla 4/0'}
-    
-
     def __init__(self):
         return None
 
@@ -42,7 +35,8 @@ class Site:
         tree = html.fromstring(site.content)
 
         articles = tree.xpath(self.xPath["newsArticle"])
-        return articles
+        links = tree.xpath(self.xPath["newsLinks"])
+        return articles, links
 
     def read_headlines(self, url):
         site = requests.get(url)
@@ -65,10 +59,24 @@ class Golem(Site):
     siteName = "golem"
     baseURL = "https://www.golem.de/"   
     searchURLString = "https://suche.golem.de/search.php?l=10&q="
+
     Site.xPath["searchArticle"] = '//span[@class="dh2 head2"]/text()'
     Site.xPath["searchLinks"] = '//ol[@class="list-articles"]/li/header//@href'
-    Site.xPath["newsArticle"] = '//h2[@class="head2"]/text()'
+    Site.xPath["newsArticle"] = '//li//h2/text()'
+    Site.xPath["newsLinks"] = '//header[@class="cluster-header"]//@href'
     Site.xPath["readHeadlineTitle"] = '//header/h1/span[@class="dh1 head5"]/text()'
     Site.xPath["readHeadlineText"] = '//header/p/text()'
     Site.xPath["readArticleText"] = '//div[@class="formatted"]/p/text()'
 
+class Spiegel(Site):
+    siteName = "spiegel"
+    baseURL = "https://www.spiegel.de/"   
+    searchURLString = "https://www.spiegel.de/suche/?suchbegriff="
+
+    Site.xPath["searchArticle"] = '//span[@class="dh2 head2"]/text()'
+    Site.xPath["searchLinks"] = '//ol[@class="list-articles"]/li/header//@href'
+    Site.xPath["newsArticle"] = '//div[@class="column-wide pano_xxl"]//div[@class="teaser"]//h2[@class="article-title"]//span[@class="headline"]/text()'
+    Site.xPath["newsLinks"] = '//div[@class="column-wide pano_xxl"]//div[@class="teaser"]//h2[@class="article-title"]//@href'
+    Site.xPath["readHeadlineTitle"] = '//div[@class="column-both"]//span[@class="headline"]//text()'
+    Site.xPath["readHeadlineText"] = '//div[@class="column-both"]/p/strong/text()'
+    Site.xPath["readArticleText"] = '//div[@class="formatted"]/p/text()'