Speed up Userscript by not using BeautifulSoup. fix #796

couchpotato/core/providers/userscript/allocine/main.py

@@ -1,5 +1,9 @@
-from bs4 import BeautifulSoup
+from couchpotato.core.logger import CPLog
 from couchpotato.core.providers.userscript.base import UserscriptBase
+import traceback
+
+log = CPLog(__name__)
+
 
 class AlloCine(UserscriptBase):
 
@@ -15,11 +19,17 @@ class AlloCine(UserscriptBase):
         except:
             return
 
-        html = BeautifulSoup(data)
-        title = html.find('title').contents[0].strip()
-        split = title.split(') - ')
+        name = None
+        year = None
 
-        name = split[0][:-5].strip()
-        year = split[0][-4:]
+        try:
+            start = data.find('<title>')
+            end = data.find('</title>', start)
+            page_title = data[start + len('<title>'):end].strip().split('-')
+
+            name = page_title[0].strip()
+            year = page_title[1].strip()[-4:]
+            return self.search(name, year)
+        except:
+            log.error('Failed parsing page for title and year: %s', traceback.format_exc())
 
-        return self.search(name, year)

couchpotato/core/providers/userscript/rottentomatoes/main.py

@@ -1,6 +1,10 @@
-from bs4 import BeautifulSoup
-from couchpotato.core.event import fireEvent
+from couchpotato.core.logger import CPLog
 from couchpotato.core.providers.userscript.base import UserscriptBase
+import re
+import traceback
+
+log = CPLog(__name__)
+
 
 class RottenTomatoes(UserscriptBase):
 
@@ -16,7 +20,20 @@ class RottenTomatoes(UserscriptBase):
         except:
             return
 
-        html = BeautifulSoup(data)
-        title = html.find('span', {'itemprop':'name'}).text
-        info = fireEvent('scanner.name_year', title, single = True)
-        return self.search(info['name'], info['year'])
+        try:
+            name = None
+            year = None
+            metas = re.findall("property=\"(video:release_date|og:title)\" content=\"([^\"]*)\"", data)
+
+            for meta in metas:
+                mname, mvalue = meta
+                if mname == 'og:title':
+                    name = mvalue.decode('unicode_escape')
+                elif mname == 'video:release_date':
+                    year = mvalue[:4]
+
+            if name and year:
+                return self.search(name, year)
+
+        except:
+            log.error('Failed parsing page for title and year: %s', traceback.format_exc())