trailer

scrapeit/imdb.py

@@ -110,6 +110,8 @@ class IMDb:
     self.locationUrl = "%slocations" % self.pageUrl
     self.externalreviewsSource = None
     self.externalreviewsUrl = "%sexternalreviews" % self.pageUrl
+    self.trailerSource = None
+    self.trailerUrl = "%strailers" % self.pageUrl
     
   def getPage(self, forcereload = False):
     if forcereload or not self.pageSource:
@@ -272,6 +274,7 @@ class IMDb:
     IMDbDict['release_date'] = self.parseReleaseinfo()
     IMDbDict['business'] = self.parseBusiness()
     IMDbDict['reviews'] = self.parseExternalreviews()
+    #IMDbDict['trailers'] = self.parseTrailers()
     self.IMDbDict = IMDbDict
     
     if IMDbDict['episode_of']:
@@ -492,6 +495,24 @@ class IMDb:
       return ret
     return {}
   
+  def getTrailers(self, forcereload = False):
+    if forcereload or not self.trailerSource:
+      self.trailerSource = read_url_utf8(self.trailerUrl)
+    return self.trailerSource
+  
+  def parseTrailers(self):
+    ret = {}
+    soup = BeautifulSoup(self.getTrailers())
+    for p in soup('p'):
+      if p('a') and p.firstText():
+        a = p('a')[0]
+        href = a['href']
+        if href and href.startswith('http'):
+          title = a.string
+          title = title.replace('www.', '')
+          ret[href] = title
+    return ret
+
 def guess(title, director=''):
   #FIXME: proper file -> title
   title = title.split('-')[0]

scrapeit/yahootrailers.py

@@ -0,0 +1,25 @@
+# -*- coding: utf-8 -*-
+# -*- Mode: Python; -*-
+# vi:si:et:sw=2:sts=2:ts=2
+from urllib import quote
+import re
+
+from BeautifulSoup import BeautifulSoup
+
+from utils import read_url, stripTags
+
+
+def trailerByTitle(title):
+  title = title.strip()
+  url = "http://movies.yahoo.com/mv/search?p=%s" % quote(title)
+  data = read_url(url)
+  soup = BeautifulSoup(data)
+  movies = soup('a', {'href': re.compile('http://movies.yahoo.com/movie.*?')})
+  if movies and movies[0].firstText() and title in movies[0].firstText():
+    info = movies[0]['href']
+    trailer = info.replace('/info', '/video')
+    data = read_url(info)
+    if trailer in data:
+      return trailer
+  return ''
+