From abf263af138679121d01aed5268d10198cb9c92d Mon Sep 17 00:00:00 2001
From: j <j@0xdb.org>
Date: Tue, 29 Apr 2008 21:09:10 +0200
Subject: [PATCH] get Trailers

---
 ox/imdb.py | 146 ++++++++++++++++++++++-------------------------------
 1 file changed, 61 insertions(+), 85 deletions(-)

diff --git a/ox/imdb.py b/ox/imdb.py
index bcb77e1..8fc9ade 100644
--- a/ox/imdb.py
+++ b/ox/imdb.py
@@ -4,7 +4,7 @@
 
 from oxutils import *
 import urllib2
-from urllib import quote
+from urllib import quote, unquote
 import re, time
 import os
 import time
@@ -39,8 +39,10 @@ def getUrlBase(imdbId):
 def getRawMovieData(imdbId):
   imdbId = normalizeImdbId(imdbId)
   data = dict()
-  data['credits'] = parseCredits(imdbId)
-  data['poster'] = findRegexp(data, 'name="poster".*?<img .*?src="(.*?)"')
+  data['title'] = getTitle(imdbId)
+  data['credits'] = getCredits(imdbId)
+  data['poster'] = getPoster(imdbId)
+  data['trailers'] = getMovieTrailers(imdbId)
 
 def parseBase(imdbId):
   data = getUrl(getUrlBase(imdbId))
@@ -59,6 +61,10 @@ def parseBase(imdbId):
   return info
   return soup
 
+def getPoster(imdbId):
+  data = getUrl(getUrlBase(imdbId))
+  return findRegexp(data, 'name="poster".*?<img .*?src="(.*?)"')
+
 def getTitle(imdbId):
   title = ''
   data = getUrl(getUrlBase(imdbId))
@@ -96,7 +102,7 @@ def creditList(data, section=None):
     credits.append(c)
   return credits
 
-def parseCredits(imdbId):
+def getCredits(imdbId):
   credits = dict()
   url = "%s/fullcredits" % getUrlBase(imdbId)
   data = getUrlUnicode(url)
@@ -107,24 +113,41 @@ def parseCredits(imdbId):
       credits[section[0]] = creditList(g, section[0])
   return credits
 
+def getMovieTrailers(imdbId):
+  url = "%s/trailers" % getUrlBase(imdbId)
+  data = getUrlUnicode(url)
+  soup = BeautifulSoup(data)
+  videos = soup('div', {'class':"video-gallery"})
+  trailers = []
+  if videos:
+    for a in videos[0]('a'):
+      title = stripTags(unicode(a)).strip()
+      url = 'http://www.imdb.com' + a['href']
+      videoId = findRegexp(url, '/(vi\d*?)/')
+      iframeUrl = "http://www.imdb.com/video/trailer/%s/player" % videoId
+      iframe = getUrlUnicode(iframeUrl)
+      videoUrl = unquote(findRegexp(iframe, 'addVariable\("file", "(.*?)"'))
+      trailers.append({'title': title, 'url': url, 'iframe': iframeUrl, 'flv':videoUrl})
+  return trailers
+
+def getMovieStills(imdbId):
+  url = "http://www.imdb.com/gallery/ss/%s" % imdbId
+  data = getUrlUnicode(url)
+  s_ = re.compile('''<img width="(\d*?)" height="(\d*?)" src="http://i.imdb.com/Photos/Ss/%s/th-(.*?).jpg"''' % imdbId).findall(data)
+  stills = []
+  for s in s_:
+    if int(s[0]) > int(s[1]):
+      stills.append("http://i.imdb.com/Photos/Ss/%s/%s.jpg" % (imdbId, s[2]))
+  if not stills:
+    s_ = re.compile('''<img width="(\d*?)" height="(\d*?)" src="http://(.*?)p.jpg"''').findall(data)
+    stills = []
+    for s in s_:
+      if int(s[0]) > int(s[1]):
+        stills.append("http://%sf.jpg" % s[2])
+  return stills
+
 '''the old code below'''
 
-def get_image(url):
-  return getUrl(url)
-
-def _castList(data, regexp):
-  soup = re.compile(regexp).findall(data)
-  if soup:
-    soup = BeautifulSoup(soup[0])
-    names = []
-    for i in soup('a', {'href': re.compile('/name/nm')}):
-      if i.string:
-        cast = stripTags(i.string)
-        if cast not in names:
-          names.append(cast)
-    return names
-  return []
-
 class IMDb:
   def __init__(self, imdbId):
     self.imdb = imdbId
@@ -140,7 +163,6 @@ class IMDb:
     self.triviaUrl = "%strivia" % self.pageUrl
     self.locationUrl = "%slocations" % self.pageUrl
     self.externalreviewsUrl = "%sexternalreviews" % self.pageUrl
-    self.trailerUrl = "%strailers" % self.pageUrl
 
   def getPage(self):
     return getUrlUnicode(self.pageUrl)
@@ -163,8 +185,10 @@ class IMDb:
         parsed_value = int(parsed_value) * 60
     elif key in ('country', 'language'):
       parsed_value = value.split(' / ')
+      parsed_value = [v.strip() for v in parsed_value]
     elif key == 'genre':
       parsed_value = value.replace('more', '').strip().split(' / ')
+      parsed_value = [v.strip() for v in parsed_value]
     elif key == 'tagline':
       parsed_value = value.replace('more', '').strip()
     elif key == 'plot_outline':
@@ -187,7 +211,7 @@ class IMDb:
         if m:
           parsed_value = m[0]
       parsed_value = parsed_value.split('<br />')[-1].split('(')[0]
-      director = self.parseCredits().get('director', None)
+      director = self.getCredits().get('director', None)
       if director:
         director = director[0]
         parsed_value = parsed_value.replace(director, '')
@@ -200,34 +224,17 @@ class IMDb:
     return parsed_value
 
   def parseTitle(self):
-    title = ''
-    data = self.getPage()
-    soup = BeautifulSoup(data)
-    html_title = soup('div', {'id': 'tn15title'})
-    if not html_title:
-      html_title = soup('title')
-    if html_title:
-      html_title = str(html_title[0])
-      html_title = html_title.replace('<br />', ' ').replace('  ', ' ')
-      title = stripTags(html_title)
-      title = re.sub('\(\d{4}\)', '', title)
-      title = re.sub('\(\d{4}/I*\)', '', title)
-      for t in ('TV series', 'TV-Series', 'TV mini-series', '(mini)', '(VG)', '(V)', '(TV)'):
-        title = title.replace(t, '')
-    if title.find(u'\xa0') > -1:
-      title = title[:title.find(u'\xa0')]
-    title = normalizeTitle(title.strip())
-    if title.startswith('"') and title.endswith('"'):
-      title = normalizeTitle(title[1:-1])
-    elif title.startswith('"') and title.find('"',1) > 0 and \
-        title.find('"',1) == title.rfind('"'):
-        se = re.compile("Season (\d*), Episode (\d*)\)").findall(data)
-        if se:
-          se = se[0]
-          se = ' (S%02dE%02d)' % (int(se[0]), int(se[1]))
-          title = normalizeTitle(title[1:title.rfind('"')]) + se + title[title.rfind('"')+1:]
-        else:
-          title = normalizeTitle(title[1:title.rfind('"')]) + ':' + title[title.rfind('"')+1:]
+    title = getTitle(self.imdb)
+    title = normalizeTitle(title)
+    if title.startswith('"') and title.find('"',1) > 0 and \
+      title.find('"',1) == title.rfind('"'):
+      se = re.compile("Season (\d*), Episode (\d*)\)").findall(data)
+      if se:
+        se = se[0]
+        se = ' (S%02dE%02d)' % (int(se[0]), int(se[1]))
+        title = normalizeTitle(title[1:title.rfind('"')]) + se + title[title.rfind('"')+1:]
+      else:
+        title = normalizeTitle(title[1:title.rfind('"')]) + ':' + title[title.rfind('"')+1:]
     return normalizeTitle(title)
 
   def parseYear(self):
@@ -252,7 +259,7 @@ class IMDb:
     data = self.getPage()
     IMDbDict ={}
     #Poster
-    IMDbDict['poster'] = findRegexp(data, 'name="poster".*?<img .*?src="(.*?)"')
+    IMDbDict['poster'] = getPoster(self.imdb)
     if not IMDbDict['poster']:
       IMDbDict['poster'] = 'http://i.imdb.com/Heads/npa.gif'
     #Title, Year
@@ -295,7 +302,7 @@ class IMDb:
       IMDbDict['tvshow'] = True
     else:
       IMDbDict['tvshow'] = False
-    IMDbDict['credits'] = self.parseCredits()
+    IMDbDict['credits'] = self.getCredits()
     IMDbDict['plot'] = self.parsePlot()
     IMDbDict['keywords'] = self.parseKeywords()
 
@@ -316,8 +323,8 @@ class IMDb:
           IMDbDict[key] = episode_of[key]
     return self.IMDbDict
 
-  def parseCredits(self):
-    raw_credits = parseCredits(self.imdb)
+  def getCredits(self):
+    raw_credits = getCredits(self.imdb)
     credits = {}
 
     def getNames(creditList):
@@ -489,22 +496,6 @@ class IMDb:
       return ret
     return {}
 
-  def getTrailer(self):
-    return getUrlUnicode(self.trailerUrl)
-
-  def parseTrailer(self):
-    ret = {}
-    soup = BeautifulSoup(self.getTrailer())
-    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]
@@ -574,21 +565,6 @@ def getEpisodeData(title, episode, show_url = None):
     episodeData['imdb'] = i['episodes'][episode]['imdb']
   return episodeData
 
-def getMovieStills(imdbId):
-  data = getUrl("http://imdb.com/gallery/ss/%s" % imdbId)
-  s_ = re.compile('''<img width="(\d*?)" height="(\d*?)" src="http://i.imdb.com/Photos/Ss/%s/th-(.*?).jpg"''' % imdbId).findall(data)
-  stills = []
-  for s in s_:
-    if int(s[0]) > int(s[1]):
-      stills.append("http://i.imdb.com/Photos/Ss/%s/%s.jpg" % (imdbId, s[2]))
-  if not stills:
-    s_ = re.compile('''<img width="(\d*?)" height="(\d*?)" src="http://(.*?)p.jpg"''').findall(data)
-    stills = []
-    for s in s_:
-      if int(s[0]) > int(s[1]):
-        stills.append("http://%sf.jpg" % s[2])
-  return stills
-
 if __name__ == '__main__':
   import sys
   #print parse(sys.argv[1])