update rottentomatoes and metacritic

ox/web/metacritic.py

@@ -2,10 +2,16 @@
 # vi:si:et:sw=4:sts=4:ts=4
 import re
 from urllib import quote
+from lxml.html import document_fromstring
 
 from ox.cache import readUrl, readUrlUnicode
-from ox import findRe, decodeHtml, stripTags
+from ox import findRe, stripTags
 
+def getUrlByImdb(imdb):
+    url = "http://www.imdb.com/title/tt%s/criticreviews" % imdb
+    data = readUrl(url)
+    metacritic_url = findRe(data, '"(http://www.metacritic.com/movie/.*?)"')
+    return metacritic_url or None
 
 def getMetacriticShowUrl(title):
     title = quote(title)
@@ -13,33 +19,39 @@ def getMetacriticShowUrl(title):
     data = readUrl(url)
     return findRe(data, '(http://www.metacritic.com/tv/shows/.*?)\?')
 
-def getData(title, url=None):
-  if not url:
-    url = getMetacriticShowUrl(title)
-  if not url:
-    return None
+def getData(url):
     data = readUrlUnicode(url)
-  score = findRe(data, 'ALT="Metascore: (.*?)"')
+    doc = document_fromstring(data)
+    score = filter(lambda s: s.attrib.get('property') == 'v:average',
+                   doc.xpath('//span[@class="score_value"]'))
     if score:
-    score = int(score)
+        score = int(score[0].text)
     else:
         score = -1
-
-  reviews = re.compile(
-            '<div class="scoreandreview"><div class="criticscore">(.*?)</div>'
-            '.*?<span class="publication">(.*?)</span>'
-            '.*?<span class="criticname">(.*?)</span></div>'
-            '.*?<div class="quote">(.*?)<br>'
-            '.*?<a href="(.*?)" ', re.DOTALL).findall(data)
+    authors = [a.text
+        for a in doc.xpath('//div[@class="review_content"]//div[@class="author"]//a')]
+    publications = [d.text
+        for d in doc.xpath('//div[@class="review_content"]//div[@class="source"]/a')]
+    reviews = [d.text
+        for d in doc.xpath('//div[@class="review_content"]//div[@class="review_body"]')]
+    scores = [int(d.text.strip())
+        for d in doc.xpath('//div[@class="review_content"]//div[contains(@class, "critscore")]')]
+    links = [a.attrib['href']
+        for a in doc.xpath('//div[@class="review_content"]//a[contains(@class, "external")]')]
 
     metacritics = []
-  for review in reviews:
+    for i in range(len(authors)):
         metacritics.append({
-        'score': int(review[0]),
-        'publication':review[1],
-        'critic':decodeHtml(review[2]),
-        'quote': stripTags(review[3]).strip(),
-        'link': review[4],
+            'score': scores[i],
+            'publication': publications[i],
+            'critic': authors[i],
+            'quote': stripTags(reviews[i]).strip(),
+            'link': links[i],
         })
-  return dict(score = score, critics = metacritics, url = url)
+        
+    return {
+    'score': score,
+    'critics': metacritics,
+    'url': url
+    }
 

ox/web/rottentomatoes.py

@@ -6,7 +6,7 @@ from ox.cache import getHeaders, readUrl, readUrlUnicode
 from ox import findRe, stripTags
 
 
-def readUrlByImdb(imdb):
+def getUrlByImdb(imdb):
     #this would also wor but does not cache:
     '''
     from urllib2 import urlopen
@@ -21,14 +21,33 @@ def readUrlByImdb(imdb):
             return "http://www.rottentomatoes.com" + movies[0]
     return None
 
+def get_og(data, key):
+    return findRe(data, '<meta property="og:%s".*?content="(.*?)"' % key)
+
 def getData(url):
-    data = readUrlUnicode(url)
+    data = readUrl(url)
     r = {}
     r['title'] = findRe(data, '<h1 class="movie_title">(.*?)</h1>')
     if '(' in r['title']:
         r['year'] = findRe(r['title'], '\((\d*?)\)')
-        r['title'] = re.sub('\((\d*?)\)', '', r['title']).strip()
-    r['synopsis'] = findRe(data, '<span id="movie_synopsis_all".*?>(.*?)</span>')
-    r['average rating'] = findRe(data, '<div id="bubble_allCritics".*?>(.*?)</div>').strip()
+        r['title'] = stripTags(re.sub('\((\d*?)\)', '', r['title'])).strip()
+    r['summary'] = stripTags(findRe(data, '<p id="movieSynopsis" class="movie_synopsis" itemprop="description">(.*?)</p>')).strip()
+    r['summary'] = r['summary'].replace('\t', ' ').replace('\n', ' ').replace('  ', ' ').replace('  ', ' ')
+    if not r['summary']:
+        r['summary'] = get_og(data, 'description')
+
+    meter = re.compile('<span id="all-critics-meter" class="meter(.*?)">(.*?)</span>').findall(data)
+    meter = filter(lambda m: m[1].isdigit(), meter)
+    if meter:
+        r['tomatometer'] = meter[0][1]
+    r['rating'] = findRe(data, 'Average Rating: <span>([\d.]+)/10</span>')
+    r['user_score'] = findRe(data, '<span class="meter popcorn numeric ">(\d+)</span>')
+    r['user_rating'] = findRe(data, 'Average Rating: ([\d.]+)/5')
+    poster = get_og(data, 'image')
+    if poster and not 'poster_default.gif' in poster:
+        r['posters'] = [poster]
+    for key in r.keys():
+        if not r[key]:
+            del r[key]
     return r