Fixed IMDB cast loading mechanism by using the good old indexing technique of yesteryear.

1 files changed, 14 insertions, 6 deletions

diff --git a/framed.py b/framed.py
index e36616d..75b3553 100644
--- a/framed.py
+++ b/framed.py
@@ -57,7 +57,6 @@ class AnswerQuestion(webapp.RequestHandler):
 class LoadNewTitles(webapp.RequestHandler):
 	def __init__(self):
 		self.api = AnyClipAPI("CAD58B9E-F045-492F-81B9-22CFE6B00604")
-		self.imdbPattern = re.compile("<td class=\"nm\"><a href=\"/name/nm[0-9]{7}/\" onclick=\"\\(new Image\\(\\)\\)\\.src='/rg/castlist/position-[0-9]{1,2}/images/b\\.gif\\?link=/name/nm[0-9]{7}/';\">([^><]+)</a></td><td class=\"ddd\"> \\.\\.\\. </td><td class=\"char\">(?:<a href=\"/character/ch[0-9]{7}/\">([^><]+)</a>[^><]*|([^><]*))</td>")
 	
 	def getCast(self, imdbCode):
 		imdbList = fetch("http://www.imdb.com/title/tt%s/fullcredits" % imdbCode)
@@ -73,11 +72,20 @@ class LoadNewTitles(webapp.RequestHandler):
 		i = 0
 		actors = []
 		characters = []
-		for match in self.imdbPattern.finditer(imdbList.content):
-			actors.append(match.group(1))
-			character = match.group(2)
-			if character == None:
-				character = match.group(3)
+		index = 0
+		htmlRemove = re.compile("<.*?>")
+		while True:
+			index = imdbList.content.find("<td class=\"nm\">", index) + 15
+			end = imdbList.content.find("</td>", index)
+			if index < 15 or end < 0:
+				break
+			actor = htmlRemove.sub("", imdbList.content[index:end])
+			index = imdbList.content.find("<td class=\"ddd\"> ... </td><td class=\"char\">", end) + 43
+			end = imdbList.content.find("</td>", index)
+			if index < 43 or end < 0:
+				break
+			character = htmlRemove.sub("", imdbList.content[index:end])
+			actors.append(actor)
 			characters.append(character)
 			i += 1
 			if i == 10: