From 1920a127599721dd927da984f6d2b157ef1dcbab Mon Sep 17 00:00:00 2001
From: kad <kad@irc>
Date: Fri, 30 Jul 2010 21:33:57 -0600
Subject: [PATCH] Generalize translation code in GoogleTranslate.py

---
 modules/GoogleTranslate.py | 34 +++++++++++++++++++++++++++-------
 1 file changed, 27 insertions(+), 7 deletions(-)

diff --git a/modules/GoogleTranslate.py b/modules/GoogleTranslate.py
index bcb35d7..c57aa38 100644
--- a/modules/GoogleTranslate.py
+++ b/modules/GoogleTranslate.py
@@ -16,6 +16,8 @@
 
 from urllib2 import urlopen
 from urllib import urlencode
+from htmlentitydefs import name2codepoint
+import re
 
 from extlib import irclib
 
@@ -49,13 +51,31 @@ class GoogleTranslate(Module):
             translation = translation.decode('utf-8')
 
             # do some text conversion
-            translation = translation.replace('\\u0026quot;', '"')
-            translation = translation.replace('\\u0026amp;', '&')
-            translation = translation.replace('\\u003c', '<')
-            translation = translation.replace('\\u0026lt;', '<')
-            translation = translation.replace('\\u003e', '>')
-            translation = translation.replace('\\u0026gt;', '>')
-            translation = translation.replace('\\u0026#39;', '\'')
+            for i in re.findall('\u(\w{4})', translation):
+                replacement = unichr(int(i, 16))
+                translation = translation.replace('\u' + i, replacement)
+
+            for i in re.findall('&(#)?(x)?(\w+?);', translation):
+                replacement = i[2]
+                if i[0] == "#":
+                    if i[1] == 'x':
+                        replacement = unichr(int(i[2], 16))
+                    else:
+                        replacement = unichr(int(i[2]))
+                else:
+                    cp = name2codepoint.get(i[2])
+                    replacement = unichr(cp)
+                translation = translation.replace('&' + i[0] + i[1] + i[2] + ';', replacement)
+
+            translation = translation.replace('\\\\', '\\')
+
+            #translation = translation.replace('\\u0026quot;', '"')
+            #translation = translation.replace('\\u0026amp;', '&')
+            #translation = translation.replace('\\u003c', '<')
+            #translation = translation.replace('\\u0026lt;', '<')
+            #translation = translation.replace('\\u003e', '>')
+            #translation = translation.replace('\\u0026gt;', '>')
+            #translation = translation.replace('\\u0026#39;', '\'')
 
             return self.reply(connection, replypath, translation)