Fixed regex so C++ and Objective-C are matched. Rendering html entiti…

…es as

ascii in results. Updated readme with new results

github-illiteracy-index/README

@@ -7,13 +7,15 @@ At the time of writing:
 
 #  Language    Illiteracy  
 1  C           0.02877583  
-2  Perl        0.01635618  
-3  Ruby        0.01560477  
-4  JavaScript  0.01330989  
-5  Shell       0.01235425  
-6  Python      0.01046104  
-7  PHP         0.00910218  
-8  Java        0.00736395  
+2  C++         0.01682951  
+3  Perl        0.01635618  
+4  Ruby        0.01560477  
+5  JavaScript  0.01330989  
+6  Shell       0.01235425  
+7  Python      0.01046104  
+8  Objective-C 0.00941299  
+9  PHP         0.00910218  
+10 Java        0.00736395  
 
 I'm as atonished as you are. C programmers come out worst.. proving to be more
 than twice as lexically challenged as the PHP community!

github-illiteracy-index/illiteracy.py

@@ -7,7 +7,7 @@
 SEARCH_URL  = 'https://github.com/search?type=Code&language=%s&q=%s'
 
 SEARCH_REGEX = r'<div class="title">Code \(([0-9]+)\)</div>'
-LANGS_REGEX  = r'<a href="/languages/([\w]*)"\s*class="bar"\s*style="width: [0-9]*%">\s*([0-9]*)%</a>'
+LANGS_REGEX  = r'<a href="/languages/([^\b"]*)"\s*class="bar"\s*style="width: [0-9]*%">\s*([0-9]*)%</a>'
 
 TERMS = {'length': 'lenght', 'height': 'heigth', 'hierarchy':'heirarchy'}
 
@@ -96,7 +96,7 @@ def printSorted(sorted):
     i = 0
     for result in sorted:
         i += 1
-        print format % (i, result[0], round(result[1], 8))
+        print format % (i,  urllib2.unquote(result[0]), round(result[1], 8))
 
 #langs = {'JavaScript': 20.0}
 langs      = getLangs()