diff --git a/bigdata/src/G_H_postviews.py b/bigdata/src/G_H_postviews.py
new file mode 100644
index 0000000..576eb13
--- /dev/null
+++ b/bigdata/src/G_H_postviews.py
@@ -0,0 +1,50 @@
+from mrjob.job import MRJob
+from mrjob.step import MRStep
+import re
+
+#main class
+class MRmostViews(MRJob):
+    def steps(self):
+        return [
+            MRStep(
+                mapper=self.get_posts_views, 
+                reducer=self.post_filter
+            )
+        ]
+
+    #get posts ands views
+    def get_posts_views(self, _, line):
+       
+        get_id = r'\sId="(.*?)"'
+        get_views = r'\sViewCount="(.*?)"'
+        id = re.search(get_id, line)
+        views = re.search(get_views, line)
+        # output <re.Match object; span=(4, 12), match=' Id="11"'>
+        # output <re.Match object; span=(98, 115), match=' ViewCount="1836"'>
+
+        #group() documentation: https://www.geeksforgeeks.org/re-matchobject-group-function-in-python-regex/
+        if id:
+            post_id = id.group(1)
+        else:
+            return ""
+        if views:
+            post_views = int(views.group(1).replace('"', ""))
+        else:
+            return 0
+        # output "11"
+        # output 1836
+
+        yield None, {"id": post_id, "count": post_views}
+    
+    #get posts, sort them and return top 10
+    def post_filter(self, _, line):
+        posts = list(line)
+        #sort posts
+        posts = sorted(posts, key= lambda x: x.get("count"), reverse=True)[:10]
+        #return top 10
+        for i in posts:
+            yield i.values()
+
+if __name__ == "__main__":
+    MRmostViews.run()
+    
\ No newline at end of file