diff --git a/wm_metrics/category_flattening.py b/wm_metrics/category_flattening.py
deleted file mode 100644
index 146f569..0000000
--- a/wm_metrics/category_flattening.py
+++ /dev/null
@@ -1,24 +0,0 @@
-#!/usr/bin/python
-# -*- coding: utf-8  -*-
-""" wmflabs_queries.py regroups query builder functions in order to generate queries for wmflabs databases.
-"""
-
-
-def category_flatt(category):
-    """
-    category_flattening
-            List all subcategories of given category
-    """
-    return """SELECT page.page_title
-	FROM categorylinks
-	JOIN page ON page.page_id = categorylinks.cl_from
-	WHERE page.page_namespace = 14 AND categorylinks.cl_to = '%s";""" % (category)
-
-
-def main():
-    categories = category(
-        "Content_media_by_years_and_quarters_-_Supported_by_Wikimedia_France_-_2014_-_2")
-    print categories
-
-if __name__ == "__main__":
-    main()
diff --git a/wm_metrics/category_induced.py b/wm_metrics/category_induced.py
new file mode 100644
index 0000000..79a9dd1
--- /dev/null
+++ b/wm_metrics/category_induced.py
@@ -0,0 +1,103 @@
+# -*- coding: utf-8 -*-
+#!/usr/bin/python
+
+import mw_api, mw_util, json, codecs, MySQLdb, operator
+
+class CategoryInduced:
+    def __init__(self, category):
+        self.commons = mw_api.MwWiki(url_api='https://commons.wikimedia.org/w/api.php')
+        self.category = category.replace(" ", "_").decode('utf-8')
+        self.categories = []
+        self.first_images = []
+        """"DB instantiation"""
+        self.db = MySQLdb.connect(host="commonswiki.labsdb", db="commonswiki_p", read_default_file="~/replica.my.cnf", charset='utf8')
+        self.cursor = self.db.cursor()
+        #TODO: prendre que des images
+        self.query = """SELECT page.page_title
+                        FROM page
+                        JOIN categorylinks ON page.page_id = categorylinks.cl_from
+                        WHERE categorylinks.cl_to = %s
+                        ORDER BY categorylinks.cl_timestamp ASC
+                        LIMIT 1;"""
+
+ 
+    def list_category(self):
+        import os.path
+        cache_name = "cache/%s.cache" % (self.category)
+        result = None
+        if os.path.exists(cache_name):
+            cache = codecs.open(cache_name, 'r', 'utf-8')
+            result = json.loads(cache.read())
+        else:
+            result = json.loads(self.commons.send_to_api(mw_api.MwApiQuery(
+                properties={
+                    "prop"	: "categories",
+                    "cllimit" : "max",
+                    "clir" : "ascending",
+                    "generator" : "categorymembers",
+                    "gcmtitle" : self.category,
+                    "gcmprop" : "title",
+                    "gcnamespace" : "6",
+                    "gcmlimit" : "max",
+                })))
+        dic = result[u'query'][u'pages']
+        list = sorted(dic.iteritems(), reverse=False, key=operator.itemgetter(1))
+        liste2 = [x[1][u'categories'] for x in list if u'categories' in x[1].keys()]
+        resu = set()
+        for l in liste2:
+            resu.update([x[u'title'] for x in l])
+        return resu
+
+    def first_image(self, category):
+        self.catsql = category[9:].replace(" ", "_")
+        self.cursor.execute(self.query, self.catsql)
+        cat_content = self.catsql.encode('utf-8')
+        first_content = [x[0].decode('utf-8') for x in self.cursor.fetchall()]
+        res = {'cat': cat_content, 'first': first_content}
+  #      print res
+        return res
+
+    def list_images(self):
+        import os.path
+        cache_name = "cache/%s.cache" % (self.category)
+        result = json.loads(self.commons.send_to_api(mw_api.MwApiQuery(
+                properties={
+                    "list"         : "categorymembers",
+                    "cmtitle"      : self.category,
+                    "cmprop"       : "title",
+                    "cmlimit"      : "max",
+                })))
+ #       print self.category 
+        res1 = [x[u'title'] for x in result[u'query'][u'categorymembers']]
+        res = [x.encode('utf-8') for x in res1]
+ #       print "List of images"
+ #       print res
+	return res
+
+def main():
+    from argparse import ArgumentParser
+    description = "Computes metrics about a commons category"
+    parser = ArgumentParser(description=description)
+    parser.add_argument("-c", "--category",
+        type=str,
+        dest="category",
+        metavar="CAT",
+        required=True,
+        help="The category on which we compute metrics")
+    args = parser.parse_args()
+    ci = CategoryInduced(mw_util.str2cat(args.category))
+    ci.categories = ci.list_category()
+    first_images = [ci.first_image(x) for x in ci.categories]
+    print "--------------------first images--------------------"
+    print first_images
+    images = [x.decode('utf-8')[5:] for x in ci.list_images()]
+    print "----------------------images------------------------"
+    print images
+    result = [first_images[x]['cat'] for x in range(len(first_images)) if first_images[x]['first'][0] in images]
+    print "----------------------result------------------------"
+    print first_images[1]['first'][0].encode('utf-8')
+    print first_images[1]['first'][0].encode('utf-8') in images
+    print result
+
+if __name__ == "__main__":
+    main()