Tweak spam algorithm to flag more bad accounts based on keywords in u…

…sernames
Sefaria · Mar 23, 2022 · c2cb263 · c2cb263
1 parent dba0ee6
commit c2cb263
Show file tree

Hide file tree

Showing 2 changed files with 9 additions and 3 deletions.
diff --git a/sefaria/views.py b/sefaria/views.py
@@ -867,11 +867,14 @@ def profile_spam_dashboard(request):
 
         regex = r'.*(?!href=[\'"](\/|http(s)?:\/\/(www\.)?sefaria).+[\'"])(href).*'
 
+        spam_keywords_regex = r'(?i).*support.*|.*coin.*|.*helpline.*'
+
         users_to_check = db.profiles.find(
             {'$or': [
                 {'website': {"$ne": ""}, 'bio': {"$ne": ""}, "id": {"$gt": earliest_new_user_id},
                       "reviewed": {"$ne": True}},
-                {'bio': {"$regex": regex}, "id": {"$gt": earliest_new_user_id}, "reviewed": {"$ne": True}}
+                {'bio': {"$regex": regex}, "id": {"$gt": earliest_new_user_id}, "reviewed": {"$ne": True}},
+                {'slug': {"$regex": spam_keywords_regex}, "id": {"$gt": earliest_new_user_id}, "reviewed": {"$ne": True}}
             ]
         })
 
@@ -882,7 +885,9 @@ def profile_spam_dashboard(request):
         for user in users_to_check:
             history_count = db.user_history.find({'uid': user['id']}).count()
             if history_count < 10:
-                profiles_list.append({"id": user["id"], "slug": user["slug"], "bio": strip_tags(user["bio"][0:250]), "website": user["website"][0:50]})
+                profile = model.user_profile.UserProfile(id=user["id"])
+
+                profiles_list.append({"name": f"{profile.first_name} {profile.last_name}", "email": profile.email, "id": user["id"], "slug": user["slug"], "bio": strip_tags(user["bio"][0:250]), "website": user["website"][0:50]})
 
         return render_template(request, 'spam_dashboard.html', None, {
             "title": "Potential Spam Profiles since %s" % date.strftime("%Y-%m-%d"),

diff --git a/templates/spam_dashboard.html b/templates/spam_dashboard.html
@@ -71,7 +71,8 @@ <h1>{{ title }}</h1>
                     <tbody>
                     {% for profile in profiles %}
                         <tr>
-                            <td><a href='/profile/{{ profile.slug }}' target='_blank'>{{ profile.slug }}</a></td>
+                            <td><a href='/profile/{{ profile.slug }}' target='_blank'>{{ profile.name }}</a></td>
+                            <td>{{ profile.email }}</td>
                             <td>{{ profile.bio }}</td>
                             <td>{{ profile.website }}</td>
                             <td><input type='checkbox' name='spam_profiles[]' value='{{ profile.id }}' form='spam_form'/></td>