grade as an int

self_hosted_env/voice_agent/apply_test_schema.py

@@ -7,7 +7,7 @@ CREATE TABLE IF NOT EXISTS students (
     id SERIAL PRIMARY KEY,
     student_id VARCHAR(50) UNIQUE NOT NULL,
     student_name VARCHAR(100),
-    grade VARCHAR(20),
+    grade INTEGER NOT NULL,
     language BOOLEAN,
     nationality VARCHAR(20) NOT NULL DEFAULT 'EGYPTIAN'
 );
@@ -26,13 +26,15 @@ CREATE TABLE IF NOT EXISTS chat_history (
 CREATE INDEX IF NOT EXISTS idx_chat_history_student_id ON chat_history(student_id);
 CREATE INDEX IF NOT EXISTS idx_chat_history_created_at ON chat_history(created_at);
 CREATE INDEX IF NOT EXISTS idx_students_nationality ON students(nationality);
+CREATE INDEX IF NOT EXISTS idx_students_grade ON students(grade);
+CREATE INDEX IF NOT EXISTS idx_students_grade_language ON students(grade, language);
 
 -- Insert dummy data for testing
 INSERT INTO students (student_id, student_name, grade, language, nationality) VALUES 
-    ('student_001', 'Ahmed Ali', 'prime4', TRUE, 'EGYPTIAN'),
-    ('student_002', 'Sara Hassan', 'prime6', FALSE, 'SAUDI'),
-    ('student_003', 'Mona Adel', 'prime5', TRUE, 'EGYPTIAN'),
-    ('student_004', 'Omar Youssef', 'prime6', FALSE, 'SAUDI')
+    ('student_001', 'Ahmed Ali', 4, TRUE, 'EGYPTIAN'),
+    ('student_002', 'Sara Hassan', 6, FALSE, 'SAUDI'),
+    ('student_003', 'Mona Adel', 5, TRUE, 'EGYPTIAN'),
+    ('student_004', 'Omar Youssef', 6, FALSE, 'SAUDI')
 ON CONFLICT (student_id) DO NOTHING;
 """
 
@@ -100,7 +102,7 @@ def setup_database(drop_existing_tables: bool = False):
 
 if __name__ == "__main__":
     # To run with a clean slate, pass True
-    # setup_database(drop_existing_tables=True)
+    setup_database(drop_existing_tables=True)
     
     # To run without dropping tables (default)
-    setup_database()
\ No newline at end of file
+    #setup_database()
\ No newline at end of file

self_hosted_env/voice_agent/insert_csv_embeddings.py

@@ -26,7 +26,7 @@ def create_schema_and_table(conn, drop_existing_table: bool):
     create_table = """
     CREATE TABLE IF NOT EXISTS educational_chunks (
       id SERIAL PRIMARY KEY,
-      grade TEXT NOT NULL,
+      grade INTEGER NOT NULL,
       subject TEXT,
       unit TEXT,
       concept TEXT,
@@ -64,6 +64,40 @@ def create_schema_and_table(conn, drop_existing_table: bool):
     conn.commit()
     cur.close()
 
+def extract_grade_number(grade_value) -> int:
+    """Extract numeric grade from various formats"""
+    if pd.isna(grade_value):
+        return 1  # default
+    
+    grade_str = str(grade_value).lower().strip()
+    
+    # Try to extract number directly
+    if grade_str.isdigit():
+        return int(grade_str)
+    
+    # Handle common formats like "grade4", "prime4", "4th grade", etc.
+    import re
+    numbers = re.findall(r'\d+', grade_str)
+    if numbers:
+        return int(numbers[0])
+    
+    # Fallback mapping for word-based grades
+    grade_mapping = {
+        'one': 1, 'first': 1,
+        'two': 2, 'second': 2,
+        'three': 3, 'third': 3,
+        'four': 4, 'fourth': 4,
+        'five': 5, 'fifth': 5,
+        'six': 6, 'sixth': 6,
+    }
+    
+    for word, num in grade_mapping.items():
+        if word in grade_str:
+            return num
+    
+    # Final fallback
+    print(f"Warning: Could not parse grade '{grade_value}', defaulting to 1")
+    return 1
 
 def insert_chunks_from_csv(csv_file: str):
     df = pd.read_csv(csv_file)
@@ -95,8 +129,11 @@ def insert_chunks_from_csv(csv_file: str):
     for idx, row in df.iterrows():
         try:
             embedding = json.loads(row["Embedding"])
+            # Convert grade to integer
+            grade_int = extract_grade_number(row["Grade"])
+            
             buffer.append((
-                str(row["Grade"]),
+                grade_int,  # Now properly converted to integer
                 row["Subject"],
                 row.get("Unit"),
                 row.get("Concept"),
@@ -161,7 +198,7 @@ def setup_embeddings_database(drop_existing_tables: bool = False):
 
 if __name__ == "__main__":
     # To run with a clean slate, pass True
-    # setup_embeddings_database(drop_existing_tables=True)
+    setup_embeddings_database(drop_existing_tables=True)
     
     # To run without dropping the table (default)
-    setup_embeddings_database()
\ No newline at end of file
+    #setup_embeddings_database()
\ No newline at end of file

self_hosted_env/voice_agent/services/chat_database_service.py

@@ -43,8 +43,8 @@ class ChatDatabaseService:
             if result:
                 return {
                     'student_id': result['student_id'],
-                    'student_name': result['student_name'],  # Added this line
-                    'grade': result['grade'],
+                    'student_name': result['student_name'],
+                    'grade': result['grade'],  # This is now an integer
                     'is_arabic': result['language'],  # Convert language boolean to is_arabic
                     'nationality': result['nationality']
                 }
@@ -150,7 +150,7 @@ class ChatDatabaseService:
                 )
                 self.conn.commit()
 
-    def create_student(self, student_id: str, student_name: str, grade: str, 
+    def create_student(self, student_id: str, student_name: str, grade: int, 
                       language: bool, nationality: str = 'EGYPTIAN'):
         """Create a new student record"""
         with self.conn.cursor() as cur:

self_hosted_env/voice_agent/services/pgvector_service.py

@@ -117,33 +117,18 @@ class PGVectorService:
             )
             return cur.fetchall()
 
-    def get_subjects_by_grade_and_language(self, grade: str, is_arabic: bool) -> List[str]:
+    def get_subjects_by_grade_and_language(self, grade: int, is_arabic: bool) -> List[str]:
         """Get available subjects for a specific grade and language"""
         with self.conn.cursor(cursor_factory=RealDictCursor) as cur:
-            # Extract numeric part from grade string
-            grade_number = ''.join(filter(str.isdigit, grade)) if grade else None
-            
-            if grade_number:
-                cur.execute(
-                    """
-                    SELECT DISTINCT subject
-                    FROM educational_chunks
-                    WHERE grade = %s AND is_arabic = %s
-                    ORDER BY subject;
-                    """,
-                    (int(grade_number), is_arabic)
-                )
-            else:
-                # Fallback if grade parsing fails
-                cur.execute(
-                    """
-                    SELECT DISTINCT subject
-                    FROM educational_chunks
-                    WHERE is_arabic = %s
-                    ORDER BY subject;
-                    """,
-                    (is_arabic,)
-                )
+            cur.execute(
+                """
+                SELECT DISTINCT subject
+                FROM educational_chunks
+                WHERE grade = %s AND is_arabic = %s
+                ORDER BY subject;
+                """,
+                (grade, is_arabic)
+            )
             return [row['subject'] for row in cur.fetchall()]
 
     def close(self):

self_hosted_env/voice_agent/static/audio-recorder.html

@@ -187,8 +187,8 @@
     <script>
         // Configuration
         const Config = {
-            BACKEND_URL: "http://teamtestingdocker.caprover.al-arcade.com:8000/chat",
-            AUDIO_RESPONSE_URL: "http://teamtestingdocker.caprover.al-arcade.com:8000/get-audio-response"
+            BACKEND_URL: "http://localhost:8000/chat",
+            AUDIO_RESPONSE_URL: "http://localhost:8000/get-audio-response"
         };
 
         // Enums