aggiunto monitoraggio accessi

app.py

@@ -1,3 +1,5 @@
+from modules.utilities import logger
+import time
 import gradio as gr
 import cv2
 import os
@@ -37,42 +39,95 @@ if not os.path.exists(demo_csv_path):
     with open(demo_csv_path, "w") as f:
         f.write(csv_content)
 
-def forecast_logic(file):
+def forecast_logic(file, request: gr.Request):
+    start_time = time.time()
+    
     if file is None: raise gr.Error("Seleziona un file CSV")
     if isinstance(file, list): file = file[0]
     
     img, text = forecast.predict_workload(file)
+    
     if img is None: raise gr.Error(text)
+    
+    elapsed_time = time.time() - start_time
+    logger.log_interaction(
+        request=request,
+        module_name="AI Forecaster",
+        action="Prediction",
+        input_data=file,
+        execution_time=elapsed_time
+    )
     return img, text
 
-def binary_classification(text):
-    if text.strip(): return binary(text)
-    raise gr.Error('Il testo è obbligatorio!')
+def binary_classification(text, request: gr.Request):
+    start_time = time.time()
+    
+    if not text.strip(): 
+        raise gr.Error('Il testo è obbligatorio!')
+    result = binary(text)
+
+    elapsed_time = time.time() - start_time
+    logger.log_interaction(
+        request=request,
+        module_name="Sentiment Analysis (BPO)",
+        action="Prediction",
+        input_data=text,
+        execution_time=elapsed_time
+    )
+    return result
+
+def multi_classification(text, request: gr.Request):
+    start_time = time.time()
+    
+    if not text.strip():
+        raise gr.Error('Il testo è obbligatorio!')  
+
+    try: 
+        result = multi(text)
+
+        elapsed_time = time.time() - start_time
+        logger.log_interaction(
+            request=request,
+            module_name="Smart Content Tagger",
+            action="Prediction",
+            input_data=text,
+            execution_time=elapsed_time
+        )
+        return result
+        
+    except Exception as e: 
+        raise gr.Error(f'Errore nel modello: {str(e)}')
+
+def image_classification(img, request: gr.Request):
+    start_time = time.time()
+    result = image(img)    
+    elapsed_time = time.time() - start_time
+    logger.log_interaction(
+        request=request,
+        module_name="Chest X-Ray",
+        action="Prediction",
+        input_data=img, 
+        execution_time=elapsed_time
+    )
+    return result
+
+def retina_classification(retina, request: gr.Request):
+    start_time = time.time()
+    result = retina_detector(retina)
+    elapsed_time = time.time() - start_time
+    logger.log_interaction(
+        request=request,
+        module_name="Diabetic Retinopathy",
+        action="Prediction",
+        input_data=retina,
+        execution_time=elapsed_time
+    )
+    return result
 
-def multi_classification(text):
-    if text.strip():
-        try: return multi(text)
-        except Exception as e: raise gr.Error(f'Errore nel modello: {str(e)}')
-    raise gr.Error('Il testo è obbligatorio!')  
+def bpo_dispatch_logic(text, request: gr.Request):
 
-def file_change(file):
-    if isinstance(file, list): file = file[0]
-    if file: return cv2.imread(file)
-    return None
-
-def image_classification(img):
-    if img is not None: return image(img)
-    raise gr.Error('L\'immagine è obbligatoria!')
-
-def retina_classification(retina):
-    if retina is not None: return retina_detector(retina)
-    raise gr.Error('L\'immagine è obbligatoria!')
-
-def bpo_dispatch_logic(text):
-    """
-    Funzione Ponte: Chiama il modulo AI e decide l'azione di business.
-    Restituisce un aggiornamento COMPLETO del componente NER per pulire la grafica.
-    """
+    start_time = time.time()
+    
     try:
         intent, urgency, entities = predict_bpo_ticket(text)
         
@@ -91,6 +146,15 @@ def bpo_dispatch_logic(text):
 
         html_output = utils.render_ner_html(entities)
 
+        elapsed_time = time.time() - start_time
+        logger.log_interaction(
+            request=request,
+            module_name="BPO Dispatcher",
+            action="Prediction",
+            input_data=text,
+            execution_time=elapsed_time
+        )
+
         return intent, urgency, action, html_output
 
     except Exception as e:
@@ -273,7 +337,7 @@ with gr.Blocks(title="NGT AI Platform", theme=theme, css_paths="style.css") as d
             
             # 4. COLLEGAMENTO FUNZIONE
             sentiment_btn.click(
-                fn=binary, 
+                fn=binary_classification, 
                 inputs=sentiment_input, 
                 outputs=sentiment_output
             )
@@ -322,7 +386,7 @@ with gr.Blocks(title="NGT AI Platform", theme=theme, css_paths="style.css") as d
                         # Usiamo un Label con top_classes=5 per vedere la distribuzione completa
                         multi_output = gr.Label(num_top_classes=5, label="Confidenza del Modello")
                     
-            analyze_btn_multi.click(multi, inputs=multi_input, outputs=multi_output)
+            analyze_btn_multi.click(multi_classification, inputs=multi_input, outputs=multi_output)
         
         # --- Chest X-Ray Diagnostics ---
         with gr.Tab("🩻 Chest X-Ray Diagnostics") as tab_xray:
@@ -351,7 +415,7 @@ with gr.Blocks(title="NGT AI Platform", theme=theme, css_paths="style.css") as d
                         gr.Markdown("#### 📋 Referto AI", elem_classes="h4-margin")
                         output_label = gr.Label(num_top_classes=4, label="Probabilità Patologia")
                     
-            analyze_btn_img.click(image, inputs=image_input, outputs=output_label)
+            analyze_btn_img.click(image_classification, inputs=image_input, outputs=output_label)
 
         # --- Diabetic Retinopathy ---
         with gr.Tab("👁️ Diabetic Retinopathy") as tab_retina:
@@ -384,7 +448,7 @@ with gr.Blocks(title="NGT AI Platform", theme=theme, css_paths="style.css") as d
                         output_dr_prob = gr.Label(label="Livello di Confidenza (Rischio)")
                     
             analyze_btn_dr.click(
-                retina_detector, 
+                retina_classification, 
                 inputs=image_input_dr, 
                 outputs=[output_dr_diagnosis, output_dr_prob]
             )

modules/utilities/logger.py

@@ -0,0 +1,76 @@
+import os
+import uuid
+import csv
+import threading
+from datetime import datetime
+from pathlib import Path
+from huggingface_hub import CommitScheduler
+import gradio as gr
+
+# --- CONFIGURAZIONE ---
+DATASET_REPO_ID = "NextGenTech/ngt-ai-platform-logs" 
+LOG_DIR = Path("data/logs")
+LOG_FILE = LOG_DIR / "access_logs.csv"
+TIME = 5 #intervallo di scrittura dei log sul dataset
+
+LOG_DIR.mkdir(parents=True, exist_ok=True)
+
+if not LOG_FILE.exists():
+    with open(LOG_FILE, "w", newline="", encoding="utf-8") as f:
+        writer = csv.writer(f)
+        writer.writerow([
+            "timestamp", "session_id", "module", "action", 
+            "ip_address", "user_agent", "language", "input_size", "processing_time"
+        ])
+
+scheduler = CommitScheduler(
+    repo_id=DATASET_REPO_ID,
+    repo_type="dataset",
+    folder_path=LOG_DIR,
+    path_in_repo="logs",
+    every=TIME,  
+    token=os.environ.get("HF_TOKEN_WRITE") 
+)
+
+def log_interaction(request: gr.Request, module_name: str, action: str, input_data=None, execution_time=0.0):
+    """
+    Registra un evento di analytics in modo invisibile.
+    """
+    try:
+
+        if request:
+            headers = request.headers
+            ip = headers.get("x-forwarded-for", request.client.host)
+            user_agent = headers.get("user-agent", "Unknown")
+            language = headers.get("accept-language", "Unknown").split(',')[0]
+        else:
+            ip, user_agent, language = "LOCAL", "Dev-Mode", "it"
+
+        session_raw = f"{ip}{user_agent}{datetime.now().date()}"
+        session_id = str(uuid.uuid5(uuid.NAMESPACE_DNS, session_raw))[:8]
+
+        input_meta = "0"
+        if isinstance(input_data, str):
+            input_meta = f"{len(input_data)} chars"
+        elif hasattr(input_data, 'shape'): # Immagini numpy
+            input_meta = f"{input_data.shape}"
+        elif input_data is not None:
+            input_meta = "Binary/File"
+
+        with scheduler.lock:
+            with open(LOG_FILE, "a", newline="", encoding="utf-8") as f:
+                writer = csv.writer(f)
+                writer.writerow([
+                    datetime.now().isoformat(),
+                    session_id,
+                    module_name,
+                    action,
+                    ip, 
+                    user_agent,
+                    language,
+                    input_meta,
+                    f"{execution_time:.4f}s"
+                ])
+                
+    except Exception as e:
+        print(f"⚠️ Errore durante il logging: {e}")