Update app.py

app.py

@@ -30,21 +30,34 @@ ga = BetaAnalyticsDataClient(credentials=creds)
 # -------------------------
 # FETCH ALL PAGE PATHS
 # -------------------------
-def fetch_all_page_paths(start, end):
+@lru_cache(maxsize=3)
+def fetch_top_page_paths(start, end, limit=300):
+    """
+    Fetch TOP page paths by traffic and cache results.
+    """
     req = RunReportRequest(
         property=f"properties/{GA4_PROPERTY_ID}",
         dimensions=[{"name": "pagePath"}],
         metrics=[{"name": "screenPageViews"}],
         date_ranges=[{"start_date": start, "end_date": end}],
-        limit=5000
+        order_bys=[{
+            "metric": {"metric_name": "screenPageViews"},
+            "desc": True
+        }],
+        limit=limit
     )
+
     res = ga.run_report(req)
 
-    pages = sorted(
-        {r.dimension_values[0].value for r in res.rows if r.dimension_values[0].value}
-    )
+    pages = sorted({
+        r.dimension_values[0].value
+        for r in res.rows
+        if r.dimension_values[0].value
+    })
+
     return pages
 
+
 # -------------------------
 # FETCH JOURNEY STEPS
 # -------------------------
@@ -180,33 +193,99 @@ def run_journey(start, end, landing_page):
 # -------------------------
 # UI
 # -------------------------
+# -------------------------
+# UI — JOURNEY EXPLORER
+# -------------------------
 with gr.Blocks() as app:
     gr.Markdown("## 🚀 Customer Journey Explorer (GA4)")
+    gr.Markdown(
+        """
+        Analyze **end-to-end customer journeys**:
+        - Where users **started**
+        - Which pages they **navigated through**
+        - Where they **exited**
+        - Whether they **submitted any form before exit** (Yes / No)
+        """
+    )
 
+    # -------------------------
+    # DATE CONTROLS
+    # -------------------------
     with gr.Row():
-        sd = gr.Textbox(value=str(date.today() - timedelta(days=30)), label="Start Date")
-        ed = gr.Textbox(value=str(date.today()), label="End Date")
+        sd = gr.Textbox(
+            label="Start Date (YYYY-MM-DD)",
+            value=str(date.today() - timedelta(days=30))
+        )
+        ed = gr.Textbox(
+            label="End Date (YYYY-MM-DD)",
+            value=str(date.today())
+        )
 
-    page_dd = gr.Dropdown(label="Select Starting Page", choices=[])
+    # -------------------------
+    # PAGE SELECTION
+    # -------------------------
+    with gr.Row():
+        page_dd = gr.Dropdown(
+            label="Starting Page Path",
+            choices=[],
+            allow_custom_value=True,
+            interactive=True,
+            info="Loads top pages by traffic. You can also type a custom page path."
+        )
 
-    load_btn = gr.Button("Load All Page Paths")
-    run_btn = gr.Button("Analyze Journey", variant="primary")
+    with gr.Row():
+        load_pages_btn = gr.Button(
+            "⚡ Load Top Pages (Fast)",
+            variant="secondary"
+        )
+        run_journey_btn = gr.Button(
+            "Analyze Journey",
+            variant="primary"
+        )
+
+    # -------------------------
+    # OUTPUTS
+    # -------------------------
+    gr.Markdown("### 📊 Journey Visualization")
+    journey_chart = gr.Plot()
+
+    gr.Markdown("### 🔁 Exit Pages")
+    exit_df = gr.Dataframe(
+        interactive=False,
+        wrap=True
+    )
 
-    chart = gr.Plot()
-    exit_df = gr.Dataframe()
-    summary = gr.Markdown()
+    gr.Markdown("### 🧠 Journey Summary")
+    journey_summary = gr.Markdown()
 
-    load_btn.click(
-        lambda s, e: gr.update(choices=fetch_all_page_paths(s, e)),
+    # -------------------------
+    # BUTTON ACTIONS
+    # -------------------------
+
+    # Load top page paths (cached + limited)
+    load_pages_btn.click(
+        fn=lambda s, e: gr.update(
+            choices=fetch_top_page_paths(s, e),
+            value=None
+        ),
         inputs=[sd, ed],
         outputs=page_dd
     )
 
-    run_btn.click(
-        run_journey,
+    # Run full journey analysis
+    run_journey_btn.click(
+        fn=run_journey,
         inputs=[sd, ed, page_dd],
-        outputs=[chart, exit_df, summary]
+        outputs=[journey_chart, exit_df, journey_summary]
     )
 
+# -------------------------
+# APP LAUNCH
+# -------------------------
 if __name__ == "__main__":
-    app.launch(server_name="0.0.0.0", server_port=7860)
+    app.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        show_api=False,
+        show_error=False
+    )