Update app.py

app.py

@@ -9,15 +9,15 @@ import plotly.express as px
 from plotly.subplots import make_subplots
 import plotly.graph_objects as go
 from datetime import date, timedelta
-from huggingface_hub import HfApi # ADDED: For permanent saving to Hugging Face
+from huggingface_hub import HfApi 
 
 # =====================================================
 # CONFIG / SECRETS
 # =====================================================
 API_TOKEN = os.getenv("LEADFEEDER_API_TOKEN")
 APP_PASSWORD = os.getenv("APP_PASSWORD")
-HF_TOKEN = os.getenv("HF_TOKEN") # ADDED: Your Hugging Face write token
-SPACE_ID = os.getenv("SPACE_ID") # ADDED: Automatically provided by HF Spaces
+HF_TOKEN = os.getenv("HF_TOKEN")
+SPACE_ID = os.getenv("SPACE_ID") 
 
 if not API_TOKEN or not APP_PASSWORD:
     print("⚠️ WARNING: Secrets missing. App will launch but API calls will fail.")
@@ -27,45 +27,36 @@ if not API_TOKEN or not APP_PASSWORD:
 ACCOUNT_ID = "255333" 
 BASE_URL = "https://api.leadfeeder.com"
 PAGE_SIZE = 100 
+
 HEADERS = {
     "Authorization": f"Token token={API_TOKEN}",
     "Accept": "application/json"
 }
+
 CAMPAIGN_CONFIG_FILE = "campaign_rules.json"
 
 # =====================================================
 # 1. CAMPAIGN MANAGER & LOGIC
 # =====================================================
-DEFAULT_CAMPAIGNS = {
-    "MS Tech": [
-        "microsoft", "power", "dynamics", "crm", "365", "xamarin", "sharepoint", 
-        "dot", ".net", "asp.net", "azure", "copilot", "real-estate-website-development", 
-        "manufacturing-production", "legacy-hrms", "healthcare-appointment", "eld-fleet",
-        "legacy-erp", "clinical-trial", "digital-e-learning", "bi-analytics", 
-        "application-modernization", "data-warehouse", "workforce", "contingent"
-    ],
-    "Fintech": [
-        "fintech", "finance", "money", "credit", "debit", "card", "invest", "bank", 
-        "wallet", "account", "bookkeeping", "payment", "loan", "nfc", "lend", "debt", 
-        "tax", "invoice", "quot", "borrow", "bill", "coin", "currency", "aml", "kyc", 
-        "blockchain", "cryptocurrency", "budget", "expense", "wealth", "sox", "sarbanes"
-    ],
-    "Adtech": [
-        "programmatic", "publisher", "media", "entertainment", "trading", "ott", 
-        "email", "news", "social", "marketing", "video", "audio", "radio", "livestream", 
-        "bid", "anime", "dooh", "vod", "rtb", "supply-side-platform", "demand-side-platform", 
-        "data-management-platform"
-    ]
-}
-
 def load_campaign_rules():
+    """Loads rules from JSON. Auto-migrates old list-based format to new include/exclude format."""
     if os.path.exists(CAMPAIGN_CONFIG_FILE):
         try:
             with open(CAMPAIGN_CONFIG_FILE, "r") as f:
-                return json.load(f)
-        except:
-            return DEFAULT_CAMPAIGNS
-    return DEFAULT_CAMPAIGNS
+                rules = json.load(f)
+                
+                # Backward compatibility: Convert flat lists into include/exclude dicts
+                for campaign, config in rules.items():
+                    if isinstance(config, list):
+                        rules[campaign] = {
+                            "include": config,
+                            "exclude": []
+                        }
+                return rules
+        except Exception as e:
+            print(f"Error loading {CAMPAIGN_CONFIG_FILE}: {e}")
+            return {}
+    return {}
 
 def categorize_quality(score):
     if pd.isna(score): return "Unknown"
@@ -81,30 +72,42 @@ def categorize_quality(score):
 def get_campaign_match(text, rules):
     if not text or not isinstance(text, str):
         return None
+        
     text_lower = text.lower()
-    for campaign_name, keywords in rules.items():
-        for kw in keywords:
-            if kw.lower() in text_lower:
-                return campaign_name
+    
+    for campaign_name, config in rules.items():
+        includes = config.get("include", [])
+        excludes = config.get("exclude", [])
+        
+        # 1st: Check Exclusions. If an excluded word is present, skip this campaign entirely
+        has_exclusion = any(ex.lower() in text_lower for ex in excludes if ex.strip())
+        if has_exclusion:
+            continue
+            
+        # 2nd: Check Inclusions. If an included word is present, it's a match
+        has_inclusion = any(inc.lower() in text_lower for inc in includes if inc.strip())
+        if has_inclusion:
+            return campaign_name
+            
     return None
 
 def apply_business_logic(df):
     if df is None or df.empty: return df
-        
+    
     # 1. Quality Group
     if "lead_quality_score" in df.columns:
         df["Quality_Group"] = df["lead_quality_score"].apply(categorize_quality)
-            
-    # 2. Campaign Logic (Smart Fallback)
-    rules = load_campaign_rules()
         
+    # 2. Campaign Logic (Strict Fallback)
+    rules = load_campaign_rules()
+    
     def resolve_campaign(row):
         # Priority 1: Landing Page
         if row.get("landing_page_path"):
             match = get_campaign_match(row["landing_page_path"], rules)
             if match: return match
             
-        # Priority 2: Exit Page (NEW LOGIC)
+        # Priority 2: Exit Page (Only checked if Landing Page DID NOT match)
         if row.get("exit_page_path"):
             match = get_campaign_match(row["exit_page_path"], rules)
             if match: return match
@@ -118,56 +121,40 @@ def apply_business_logic(df):
         if row.get("primary_industry"):
             match = get_campaign_match(row["primary_industry"], rules)
             if match: return match
-                
+            
         return "Uncategorized"
         
     df["Campaign"] = df.apply(resolve_campaign, axis=1)
     return df
 
 # =====================================================
-# 2. PRESETS (UPDATED WITH COUNTRY/CITY/INDUSTRY)
+# 2. PRESETS
 # =====================================================
 DASHBOARD_PRESETS = {
-    # -- SPECIAL VIEWS --
-    "Key Campaigns Bifurcation (MS Tech, Fintech, Adtech)": ("SPECIAL_KEY_BIFURCATION", None, None),
+    "Key Campaigns Bifurcation (Top Campaigns)": ("SPECIAL_KEY_BIFURCATION", None, None),
     "All Campaigns Performance": ("SPECIAL_ALL_PERFORMANCE", None, None),
-    
-    # -- CAMPAIGN & QUALITY --
     "Visits by Campaign": ("Campaign", "total_visits", "sum"),
     "Leads by Campaign": ("Campaign", "company_name", "count"),
     "Leads by Quality Group": ("Quality_Group", "company_name", "count"),
-    
-    # -- GEOGRAPHY (NEW) --
     "Visits by Country": ("country", "total_visits", "sum"),
     "Leads by Country": ("country", "company_name", "count"),
     "Visits by City": ("city", "total_visits", "sum"),
     "Leads by City": ("city", "company_name", "count"),
-    
-    # -- INDUSTRY (NEW) --
     "Visits by Industry": ("primary_industry", "total_visits", "sum"),
     "Leads by Industry": ("primary_industry", "company_name", "count"),
-    
-    # -- OTHERS --
     "Top Accounts by Visits": ("company_name", "total_visits", "sum"),
 }
 
 TREND_PRESETS = {
-    # -- CAMPAIGN & QUALITY --
     "Visits Trend by Campaign": ("total_visits", "sum", "Campaign"),
     "Leads Trend by Campaign": ("company_name", "count", "Campaign"),
     "Visits Trend by Quality": ("total_visits", "sum", "Quality_Group"),
-    
-    # -- GEOGRAPHY (NEW) --
     "Visits Trend by Country": ("total_visits", "sum", "country"),
     "Leads Trend by Country": ("company_name", "count", "country"),
     "Visits Trend by City": ("total_visits", "sum", "city"),
     "Leads Trend by City": ("company_name", "count", "city"),
-    
-    # -- INDUSTRY (NEW) --
     "Visits Trend by Industry": ("total_visits", "sum", "primary_industry"),
     "Leads Trend by Industry": ("company_name", "count", "primary_industry"),
-    
-    # -- GENERAL --
     "Total Visits Trend": ("total_visits", "sum", None),
     "Active Accounts Trend": ("company_name", "count", None),
 }
@@ -281,7 +268,7 @@ def enrich_leads_with_visits(df, start_date, end_date, max_rows=None, progress=g
                     if not landing:
                         first_step = visit_route[0]
                         landing = first_step.get("page_path") or first_step.get("page_url")
-                
+                        
                 if not landing or not exit_p:
                     pv_map = {p["id"]: p["attributes"] for p in included if p["type"] == "page_views"}
                     pv_ids = [r["id"] for r in visit.get("relationships", {}).get("page_views", {}).get("data", [])]
@@ -353,8 +340,9 @@ def build_dashboard(df, preset, top_n):
     if df is None or df.empty: return px.bar(title="No Data")
     if "Campaign" not in df.columns: df = apply_business_logic(df)
     
-    if preset == "Key Campaigns Bifurcation (MS Tech, Fintech, Adtech)":
-        target = ["MS Tech", "Fintech", "Adtech"]
+    if preset == "Key Campaigns Bifurcation (Top Campaigns)":
+        rules = load_campaign_rules()
+        target = list(rules.keys())[:3] if rules else [] 
         filtered = df[df["Campaign"].isin(target)].copy()
         if filtered.empty: return px.bar(title="No Data for Key Campaigns")
         
@@ -423,17 +411,15 @@ def get_trend_filter_options(df, preset):
     metric, agg, segment = TREND_PRESETS[preset]
     if not segment: return gr.update(choices=[], value=None, visible=False)
     
-    # FORCE LOGIC
     if segment not in df.columns: 
         df = apply_business_logic(df)
         
     options = sorted(df[segment].astype(str).unique().tolist())
     
-    # Smart Defaults for specific columns
     if segment == "Campaign":
-        defaults = [o for o in options if o in ["MS Tech", "Fintech", "Adtech"]]
+        defaults = options[:3] # Pick top 3 dynamic campaigns automatically
     else:
-        defaults = options[:5] # Top 5 for City/Country/Industry
+        defaults = options[:5] 
         
     if not defaults: defaults = options[:5]
     return gr.update(choices=options, value=defaults, visible=True, label=f"Filter {segment}")
@@ -447,9 +433,10 @@ with gr.Blocks(title="Leadfeeder Campaign Pro") as demo:
     with gr.Row():
         pwd = gr.Textbox(type="password", label="App Password")
         gr.Button("Auth").click(lambda p: gr.Info("Success") if p==APP_PASSWORD else gr.Error("Invalid"), pwd, None)
-        df_state = gr.State()
-        status = gr.Markdown()
-        
+    
+    df_state = gr.State()
+    status = gr.Markdown()
+    
     with gr.Tabs():
         # --- TAB 1: DATA ---
         with gr.Tab("📋 Data & Report"):
@@ -462,6 +449,7 @@ with gr.Blocks(title="Leadfeeder Campaign Pro") as demo:
                 file_dl = gr.File(label="Download Excel")
                 
             table = gr.Dataframe(label="Preview (Top 50 Enriched)", interactive=True)
+            
             btn_load.click(load_preview, [start, end], [df_state, table, status])
             btn_dl.click(download_full_excel, [df_state, start, end], file_dl)
             
@@ -472,7 +460,7 @@ with gr.Blocks(title="Leadfeeder Campaign Pro") as demo:
             
             gr.Markdown("### 📊 Charts")
             with gr.Row():
-                preset = gr.Dropdown(choices=list(DASHBOARD_PRESETS.keys()), label="Chart View", value="Key Campaigns Bifurcation (MS Tech, Fintech, Adtech)")
+                preset = gr.Dropdown(choices=list(DASHBOARD_PRESETS.keys()), label="Chart View", value="Key Campaigns Bifurcation (Top Campaigns)")
                 top_n = gr.Slider(5, 50, value=10, label="Top N Items")
             chart = gr.Plot()
             gr.Button("Build View").click(build_dashboard, [df_state, preset, top_n], chart)
@@ -493,41 +481,49 @@ with gr.Blocks(title="Leadfeeder Campaign Pro") as demo:
             gr.Markdown("### Manage Campaign Groups")
             
             init_rules = load_campaign_rules()
-            camp_choices = list(init_rules.keys()) + ["+ Create New Campaign"]
-            default_kws = ", ".join(init_rules.get(camp_choices[0], [])) if init_rules else ""
+            camp_choices = list(init_rules.keys()) + ["+ Create New Campaign"] if init_rules else ["+ Create New Campaign"]
+            
+            default_inc = ""
+            default_exc = ""
+            if init_rules and camp_choices[0] in init_rules:
+                default_inc = ", ".join(init_rules[camp_choices[0]].get("include", []))
+                default_exc = ", ".join(init_rules[camp_choices[0]].get("exclude", []))
             
             with gr.Row():
                 camp_dropdown = gr.Dropdown(choices=camp_choices, label="Select Campaign to Edit", value=camp_choices[0])
-                new_camp_name = gr.Textbox(label="New Campaign Name", visible=False)
+                new_camp_name = gr.Textbox(label="New Campaign Name", visible=(not init_rules))
+                
+            with gr.Row():
+                inc_kw_input = gr.Textbox(label="Include Keywords (comma separated)", lines=3, value=default_inc)
+                exc_kw_input = gr.Textbox(label="Exclude Keywords (comma separated)", lines=3, value=default_exc)
                 
-            kw_input = gr.Textbox(label="Keywords (comma separated)", lines=5, value=default_kws)
-            
             def update_ui_on_select(selected_camp):
                 rules = load_campaign_rules()
                 if selected_camp == "+ Create New Campaign":
-                    return gr.update(visible=True, value=""), gr.update(value="")
+                    return gr.update(visible=True, value=""), gr.update(value=""), gr.update(value="")
                 else:
-                    kws = rules.get(selected_camp, [])
-                    return gr.update(visible=False), gr.update(value=", ".join(kws))
+                    inc_kws = rules.get(selected_camp, {}).get("include", [])
+                    exc_kws = rules.get(selected_camp, {}).get("exclude", [])
+                    return gr.update(visible=False), gr.update(value=", ".join(inc_kws)), gr.update(value=", ".join(exc_kws))
                     
-            camp_dropdown.change(update_ui_on_select, inputs=[camp_dropdown], outputs=[new_camp_name, kw_input])
+            camp_dropdown.change(update_ui_on_select, inputs=[camp_dropdown], outputs=[new_camp_name, inc_kw_input, exc_kw_input])
             
             save_config_btn = gr.Button("💾 Save Configuration to Hugging Face", variant="primary")
             config_status = gr.Markdown()
             
-            def save_easy_config(selected_camp, new_name, kw_string):
+            def save_easy_config(selected_camp, new_name, inc_string, exc_string):
                 rules = load_campaign_rules()
                 
-                raw_kws = kw_string.split(",")
-                clean_kws = [k.strip().lower() for k in raw_kws if k.strip()]
+                clean_inc = [k.strip().lower() for k in inc_string.split(",") if k.strip()]
+                clean_exc = [k.strip().lower() for k in exc_string.split(",") if k.strip()]
                 
                 target_camp = new_name.strip() if selected_camp == "+ Create New Campaign" else selected_camp
                 
                 if not target_camp:
                     return "❌ Error: Campaign name cannot be empty.", gr.update()
                     
-                # 1. Update the dictionary
-                rules[target_camp] = clean_kws
+                # 1. Update the dictionary format
+                rules[target_camp] = {"include": clean_inc, "exclude": clean_exc}
                 
                 # 2. Save file locally first
                 with open(CAMPAIGN_CONFIG_FILE, "w") as f:
@@ -554,7 +550,7 @@ with gr.Blocks(title="Leadfeeder Campaign Pro") as demo:
                 updated_choices = list(rules.keys()) + ["+ Create New Campaign"]
                 return status_message, gr.update(choices=updated_choices, value=target_camp)
                 
-            save_config_btn.click(save_easy_config, inputs=[camp_dropdown, new_camp_name, kw_input], outputs=[config_status, camp_dropdown])
+            save_config_btn.click(save_easy_config, inputs=[camp_dropdown, new_camp_name, inc_kw_input, exc_kw_input], outputs=[config_status, camp_dropdown])
             
         # --- TAB 5: DEBUGGER ---
         with gr.Tab("🛠️ Debugger"):