Update app.py

app.py

@@ -1,7 +1,7 @@
 # Fichier: app.py
 
 # ===================================================================================
-# WAHIS SCRAPER - VERSION FINALE AVEC CORRECTION DE LA VISUALISATION CARTE
+# WAHIS SCRAPER - VERSION FINALE, SIMPLE ET ROBUSTE (FILTRES ET TABLEAU)
 # ===================================================================================
 
 import gradio as gr
@@ -16,7 +16,6 @@ import subprocess
 import zipfile
 from playwright.async_api import async_playwright
 from playwright_stealth import stealth_async
-import plotly.express as px
 
 # Installation du navigateur (ne change pas)
 def install_playwright_browsers():
@@ -82,87 +81,100 @@ class WAHISScraper:
                 if browser and browser.is_connected(): await browser.close()
 
 def process_data_and_create_zip(reports, outbreaks, additional_infos):
-    if not reports or not outbreaks: return pd.DataFrame(), pd.DataFrame(), None
+    if not reports: return pd.DataFrame(), [], [], [], None
+    
+    # Enrichir les données des foyers avec le nom de la maladie et du pays
+    report_map = {report['eventId']: {'disease': report['disease'], 'country': report['country']} for report in reports}
+    for outbreak in outbreaks:
+        event_info = report_map.get(outbreak.get('eventId'), {})
+        outbreak['disease'] = event_info.get('disease')
+        outbreak['country'] = event_info.get('country')
+    
     valid_additional_infos = [info for info in additional_infos if isinstance(info, dict)]
     additional_info_map = {info.get('outbreakId'): info for info in valid_additional_infos}
     for outbreak in outbreaks:
         outbreak_id = outbreak.get('outbreakId')
         if outbreak_id in additional_info_map: outbreak.update(additional_info_map[outbreak_id])
+    
     df_outbreaks = pd.DataFrame(outbreaks)
-    df_summary = pd.DataFrame(reports)
+    
+    # Créer les listes pour les filtres à partir des données complètes
+    all_countries = sorted(df_outbreaks['country'].dropna().unique()) if 'country' in df_outbreaks else []
+    all_diseases = sorted(df_outbreaks['disease'].dropna().unique()) if 'disease' in df_outbreaks else []
+    all_species = sorted(df_outbreaks['species'].dropna().unique()) if 'species' in df_outbreaks else []
+
+    # Créer le package ZIP
     timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
     zip_path = OUTPUT_DIR / f"wahis_package_{timestamp}.zip"
     with zipfile.ZipFile(zip_path, 'w', zipfile.ZIP_DEFLATED) as zipf:
+        df_summary = pd.DataFrame(reports)
         df_summary.to_excel(OUTPUT_DIR / "1_summary_events.xlsx", index=False)
         df_outbreaks.to_excel(OUTPUT_DIR / "2_outbreaks_full_details.xlsx", index=False)
         zipf.write(OUTPUT_DIR / "1_summary_events.xlsx", arcname="1_summary_events.xlsx")
         zipf.write(OUTPUT_DIR / "2_outbreaks_full_details.xlsx", arcname="2_outbreaks_full_details.xlsx")
-    return df_summary, df_outbreaks, str(zip_path)
-
-def create_map_figure(df, title="Carte des Foyers"):
-    if df.empty or 'latitude' not in df.columns or 'longitude' not in df.columns:
-        fig = px.scatter_mapbox(title="Aucune donnée à afficher")
-        fig.update_layout(mapbox_style="open-street-map")
-        return fig
     
-    df_plot = df.copy()
-    
-    size_col_name = None
-    if 'totalAnimalsAffected' in df_plot.columns: size_col_name = 'totalAnimalsAffected'
-    elif 'cases' in df_plot.columns: size_col_name = 'cases'
-    elif 'deaths' in df_plot.columns: size_col_name = 'deaths'
-
-    # CORRECTION : Utiliser 'locationName' qui est le bon nom de colonne
-    hover_name_col = 'locationName' if 'locationName' in df_plot.columns else None
+    return df_outbreaks, ["Tous"] + all_countries, ["Toutes"] + all_diseases, ["Toutes"] + all_species, str(zip_path)
 
-    if size_col_name:
-        df_plot['size_for_plotting'] = pd.to_numeric(df_plot[size_col_name], errors='coerce').fillna(1)
-        fig = px.scatter_mapbox(df_plot, lat="latitude", lon="longitude", color="diseaseName", 
-                                size='size_for_plotting', size_max=30, 
-                                hover_name=hover_name_col, title=title, zoom=1, height=600)
-    else:
-        fig = px.scatter_mapbox(df_plot, lat="latitude", lon="longitude", color="diseaseName", 
-                                hover_name=hover_name_col, title=title, zoom=1, height=600)
-
-    fig.update_layout(mapbox_style="open-street-map", margin={"r":0,"t":40,"l":0,"b":0})
-    return fig
+# --- Fonctions de l'interface Gradio ---
 
 async def run_and_update_ui():
     scraper = WAHISScraper()
     reports, outbreaks, additional_infos, logs = await scraper.run_extraction_async()
     if not reports:
-        return logs, None, None, None, None, None
-    df_summary, df_outbreaks, zip_path = process_data_and_create_zip(reports, outbreaks, additional_infos)
-    initial_map = create_map_figure(df_outbreaks)
-    return logs, initial_map, df_summary, df_outbreaks, gr.File(value=zip_path, visible=True), "Extraction réussie. Explorez les données ci-dessous."
+        return {status_textbox: logs, ui_visibility_group: gr.Group(visible=False)}
+    
+    df_outbreaks, countries, diseases, species, zip_path = process_data_and_create_zip(reports, outbreaks, additional_infos)
+    
+    return {
+        status_textbox: logs,
+        outbreak_data_state: df_outbreaks,
+        filter_country: gr.Dropdown(choices=countries, value="Tous"),
+        filter_disease: gr.Dropdown(choices=diseases, value="Toutes"),
+        filter_species: gr.Dropdown(choices=species, value="Toutes"),
+        outbreaks_table: df_outbreaks,
+        download_section: gr.File(value=zip_path, visible=True),
+        ui_visibility_group: gr.Group(visible=True)
+    }
+
+def update_table(country, disease, species, df_outbreaks):
+    if df_outbreaks is None or df_outbreaks.empty: return pd.DataFrame()
+    filtered_df = df_outbreaks.copy()
+    if country != "Tous": filtered_df = filtered_df[filtered_df['country'] == country]
+    if disease != "Toutes": filtered_df = filtered_df[filtered_df['disease'] == disease]
+    if species != "Toutes": filtered_df = filtered_df[filtered_df['species'] == species]
+    return filtered_df
 
 with gr.Blocks(theme=gr.themes.Soft(), title="WAHIS Scraper") as demo:
+    outbreak_data_state = gr.State()
     gr.Markdown("# 🤖 Scraper pour WAHIS (WOAH) - Version Robuste")
     run_button = gr.Button("🚀 Lancer l'extraction des données", variant="primary")
     
-    status_message = gr.Markdown("")
-
-    with gr.Accordion("Journal d'exécution", open=False):
+    with gr.Group(visible=False) as ui_visibility_group:
+        gr.Markdown("### 🔍 Filtrez les données")
+        with gr.Row():
+            filter_country = gr.Dropdown(label="Pays")
+            filter_disease = gr.Dropdown(label="Maladie")
+            filter_species = gr.Dropdown(label="Espèce")
+        gr.Markdown("### 📋 Tableau Détaillé des Foyers (avec Coordonnées GPS)")
+        outbreaks_table = gr.DataFrame(wrap=True)
+
+    with gr.Accordion("Journal d'exécution et Téléchargement", open=False):
         status_textbox = gr.Textbox(lines=15, label="📜 Logs", interactive=False)
-    
-    gr.Markdown("---")
-    gr.Markdown("### 🗺️ Carte des Foyers (Vue d'ensemble)")
-    map_plot = gr.Plot()
-    
-    gr.Markdown("### 📊 Tableau Résumé des Rapports")
-    summary_table = gr.DataFrame()
-    
-    gr.Markdown("### 📋 Tableau Détaillé des Foyers (avec Coordonnées GPS)")
-    outbreaks_table = gr.DataFrame()
-    
-    gr.Markdown("### 💾 Télécharger le Package Complet")
-    download_section = gr.File(label="Package de Données (.zip)")
+        download_section = gr.File(label="💾 Télécharger le Package Complet (.zip)")
 
     run_button.click(
         fn=run_and_update_ui,
         inputs=[],
-        outputs=[status_textbox, map_plot, summary_table, outbreaks_table, download_section, status_message]
+        outputs=[status_textbox, ui_visibility_group, outbreak_data_state, filter_country, filter_disease, filter_species, outbreaks_table, download_section]
     )
+    
+    filters = [filter_country, filter_disease, filter_species]
+    for f in filters:
+        f.change(
+            fn=update_table,
+            inputs=[filter_country, filter_disease, filter_species, outbreak_data_state],
+            outputs=[outbreaks_table]
+        )
 
 if __name__ == "__main__":
     demo.launch()