ui/callbacks.py

# ui/callbacks.py

# -*- coding: utf-8 -*-
#
# PROJECT:      CognitiveEDA v5.7 - The QuantumLeap Intelligence Platform
#
# DESCRIPTION:  This module contains the core logic for all Gradio event handlers.
#               The clustering callback is now updated to include persona profiling.

import gradio as gr
import pandas as pd
import logging
from threading import Thread

import plotly.graph_objects as go
import plotly.express as px

from core.analyzer import DataAnalyzer, engineer_features
from core.llm import GeminiNarrativeGenerator
from core.config import settings
from core.exceptions import DataProcessingError
from modules.clustering import perform_clustering
# --- NEW IMPORT ---
from modules.profiling import profile_clusters


# --- Primary Analysis Chain (Unchanged) ---
def run_initial_analysis(file_obj, progress=gr.Progress(track_tqdm=True)):
    if file_obj is None: raise gr.Error("No file uploaded.")
    progress(0, desc="Validating configuration...")
    if not settings.GOOGLE_API_KEY: raise gr.Error("CRITICAL: GOOGLE_API_KEY is not configured.")
    try:
        progress(0.1, desc="Loading raw data...")
        df_raw = pd.read_csv(file_obj.name) if file_obj.name.endswith('.csv') else pd.read_excel(file_obj.name)
        if len(df_raw) > settings.MAX_UI_ROWS:
            df_raw = df_raw.sample(n=settings.MAX_UI_ROWS, random_state=42)
        progress(0.5, desc="Applying strategic feature engineering...")
        df_engineered = engineer_features(df_raw)
        progress(0.8, desc="Instantiating analysis engine...")
        analyzer = DataAnalyzer(df_engineered) 
        progress(1.0, desc="Analysis complete. Generating reports...")
        return analyzer
    except Exception as e:
        logging.error(f"Error in initial analysis: {e}", exc_info=True)
        raise gr.Error(f"Analysis Failed: {str(e)}")

def generate_reports_and_visuals(analyzer, progress=gr.Progress(track_tqdm=True)):
    if not isinstance(analyzer, DataAnalyzer):
        yield (None,) * 14
        return
    progress(0, desc="Spawning AI report thread...")
    ai_report_queue = [""]
    def generate_ai_report_threaded(a):
        narrative_generator = GeminiNarrativeGenerator(settings.GOOGLE_API_KEY)
        ai_report_queue[0] = narrative_generator.generate_narrative(a)
    thread = Thread(target=generate_ai_report_threaded, args=(analyzer,))
    thread.start()
    progress(0.4, desc="Generating reports and visuals...")
    meta = analyzer.metadata
    missing_df, num_df, cat_df = analyzer.get_profiling_reports()
    fig_types, fig_missing, fig_corr = analyzer.get_overview_visuals()
    initial_updates = (
        gr.update(value="⏳ Generating AI report..."), gr.update(value=missing_df),
        gr.update(value=num_df), gr.update(value=cat_df), gr.update(value=fig_types),
        gr.update(value=fig_missing), gr.update(value=fig_corr),
        gr.update(choices=meta['numeric_cols'], value=meta['numeric_cols'][0] if meta['numeric_cols'] else None),
        gr.update(choices=meta['numeric_cols'], value=meta['numeric_cols'][0] if meta['numeric_cols'] else None),
        gr.update(choices=meta['numeric_cols'], value=meta['numeric_cols'][1] if len(meta['numeric_cols']) > 1 else None),
        gr.update(choices=meta['columns']), gr.update(visible=bool(meta['datetime_cols'])),
        gr.update(visible=bool(meta['text_cols'])), gr.update(visible=len(meta['numeric_cols']) > 1)
    )
    yield initial_updates
    thread.join()
    progress(1.0, desc="AI Report complete!")
    final_updates_list = list(initial_updates)
    final_updates_list[0] = gr.update(value=ai_report_queue[0])
    yield tuple(final_updates_list)

# --- Interactive Explorer Callbacks (Unchanged) ---
def create_histogram(analyzer, col):
    if not isinstance(analyzer, DataAnalyzer) or not col: return go.Figure()
    return px.histogram(analyzer.df, x=col, title=f"<b>Distribution of {col}</b>", marginal="box")

def create_scatterplot(analyzer, x_col, y_col, color_col):
    if not isinstance(analyzer, DataAnalyzer) or not x_col or not y_col: return go.Figure()
    df_sample = analyzer.df.sample(n=min(len(analyzer.df), 10000))
    return px.scatter(df_sample, x=x_col, y=y_col, color=color_col if color_col else None)


# --- MODIFIED CLUSTERING CALLBACK ---
def update_clustering(analyzer, k):
    """
    Orchestrates the full clustering workflow:
    1. Runs K-Means clustering.
    2. Receives cluster labels.
    3. Calls the profiling module to analyze the segments.
    4. Returns all results to the UI.
    """
    if not isinstance(analyzer, DataAnalyzer):
        # Return empty updates for all 5 clustering output components
        return go.Figure(), go.Figure(), "", "", go.Figure()

    # Step 1: Perform Clustering to get visuals and labels
    fig_cluster, fig_elbow, summary, cluster_labels = perform_clustering(
        analyzer.df, analyzer.metadata['numeric_cols'], k
    )

    if cluster_labels.empty:
        # Handle cases where clustering fails (e.g., not enough data)
        return fig_cluster, fig_elbow, summary, "Clustering failed. No personas to profile.", go.Figure()

    # Step 2: Profile the resulting clusters
    numeric_to_profile = ['Total_Revenue', 'Quantity_Ordered', 'Hour']
    cats_to_profile = ['City', 'Product', 'Day_of_Week']
    
    # Filter to only use columns that actually exist in the engineered dataframe
    numeric_to_profile = [c for c in numeric_to_profile if c in analyzer.df.columns]
    cats_to_profile = [c for c in cats_to_profile if c in analyzer.df.columns]

    md_personas, fig_profile = profile_clusters(
        analyzer.df, cluster_labels, numeric_to_profile, cats_to_profile
    )

    # Step 3: Return all 5 results in the correct order for the UI
    return fig_cluster, fig_elbow, summary, md_personas, fig_profile